The Electromagnet

fyxmll
Uttiretîtg
îlrmg
BOUGHT WITH THE INCOME FROM THE
SAGE
ENDOWMENT FUND
THE GIFT OF
U^ntu W. Sage
1891
A.SLtf^7^3
2..^/iAq.o.rf.
5901
The
D.
Van Nostrand Company

and supply
it
intend this booh to be sold to the Public

at the advertised price,
to
the Trade on terms which will not allow
of discount.
The
tine
original of
tliis
book
is in
Cornell University Library.
There are no known copyright
restrictions in
text.
the United States on the use of the
http://www.archive.org/details/cu31924031243367
arVIS/Oa*^""
""'*"y
Library
The electromagnet.
olin.anx
3 1924 031 243 367
THE
ELECTROMAGNET
CHARLES
R.
UNDERHILL
Co,
Chief Electrical Engineer
Varley Duplex Magnet
.<^HWt.
NEW YORK:
D.
VAN NOSTRAND COMPANY

23
Murray and
27
Warren
Sts.
1903
COPVKIGHTED BY
VARLEY DUPLEX MAGNET
CO.
Stanbope Ipresa
H.
GILSON BOSTON,
COMPANY
U.S.A.
PREFACE.
This book is a new and revised edition of " Ths Electromagnet " by Townsend Wolcott, A. E. Kennelly,- and
Richard Varley.
The
author has endeavored to give the facts connectedly,
so that the reader
may
easily follow the reasoning without
referring to different parts of the book, with the exception of the tables,
which are placed
in the
Appendix
for
convenience.
Much
of the data, especially that concerning windings,
has been obtained from actual practice.
As the economy and efficiency of an electromagnet depend largely on the proper design and calculation of the winding, particular attention has been paid to that detail. Acknowledgments are due to R. Varley, W. J. Varley, A. D. Scott, J. M. Knox, and W. H. Balke for data and
assistance.
C. R.
Underbill.
Providence,
R.I.,
August
29, 1903.
CONTENTS
PAGE
Notation
ix
CHAPTER
ART.
1.
I.
ELECTRIC AND MAGNETIC CIRCUIT CALCULATIONS.
2.
3.
4. 5.
6.
7.
Magnetism Magnetic Poles Magnetic Field Forms of Permanent Magnets Magnetic Induction
Electric Circuit
3 4
6 6
6
7
11
Ohms' Law
Divided or Branched Circuits
8. g.
Magnetic Units
Electromagnetism Force about a wire Ampere-turns
Effect of Iron in Magnetic Circuit
10.
13 13
17
11.
12.
13.
14.
18
Terms Expressed
Permeability
.
15.
16.
Measure General Relations between Magnetic Units

in English
19 19
21
17.
18.
Magnetic Testing
Practical Calculations of Magnetic Circuit
22
ig.
Effect of Joint in Magnetic Circuit
23 28
zo.
Magnetic Leakage
Limits of Magnetization
Hysteresis Retentiveness
29
32
21.
22.
23.
34
35
Problems
36
VI
CONTENTS.
CHAPTER
II.
WINDING CALCULATIONS.
ART.
24.
25.
PAGE
Simple Principle of Calculating Windings Copper Constants
39 40
41
26.
Most
Efficient
Winding
27.
28.
Circular Windings
42
47
Points to be Observed in Practice
29.
30.
Formulae for Turns, Resistance, and Ampere-tums Constant Resistance with Variable Insulation
Layers and Turns per Inch
....
...
49
51
31. 32. 33. 34.

35.
52
36.
37. 38. 39. 40.
Windings with Wires other than Copper Small Magnets on High-voltage Circuits Resistance Wires One Coil Wound Directly Over the Other Parallel Windings Joint Resistance of Parallel Windings RelationsHoldingfor Any Size of Wire and Winding Volume The American Wire Gauge (B & S)
. .
53 53 55 56
57
59 59
60
63
Thickness of Insulation
Ratio of Weight of Copper to Weight of Insulation
.
41.
42. 43.
63 66
Weight
of Insulation to Insulate
Any Wire
. .
General Construction of Electromagnets

Insulation of Bobbin for High Voltage Theory of Magnet Windings Paper Inserted into the Winding Duplex Windings Other Forms of Windings than Round Square or Rectangular Windings Windings with Elliptical Cross-Sections Windings Whose Cross-Sections have Parallel Sides and Rounded Ends Problems
.
44.
45. 46.
....
67
68 70
72
47.
48. 49.
50.
73 7^
76
80
51.
81
8i
CONTENTS.
VU
CHAPTER
III.
HEATING OF MAGNET
ART.
52. 53. 54. 55.
COILS.
PAGE
Effect of Heating
90
. .
Relation between Magnetomotive Force and Heating
96
103
Advantage of Thin Insulating Material
Work
at
End
of Circuit
104
107
Problems
CHAPTER
IV.
ELECTROMAGNETS AND SOLENOIDS.

56. 57.
Forms
of Electromagnets
no
114 116
.
.
Direction of Flux in Core
58.
Action of an Electromagnet 59. Calculation of Traction

60. 61.
62.
117
Solenoids
119
Action of Solenoids
Polarized Magnets
120
122
Problems
126
CHAPTER
V.
ELECTROMAGNETIC PHENOMENA.
63.
64. 65. 06.
Induction
Self-induction
127
127
Alternating Currents
Eddy Currents
.,.....,,.,.
130
130
viii
CONTENTS.
APPENDIX.
PAGB
Standard Copper Wire Table

Explanation of Table
135
136
137
Bare Copper Wire
Bare Copper Wire (Commercial Half Sizes) Weight of Copper in 100 Pounds of Cotton Covered Wire Weight of Copper in 100 Pounds of Cotton Covered Wire
.
138 138
(Commercial Half Sizes)
139
Silk Insulated
Weight of Copper
in 100
Pounds of
Wire (Com139
. .
mercial Half Sizes)
Weight of Copper in 100 Pounds of Silk Insulated Wire Data for Insulated Wire Tables lo-Mil. Double Cotton, Insulated Wire 8-Mil. Double Cotton
5-Mil. Single 4-Mil. Single
8-Mil. 4-Mil. 4-Mil.
140 140
141 141
Cotton
Cotton
142 142 143 I4j
3-Mil.
Double Cotton (Commercial Half Sizes) Single Cotton (Commercial Half Sizes) Double Silk Double Silk
144
144 145
145
2-Mil. Single Silk

i.S-Mil. Single Silk
4-Mil.
3-Mil.
Double Double
Silk
(Commercial Half Sizes) Silk (Commercial Half Sizes) (Commercial Half Sizes)
146
146
147 147
2-Mil. Single Silk

1. 5-Mil.
Single Silk (Commercial Half Sizes)
Table of Resistances of German Silver Wire Permeability Table

Traction Table
Insulating Materials
148 149
1^0
150
.
.
Decimal Equivalents of Fractional Parts of an Inch

Ix)garithms of
151
Numbers
'Sî 153
Antilogarithms
'S4i '55
NOTATION.
= percentage of copper in cotton insulated wire. = percentage of copper in silk insulated wire. A = area in square inches. A^ = area in square centimeters. d = distance between centers of cores in inches. B = magnetic induction (English system). (B = magnetic induction in gausses. = constant = .0000027107. c CM. = circular mils. Cy, = weight of cotton in pounds. = diameter of core + sleeve. = as in Fig. 36, p. 76.
a
!
(/
= as in Fig. 41, p. 80. = as in Fig. 42, p. 82. = diameter of core. = deflection of galvanometer. Z> = true outside diameter of round windings. ^^ = as in Fig. 36, p. 76.
"4
>
/
rt'j
</,,
</*
^'1=
Z>5=
as in Fig. 41, p. 80.
as in Fig. 42, p. 82.
=
e
E.M.F.
electromotive force in volts.
base of Naperian logarithms
2.7182818.
NOTATION.
P =
SF
= / = number of cycles per second. g = total diameter of insulated wire. = space factor. = lateral value of wire and insulation. gi g^ = vertical value of wire and insulation. H = magnetizing force (English system). X = magnetizing force in gausses. JI = as in Fig. 42, p. 82. H.P. = horse-power. / = current in amperes.
g''
magnetomotive force (English system). M.M.F. = magnetomotive force in gilberts.
/JV=
/r
ampere-turns.
= joint resistance. w" J^ = resistance factor = k = constant of galvanometer. L = length of winding. Lb. = pounds adv. = length of magnetic circuit in inches. / 4 = length of magnetic circuit in centimeters. = length of wrap of paper in inches. = length of wire or strand in inches. 4, L = inductance in henries. M = mean or average diameter of winding in inches. J/i = as in Fig. 42, p. 82. m = turns of wire per inch. iV = number of turns of wire in winding. n = number of layers. c = a constant (see p. 35). n,u = number of wires. P = paper allowance for duplex windings.
tt
.
/p
NOTATION.
XI
P =
=
(R
magnetic attraction or
pull.
combined resistance and space factor
jR
r
s
= magnetic reluctance in oersteds. = magnetic reluctance (English system). = radius of circle. = gauge number of wire (B. & S.).
= silk allowance for duplex windings. = radiating surface in square inches. S = weight of silk in pounds. T = thickness or depth of winding. = time constant. f i = rise in temperature. V = volume of winding in cubic inches. Fi = leakage coefficient. y^ = volumeof paper in duplex windings (cubic inches). y^ = volume of silk space in cubic inches. w = combined weight and space factor = W = watts. IVg = watts per square inch.
S
S,.
Wc=
X
watts lost per cubic centimeter of iron.
IT
= as in Fig. 15, p. 16. X = intermediate diameter in inches. Z = impedance. A = diameter of wire in inches. = ohms per pound for insulated wires. X = weight of bare wire in pounds. = permeability. =3.1416 = ratio between diameter and
fi
circum-
ference of circle.
p
p,
= =
electrical resistance.
series resistance.
NO TA TION.
S ^
<î
</)2
= =
cross-section of insulation in circular inches.

flux in
webers
lines of force.
= = ii = us' = =
lit
useful flux.
leakage in webers.
ohms per pound ohms per inch. ohms per foot.
for bare wires.
THE ELECTROMAGNET.
CHAPTER
I.
ELBCTRIC AND MAGNETIC CIRCUIT CALCULATIONS.

I.
Magnetism.
"
Magnetism
is
that peculiar property occasionally pos-
sessed by certain bodies (more especially iron and steel) whereby under certain circumstances they naturally attract or repel one another according to determinate laws." Magnetism is supposed to have first been discovered by the ancients in Magnesia, Thessaly, where they found an ore which possessed a remarkable tractive power for iron. A piece of the ore having this power they termed
a Magnet.
It was also discovered that when a piece of this ore was suspended so that it could move freely, one of its ends always pointed to the north, and the other end, of
course, pointed south.
Navigators took advantage of this
principle to steer their ships,

stone
ore.
and hence the name Lode(meaning Leading Stone) was given to the natural
were made by rubbing a bar of
a piece of lodestone.
Artificial magnets
hardened
steel with
Artificial
magnets which retain their magnetism for a
long time are called Permanent Magftets.
THE ELECTROMAGNET.
2.
Magnetic Poles.
magnet which has a tendency to point its North Pole, although this arrangement really makes the pole situated near the geographical North Pole of the earth the magnetic South Pole of the earth, since like magnetic poles repel one anof a
The end
is
north
naturally termed
other and unlike poles attract each other.
In this book the term North-seeking Pole will be used

instead of North Pole, as the latter
is
liable to
become
confused with the geographical North Pole of the earth.
The
north-seeking pole of a magnet
is
equal in strength
to its south-seeking pole, the strength gradually decreas-
ing until
at
all.
midway between
is
the poles there
is
no attraction
This place
called the Neutral Line.
Every magnet has two poles, and if the Par Magnet I should be broken into any number of pieces, each piece would be a perfect magnet, with a north-seeking and south-seeking pole of equal strength.
in Fig.
ELECTRIC AXD MAGAETlC CIRCUIT.

3.
Magnetic Field.
If
filings sprinkled
a piece of paper be laid over a bar magnet and iron over the paper, and then if the paper be
they will take positions as shown in Fig.
2.
jarred slightly so as to give the filings an opportunity to

settle freely,
From mathematical and experimental research it has been fomid that tlie magnetism passes through the inside of a magnet from pole to pole, issuing from its north-seeking pole, and returning through the air or surrounding
media
to its south-seeking pole, although
aU of the mag-
netism does not pass from the ends of the magnet (as
may
be seen by reference to Fig. 2), which fact shows that all of the magnet on one side of the center, or neutral line, is north-seeking, while all on tlie other side is south-seeking,
the poles being stronger near the ends.
THE ELECTROMAGNET.
These streams of magnetism are called Lines of Force, and the media about the poles of the magnet, through which they pass, is called the Field of Force. The lines are always closed curves hence the path through which
;
they flow
is
called the Magnetic Circuit.
4.
Forms
of
Permanent Magnets.
is
The
as
usual form of the magnetic circuit
substantially
shown in Fig. 3. This form is called a Horseshoe permanent magnet, and is a bar bent into the shape of a horseshoe. This is done to bring the two poles of the magnet near each other, and thus
shorten the magnetic circuit.
piece of iron which
is
The
attracted
its
by the magnet
ture.
is
called
Armaout
The
through
ture,
lines
of
force flow
the
north-seeking pole
of the magnet, through the arma-
and
into the south-seeking
pole,
through the
substance
of
the magnet to the starting-point.
To
obtain the largest
3
Fig. 3-
of lines of force through a

net, the
number mag-
magnetic circuit should
be as short, and have as few Air Gaps, as possible. Another form is shown in- Fig. 4, and in effect is merely
two horseshoe magnets placed with similar poles together, thus tending to repel one another. A magnet thus
arranged
results
is
said to have
Consequent Poles.
if
The same
would
be obtained
two or more horseshoe
ELECTRIC AND MAGNETIC CIRCUIT.
magnets were placed side by side with similar poles

together.
M.
I
N.
s.
s.
Fig.
nets,
Magnets of the latter form are called Compound Magand are commonly used on magneto generators.
s.)
(See Fig.
Fig.
5-
it is very important that each magnets should have the same strength otherwise, the weaker magnets would act as a return cir-
In this type of magnet
of the separate
THE ELECTROMAGNET.
and the
effective field
cuit for the stronger ones,
would
thus be weakened.
5.
Magnetic Induction.
When a magnet attracts a piece of iron, this iron itself becomes a magnet while it is being attracted, and will attract other pieces of iron, which in turn also become
magnets.
pieces
is
This
successive
magnetization of the
iron
said to be produced
by Magnetic
Induction.
6.
Electric Circuit.
The
force which causes a current of electricity to flow

is
through a conductor
viated E.M.F.),
called Electromotive Force (abbreunit used in practice

offers
is
and the
the
Volt.
Every known substance

electrical resistance is the
some
Resistance to the
practical unit of
passage of an electric current.
The
Ohm.
is
The
is
unit strength of electric current
the
Ampere and
produced by the unit electromotive force acting through
the unit resistance.
The
rule expressing the relation
force, current strength,
between electromotive and resistance is known as
7.
Ohm's Law.
is
The strength of
the current
equal
to
the electromotive
force divided by the resistance, or
'-V
Transposing,
E = Ip
(2)

and
where
'^
7'
(3)
/=
E=
p
strength of current, electromotive force,

resistance.
The
unit of electric
to the
power
is
termed the Watt.
The
watts are equal

resistance,
square of the current multiplied by the
or
=^ I^p,
(4)
whence, by substitution,
W=
(5), also,
W = EJ.
fF.
(6)
746 watts equal one horse-power, or H.P.
746
(7)
Therefore, one watt equals .00134 horse-power, or
fF=
8.
.00134 H.P.
(8)
Divided or Branched Circuits.

of equal resistances are connected
When any number

of
in multiple, the Joint Resistance is equal to the resistance
one conductor divided by the number of conductors. The joint resistance of two equal or unequal resistances
equal
to their product
contiected in multiple is
divided by their
sum, or
P
Pi
Example. The resistance of two electromagnets is ohms and 92 ohms respectively. What will be the joint 74
resistance
when they
are connected in multiple
'
THE ELECTROMAGNET.
Solution.
X
92 92
74 74
6,808
166
= 41+
ohms.
Ans.
The joint
is
resistance
of three or more conductors in multiple
equal
to the reciprocal
of their joint
conductivity.
is
Since the conductivity of a conductor

of its resistance, the conductivity
let PX1P21 îid Pa
the reciprocal
if
=9
Therefore,
we
equal the separate resistances of the three

6,
branches, as in Fig.
the conductivities will be

ft Pi
and
respectively.
/>s
Their joint conductivity
is
Px
JL
_,_
^ _ P-iP% +
Pz
PlPs
PlP2^
Pi
PiPiPs
and the reciprocal
_
P2PS
P1P.P3
+ PiPs + P1P2
(9).
^y^^
(^^)
which
is
the same as
riAAMAAMAAAH
VWVWWVVW^
MWVWWWV^M
Fig.
6.

Example. Three electromagnets having resistances of 4 ohms, 6 ohms, and 8 ohms respectively are to be connected in multiple. What will be their joint resistance? Solution.
By formula
PiPiPa
(lo),
Jr
=
kPa
4X6x8
X
8)
+ PiPz + P1P2 (6 X 8) + (4 IQ2 = -^^ = 1.84+ ohms.

,
(4
6)
104
The
current which will flow through each branch of the
circuit is
found by ascertaining the
total current flowing

{2),
through the branches, and then by formula
= Ip,
to
find
find the electromotive force across the branches

^,
from a
,
and next by applying formula
(i),
/=
P
the
current flowing through each branch.
>VWWWW\/\r
Example.
4 ohms,
pj
In
the diagram (Fig. 7),
5 ohms,
pi=
3 ohms, p^ ohm, and p^, the internal

pj
resistance of the battery
ohms.
lO
THE ELECTROMAGNET.
amperes of current
will flow
How many
branch
?
through each
Solution.
formula
By
formula (lo) the joint resistance of

is
the branched circuit

(i),
1.28
ohms
nearly.
By Ohm's
law,
the current
^=
branched
nearly.
f=
,..8
(2)
+ 1+2 =
the
^=
=
.935
^
73
-935 -'"P--^-
By formula
electromotive
force
across
the
circuit
from a to b
(i)
1.28
1.2 volts
Then by formula
.4,
/i
=
1.2
^ /j
=
1.2
.3,
1.2 = - =
.24.
Ans.
SO.
so.
Fig.
8.
From
the foregoing
it is
seen that two resistances, to be
connected in multiple and produce the same resistance to

the line as
if
they were connected in series, must have the
resistance of each increased four times.
r-mNmmMAAAAAMAAAA^
Fig. 9.
Assume two electromagnet windings

to
of
be used in
series,
having a
total resistance of
50 ohms each 100 ohms

as in Fig. 8.
If
II
they were connected in multiple as
in
Fig. 9, the resistance
\ of
would be ^-= 2<, ohms, or only what the resistance would be were they connected in
series.
9.
Magnetic Units.
In the following discussion it is assumed that there is no magnetic material in the circuit, except when the contrary
is
stated.
is
Unit Magnetic Pole in the C.G.S. system

that
so defined
when placed
if
at
it
exactly similar pole
one centimeter distance from an repels it with a force of one dyne.

at the center of
Now,
a unit magnetic pole be placed from

this pole
a sphere of one centimeter radius (two centimeters in diameter), there will radiate
one
line of force for
each square centimeter of surface on the sphere, and as the

area of the sphere
there will be 4
is
equal to 4
irr''
square centimeters,
3.1416
X i^=
12.5664 lines of force
radiating from the unit pole.
One
Density
line of force (also called

is
one Weber, symbol

unit
is
<^)
per square centimeter

of
JC).
called
Intensity
or unit
magnetization,
and
termed
the
Gauss
(symbol
Thus, loo gausses are

It is to
100 webers per
square centimeter.
be observed that the gauss has
it is the number of Per Unit Area in square centimeters. The number of gausses are found by dividing the total number of lines of force by the total cross-sectional area of the magnet in square centimeters. The force producing the flow of magnetism is called Magnetomotive Force, and the unit is the Gilbert (symbol JF).
nothing to do with the total area, but

lines of force
12
THE ELECTROMAGNET.
of gilberts per centimeter length of
The number
netic circuit
is
mag-
called the Magnetizing Force.

is
The law
potential
of the magnetic circuit
identical with that of

is
the electric circuit, inasmuch as the Flow

difference
equal to the
(See
divided by the resistance.
Ohm's
law,
page
6.)
In the case of the magnetic
circuit,
however, when com-
posed of iron or
Reluctance
steel,
the
magnetic resistance called
changes with the flow of magnetism called
Flux, or more correctly, with the magnetic density or lines
per square centimeter.
The
ation
is
property of the iron or steel which causes this varicalled

its
Permeability.
is
The
Reluctance or magnetic resistance
equal
to the
its
length of the magnetic circuit, divided by the product of

cross-sectional
area
and permeability.
(5^
Thus,
(ii)
where
= length of magnetic circuit in centimeters, = reluctance in oersteds, A^ = cross-sectional area in square centimeters, = permeability.
/
(R
/A
is
The magnetomotive force (abbreviated M.M.F.) in gilberts equal to the number of lines of force, multiplied by the
reluctance.
Thus,
g^
^(R,
(12)
where
= gilberts, = webers, (R = oersteds.

F
1^
'

g:
Also,
"^
m:
*^'^^
a=
|.
(.4)
Substituting the value of (R from (11) in (12),
l^cW
When
the magnetic circuit consists of several
is
parts,
the total reluctance

reluctances
;
equal to the
sum
of
all
of the
thus,
-'Self*!
-'^cara
-^csrs
10.
Electromagnetism.
if
In 1819, Oersted discovered that

to take
a compass needle be
it
brought near a wire carrying an electric current,
tends
up a
position at right angles to the direction of
the wire.
The relation which exists between direction of current and deflection of compass needle is as follows If the
:
current flows through the wire
from
left to right,
is
and
the
needle is above the wire, the north-seeking pole
deflected
toward
pole
the observer.
If below the wire, the north-seeking

the observer.
is deflected
from
11.
Force about a Wire.

is
The
explanation of the foregoing

is
that the wire carrycircles of
ing the current
surrounded by concentric
14
force,
THE ELECTROMA GNE T.

and the compass needle, being a magnet, tends to This is itself in the direction of these lines of force.
illustrated in Fig. lo.
set
The compass needle can

never set
itself
exactly in the
direction of the lines of force
on account of the earth's magnetism, unless the earth's magnetism be neutralized.

Fig.
tion
Fig. 10.
1 1 also shows the relabetween direction of cur-
rent and direction of lines of

force.
The
relation
between
the current in the wire and the intensity of magnetization, or density of lines of force,
is illustrated
in Fig. 12.
Fig.
II.
When
the wire carries 10 amperes, at one centimeter
from the center of the wire there are two lines of force (webers) per square centimeter for each centimeter length that is, two gausses and at two centimeters from of wire
the center of the wire there
square centimeter
that
is
but one line of force per

is
is,
there
but one gauss.
ELECTRIC AXD MAGNETIC CIRCUIT.

Hence the
air
is
15
following law
The
tlie
intensity in gausses in
equal
to
two-tenths times
current in amperes flow-
i6
THE ELECTROMAGNET.
meters,
At two centimeters, the circumference is 12.5664 centiand the M.M.F. is i x 12.5664 = 12.5664 gilberts. Therefore, the total M.M.F.
is
always
12.5664
is
gilberts
when 10 amperes
through the wire.
flowing
When
through
are
the wire
is
bent into
a circle and a current passed

it,
the lines of force

circles
no longer simple
but are distorted, assuming

positions as
shown in
Fig. 14,
so that the force at any point
can only be calculated
by
means
of the higher mathe-
matics, but in the center, the intensity
3C
(20)
where r
Fig- IS-
is
the radius of the loop or turn of wire as in
Fig. 15.

At any distance x from the center
X, the force
is
of loop,
on the axis
3C
=
it
.2 irlr^
-.,
(21)
but at points
algebra, but
ofl
is
the axis,
cannot be calculated by simple
approximately uniform near the center of
the loop, increasing somewhat toward the wire, while very
near the wire

of the wire
loop.
is
it is
much
greater, especially
if
the diameter
small as compared with the diameter of the
The magnetomotive
straight wire
force,
however,
is still
SF
.\-irI
(22) gilberts, as in the case of the simple circle about a
shown
in Fig. 13.
is
From
the above
deduced the following law
If a
centi-
wire one centimeter in length be bent into an arc of one
meter radius,
and a
current of 10 amperes passed through
the wire, at the center of the arc there will be one line of
force per square centimeter,

gauss.
12.
i.e.,
the intensity
will be
one
Ampere-Turns.
is
In practice the wire
wound
and one turn

through
it
of wire with
in spirals on a bobbin, one ampere of current flowing
is
called
one Ampere-turn, and
the
same
any number of turns and any number of One ampere flowing through one hundred turns amperes. gives exactly the same results as one hundred amperes
relation holds for
flowing through one turn.
The ampere-turns, then, are found by multiplying the number of turns by the current in amperes flowing through
the turns.
THE ELECTROMAGNET.
The symbol
for ampere-turns
is
IN.
Where
N= number of turns.
It
/=
current in amperes,
has already been stated that a unit magnet pole sends

is
out 12.5664 lines of force, and that a force of one dyne
exerted along each one
of these lines, or
12.5664 dynes for

that
also 12.5664
the 12.5664 lines of force.
Now,
it
may be shown
is
the force produced by ten ampere-turns
dynes or 12.5664 gilberts. produces 1.25664 gilberts.
Therefore, one ampere-turn
13.
Effect of Iron in
is
Magnetic
Circuit.
cir-
When
cuit,
iron or steel
introduced into the magnetic
the conductivity of the magnetic circuit called Perincreased.
meahility is greatly
is
The
all
permeability of air
taken as unity, and since nearly
substances excepting
air,
iron
and
steel
have the same permeability as
only the
two
latter will
is
be considered.
Lines of force
There
no insulator of magnetism.
pass through or permeate every
known
substance.
In order to distinguish the lines per square centimeter

in air
from
lines per square centimeter in iron

is
or
steel,
the symbol (B
given to the
latter,
and they are called
Lines of Induction.
Thus,
(B
= /*3C,
(23) (24)
kLECTRIC AND MAGNETIC CIRCUIT.

where
3e
(B
ju.
10
= gausses in air, = gausses in iron = permeability.
or steel,
14.
Terms Expressed
in English Measure.
Since in America the units used are in English measure,
a great many engineers prefer to change the magnetic

units into terms of English measure also.
In Metric measure
JF
1.25664 IN.
(26)
Therefore, in English measure,
F= 3-192/^
Unless otherwise specified in what follows,
will represent units in
all
(27)
symbols
English measure
and
in order to
avoid confusion the same symbols as applied to the units

in
metric measure will be used, but in heavy tjrpe and
English characters.
Thus,
F= 3-192 -^^
/i\^=.3i32F,
(27)
(28)
also
B=
15.
i^H.
(23)
General Relations between Magnetic Units.

(is)
17
From
Substituting the value of
F in (28)
ZV=.3i32.^
That
is,
(29)
the ampere-turns required to produce the fiux
<j)
are equal to the product of .3132 times the fiux and the
reluctance,
20
THE ELECTROMAGNET.
also
^^3:193^.
is,
(30)
That
the total flux
is
equal to 3.193 times the ampere-
turns divided by the reluctance.
The number
netic field, or
of lines per square inch is equal to the
total flux divided
by the
cross-sectional area of the
mag-
= ! Substituting the value of B from
(3.)
(31) in (29),
(33)
IN=-^1^
Bl
3-193/*
(33)
whence,
B=^-^^^.
is,
(34)
is
That
the
number
of
of lines per square inch
equal
to the product
3.193 times the ampere-turns and permeability, divided by the total length of the magnetic
circuit.
Fig.
16..

The above
21
formulae apply to the magnetic circuit con-
sisting of a continuous iron or steel ring as in Fig. i6.
i6.
Permeability.
The
permeability
ju.
decreases as the magnetic density
increases,
and
by actual test, made.
found for various grades of iron or steel and then curves plotted on charts, or tables
is
(See permeability table on
p. 149.)
Example.
ring
in
Assume the
16
to
.6
mean diameter
inches,
of
the
iron
cross-
Fig.
be three
and the
it is
sectional area to
be
square inch, and that
required
to force 60,000 lines of force through the iron.
How
many
ampere-turns are required
Solution.
inches
Since the mean

.6
diameter
is
is
three inches,
the length of the magnetic circuit
3-1416
9.4248
/.
60,000 lines through
square inch
is
equivalent to
Assuming the ring to be made of annealed wrought iron, and referring to formula (32) and substituting the values of B, /, and
100,000 lines per square inch
3-
//,,
_ =
.3132
100,000
X 9.4248
Ans.
289,000
360
805 ampere-turns.
is
360
Since there
a wide variation in the permeability in

steel,
the same grade of iron or
the above result would
only be approximate unless a very careful test was

of tlie ring
itself.
made
22
THE ELECTROMAGiYET.
Vj.
M^netic
Testing.
in Fig. 17.
is
One method
of doing this
is illustrated
is
the iron ring to be tested, and

coil C,
wound with a
re-
magnetizing
which
is
in series with a source of
current B, adjustable rheostat R,

versing-switch S.
and double-throw
secondary winding W, called the exploring

ballistic
coil, is
connected to a
able rheostat R.
galvanometer through an adjust-
Fig. 17.
The magnetizing
force
H
where /
ringis
3.192
-J
zy
,
(35)
the
mean
length of the magnetic circuit in the
When
flection is
the primary circuit
is
closed or broken, a de-
produced
in the ballistic
galvanometer propor-
tional to the magnetic flux.
Then
where
B = Kd, d^ = deflection of galvanometer,

K
(36)
constant of galvanometer.

The
permeability
is
23
then found by formula (24),
18.
It
Practical Calculations of Magnetic Circuit.

is
has be'en stated that the induction

force
equal to the
magnetizing
multiplied
by the permeability, or
B = jiH, and the magnetizing force is equal to the M.M.F. per inch. Now, since the M.M..F. is proportional
for any specific case depends upon the ampere-turns per inch of magnetic circuit, and nothing else.
to the ampere-turns, the induction
On
inch.
this principle curves
show the value
of
for
have been constructed which any number of ampere-turns per

steel.
In Figs. 18 and 19 there are several curves comuse the curves, find the point on the curve horizonper square inch, and then
vertically
this point
bined representing difEerent grades of iron and
To
from
tally opposite the induction
on the curve trace
downwards
to the ampere-turns per linear inch.
The product
of the length of the magnetic circuit into
the ampere-turns per linear inch gives the total
number
of ampere-turns required to maintain the induction
B in
of
the iron.
As an example, assume a
form of a Swedish iron
the magnetic circuit
is
closed magnetic circuit in the
ring,
and the average length
10 inches, and that 87,000 lines
per square inch are required.

Referring to the chart, Fig. 18, for 87,000 lines per square inch there are required 20 ampere-tiurns for each
24
THE ELECTROMAGNET.
Induction per Square
CM.
-rr~
i;
i
ELECTRIC AND MAGNETIC CIRCUIT,

iiduction per Square
25
CM.
(.OoMsses.)
26
THE ELECTROMAGNET.
is
inch of circuit, and since the length of the circuit

induction of 87,000 lines per square inch
200.
If
lo inches, the total ampere-turns required to maintain an

is
10
X 20
=
.5
the cross-sectional
area of the
iron
ring was
square inch, the total number of lines of force would be
87,000
If
-S
the
43.50area was 2
inches, the
total
flux
would be
87,000
174,000 webers.
When
the magnetic circuit consists of the
same
quality
of iron, but of parts of different cross-section, calculate the
circuit,
induction per square inch for one of the parts of the and then the induction per square inch for
the other parts will depend simply on the ratio of their cross-sections to the cross-section of the first part

Example.
inch
is
2/
An induction
in
of 100,000 lines per square
required
the
cores of the magnetic circuit
shown
in Fig. 20, the cores consisting of
Swedish
iron.
How many
ampere-turns are required ?
Solution.
The
cross-sectional area of the cores
is
.7854 =
.H05 square
inches
each.
is
sectional area of the
yoke and armature

is
The \ X \
cross-
.125
square inches each.

Since the cross-section of the cores
the smaller, the
is
induction required in the yoke and armature
100,000
.1105
1 88,400 hnes per square inch.

,
Referring to the chart, Fig. 18, the ampere-turns per

linear inch required for 100,000 lines per square inch are
34.1,
and
20.
is
for
88,400 lines the ampere-turns per inch

total
are
The
length of the circuit through the
cores
4 inches, and through the yoke and armaTherefore, the ampere-turns required
ture 3.25 inches.
for the cores are 34.1
4=
-|-
136.4,
and
for the
yoke and
armature 20
3.25
65,
and the
65
total ampere-turns for
the whole magnet 136.4
201.4.
Ans.
The
and
lines of induction are nearly straight in the cores,
for that reason the exact length of the cores
was
used in the calculation.

ever, the lines
In the yoke and armature, howafter the
bend around something
form of
the dotted lines in Fig. 20, so if" was considered as a

fair
average for the length of the circuit in the yoke and
armature.
In the above example no allowance was

leakage, or for reluctance at the joints.
made
for
28
If the
THE ELECTROMAGNET.
armature was removed even .001" from both
cores, the total length of the air
gap would be .002", and

is i,
since the
permeability
of
air
the
ampere-turns
required for the air gap alone would be
IN = Bl =
3-193
100,000
.002
62.6,
3-193
or a total of 264 ampere-turns, including the air gap, an

increase of 13.1 per cent over the ampere-turns required
for the iron alone.
19.
Efiect of Joint in
Magnetic Circuit.
circuit,
In addition to increasing the reluctance of the
an
air
gap introduces leakage and a demagnetizing action
due to the iniiuence of the poles induced at the ends of

the cores.
The
distance between the two faces of an air gap
is
not
the exact length of the gap, as the lines bulge somewhat, as in Fig. 2
1
Fig. 21.
Joints or cracks in the magnetic circuit
have the same
general effect as
demagnetization.
air
an
air
gap, introducing leakage and
gap
for
Ewing and Low found the equivalent two wrought-iron bars to be about .0012 of
an inch.

The
eifect
29
for low
of
the joint
is
more noticeable
as
magnetizations than for high ones,

attraction
the
increased
decreases the distance between the faces of
the joint, thus reducing the reluctance.
From
the
the above
is
seen the importance of having
all
joints faced as nearly perfect as possible; furthermore,
area of the joint should at least equal the cross-
sectional area of the part having the lowest permeability.
20.
It
Magnetic Leakage.
was
stated in art. 13, p. 18, that there
sulator of
is no inmagnetism and that the permeability of air is
taken as unity.
some reluctgap of the magnetic circuit, some of the lines of force must pass between the cores or other parts of the magnetic circuit, where there is a difference
It is therefore evident that since there is in the iron
ance
and
air
of potential.
may be applied by finding the relative reluctances of the magnetic circuit proper and of the path of the leakage lines.
The law
of the divided electric circuits
to magnetic circuits in this case also,
The leakage may be
calculated with great accuracy by
plotting the probable leakage paths.
The ratio between the total number of lines generated and the number of useful lines is called the leakage coThus, efficient, and is denoted by the symbol Vi.
"'
= %
to)
30
where
THE ELECTROMAGNET.
<^
<j!)i
= =
total flux,
useful flux.
flat sur'
The
faces
reluctance of the air space between two

:
is (R
A,u
logic
(ii),
but between two cylinders
it is
737
(R
=
d^
-^
(38),
where
-=
where
= = 4=
diameter of the cylinders,

distance between centers, length of cylinders.
of
is
The numerical value

as long as the ratio
is
constant for
all
dimensions
constant.
The
following table f gives the magnetic reluctance
per inch between unit lengths of two equal parallel cylinders surrounded by air and having various values of the
.
ratio -T
ELECTRIC AND MAGNETIC CIRCUIT

To
find the air reluctance
between two equal
parallel
cylinders, find the ratio
and opposite
this value in the
table
is
the reluctance per inch of length.
Therefore, the
total reluctance
between them
is
the reluctance per inch
divided by the length of each cylinder in inches.

Since at the yoke the difference of magnetic potential
is
approximately zero, and at the poles
it
is
approximately
or
maximum, the average difference of magnetic potential M.M.F. is equal to the total M.M.F. divided by 2,
Average M.M.F.
or
3-i93
^N ^
is
^^^^^
^^
Therefore, the leakage in webers
_ 92 =
1.5965
IN
>
(39)
being found in the table as explained above.
From this the leakage coejfficient Vi is found. The leakage may be included in the total reluctance by multiplying the sum of the reluctances by the leakage coefficient.
Thus,
i?=^(ii + :4 + :4).etc.
(40)
Example.
What
is
the leakage coefficient of an eleccircuit,
tromagnet, the reluctance of the magnetic

the air gap, being .05, the cores .5"
apart, center to center,
dia.,
including
3"
long,
?
and 2"
and the M.M.F. 4,000

4.
Solution.
=-
==
is
From
table,
when
=
3"
4,
is
reluctance per inch
.258,
and the reluctance
for
:^ = .086.
3
32
If
THE ELECTROMAGNET.
there
equal the total
was no leakage the flux, which is

4,000
OS
total
useful lines
would
80,000,
but the leakage
is,
(*?)
.086
therefore the useful lines are
23>2S;
80,000
23,250
56,750,
and the leakage
coefficient
^=^/37)=^6:^ =
Therefore, the total reluctance
1. 41
<^
80,000
x.4i.
Ans.
said to be .05
may be
.0705 including the leakage.
Also the M.M.F. must be increased approximately 1.41

times to produce 80,000 useful lines through the poles
and armature
of the magnet.
A high
as the
reluctance in the cores complicates the problem,
M.M.F. between the poles can not be considered as the total M.M.F. In general, the leakage may be reduced by the uniform
distribution of
the
winding over the magnetic

circuit,
circuit,
roundness and evenness of the magnetic
and the.
avoidance of sharp corners and abrupt turns.

21.
Limits of Magnetization.
When
is
the magnetizing force about an iron or steel core
gradually increased from zero, the magnetization in the
iron also increases,
though not in the same proportion,

until
it
33
reaches a point where
it is
not affected by a very
material increase in the magnetizing force.
The
iron
is
then said to be saturated, and the point at which the

duction reaches the
in-
maximum
is
called the
saturation
point, or the limit of magnetization.
The
following table * shows the various values of
B for
different grades of iron
and
steel at the saturation point.
Values of
Wrought
Cast
steel
iron
130,000
...
. .
Mitis iron
....
...
127,500
122,500
77i500
Ordinaiy cast iron
The
practical
the densities given in the above table.
working densities are about two-thirds of For practical work-
ing densities see table, p. 117.
The
relation
between the values of
H and B can be
plotted as a curve which has the general form as in Fig. 22.
(Also see Figs. 18 and 19.)
* wiener, Dynamo-Elec, Macliinery.
34
THE ELECTROMAGNET.
22. Hysteresis.
If
the
now the magnetizing force is gradually reduced from maximum value to zero, the magnetization will be
found to have a higher value in the decreasing series of

3C than in the increasing series.

again, the corresponding values of (B
35
wUl be found on the

tests of
curve at the right.
This diagram
is
taken from one of Ewing's

is
a very soft iron ring, and
known
as the Hysteresis Loop.
The
area of the loop will vary for different grades of
iron or steel, but the general form will always be the
same, and the values of
(B in
the increasing series will
never be quite equal to the decreasing value.
The
area of the loop represents the loss due to heat in

is
the iron, and
called the Hysteresis Loss,

is
which in
alter-
nating-current apparatus
is
very serious.
This energy loss
expressed by the following formula due to Steinmetz
fF,=/(B where W^=- watts

(S>
(41)
lost per cubic centimeter of iron,
= maximum induction per square centimeter, = a constant varying from .002 for soft irons
to .0045 for transformer irons.
f= number of
cycles per second,
23. Retentiveness.
That property which tends to retain magnetization is as Retentiveness, and that portion of magnetization which remains is called Residual Magnetization, and the force which maintains the magnetization is called the
known
Coercive Force.
The reason why

tion coils
rapid-acting electromagnets or inducin the
have openings
magnetic circuit
is
because
the iron
ness.
makes the
it
action sluggish
due
to the retentive-
If the
armature of an electromagnet actually touches

will stick after the current ceases to
the pole pieces,
flow through the winding.
36
THE ELECTROMAGNET.
Problems.
1.
The E.M.F.
is
of an electric circuit
is
no
volts
and
the resistance
will flow
2.
5 ohms.
How many
?
amperes of current
22 amperes.
is
through the circuit
The
?
current strength of a circuit

is
20 amperes and
the E.M.F.
circuit
3.
220
volts.
What
is
the resistance of the

II
ohms.
A coil
of wire has a resistance of 100 ohms.
What
will
.5
be the E.M.F. across its terminals when a current of ampere is flowing through the winding? 50 volts. 4. In Problem 3, how many watts would be expended
on the winding? 25 watts. 5. What would be the expenditure in watts in Problem I ? 2,420 watts. 6. How many watts would be expended in Problem 2 ?
4,400 watts.
7.
How many
amperes would be required
to
produce
S horse-power in a circuit at
8.
no
volts?
33-91 amperes.
What would be
7
?
the resistance of the circuit in
Problem
9.
3.24 ohms.
How many
Two
coils of
horse-power in the circuit in Prob3.24 H.P.

25
lem
1 ?
10.
ohms and 50 ohms
respectively
are connected in series in a 10-volt circuit.
How many
?
watts will they consume, assuming the resistance of the

coils to
be the entire resistance of the
circuit
1.33 watts.
11.
In Problem 10,
if
how many
watts would be con?
sumed
the two coils were connected in parallel
6 watts.

12.
37
Three
coils
of 5
ohms, 12 ohms, and 17 ohms
respectively are connected in multiple.
What
is
the joint
resistance?
13.
2.92 ohms.
7,
In Fig.
terminals of
p^, if p^
what would be the voltage across the = 4, p^ = 7, p, = n, and p^ = 3 ?

1.7 volts.
14.
The M.M.F.
of a magnetic circuit
is
is
1,200 gilberts
and the reluctance

will flow
.001 oersted.
?
How many
webers
through the circuit

flux of
1,200,000 webers.
is
15.
10,000 webers
obtained with 1,500

.15 oersted.
gilberts.
16.
What is the reluctance? How many gilberts will be required
to force 20,000
webers through a reluctance of .025 oersted?
500
17.
gilberts.
What
if
is
the reluctance of a magnetic circuit 10

i
centimeters in length and

section,
18.
square centimeter in cross1,800?
the permeability
gilberts
air
is
-00555 H" oersted.
How many
100 webers through an

square centimeters
19.
would be required to force gap .1 centimeter long by 2

?
in cross-section
5 gilberts.
What What
is
the
M.M.F.
in
2
?
centimeters of length
2 gilberts.
measured along one

20.
is
line of force
the permeability of an iron core in which

is
the induction
6,200, with a magnetizing force of
H= 4?
21.
/*=
intensity will
i>SS-
What
be required for an induction
of 5,000 gausses
22.
How
2
when p, = 1,500 ? many ampere-turns would be
H= 3-33required to force
16,000 lines of force through a magnetic circuit 10 inches
long and
meability
square inches in cross-section, and with per-
2,100
11-93 ampere-turns.
38
23.
THE ELECTROMAGNET.
What would be
the total flux produced by 200
1.5
/a
ampere-turns in a magnetic circuit 16 inches long,
square inches in cross-section, and with permeability 2,000?

24.
119,738 webers.
In Problem
23,
what would be the
intensity of
induction
25.
B?
5=79,825.
ampere-turns would be required to pro-
How many
duce a density of magnetization = 1,700 square inch if /= 8 and

fx,
B
?
of 55,000 lines per
81.06 ampere-turns.
26.
What would be
the density of magnetization
when
27.
IN =
What
200,
is
/=
11,
and
/x
1,300
5=
/=
7
75,471.
the magnetizing force
when
and
IN
3)o?
28.
ii=
?
1,368.
In 25, what are the ampere-turns required per

10.13 ampere-turns.
linear inch of magnetic circuit

29.
What
is
the leakage coefficient where the useful
flux is 120,000 lines
and the
total flux is
180,000 lines
7i
1.5.
30.
lines
What
is
the useful flux
when
the total flux

?
is
90,000
64,290.
and the leakage
coefficient 1.4
<î
the air reluctance between the cores of a magnet each i" diameter, 3" long, and 2" apart, center
31.
is
'
What
to center?
Jl
.055.
ampere-turns would be required in Fig. 20 to force 50,000 lines per square inch through the cores
32.
How many
when
the armature
is
?
removed
^"
from the poles, not
considering leakage
Approximately 2,000 ampere-turns.
39
CHAPTER
II.
24.
Simple Principle of Calculating Windings.
The
length of any strand which
may be wound
of
in
any
given bobbin of any shape or form depends upon two

things only,
viz.,
the
available volume
the bobbin,
and the
Let
cross-sectional constant.
= length of strand in inches,

in square inches,
V = volume of winding space

g^
= cross-sectional constant.
(42),
Then,
4=-^
^^==-^(43),
V=g'l^{^^).
the cross-section of the strand.
must not be confused with To make the meaning clear, assume that a strand of roimd insulated wire is wound in two layers on a tube, as in Fig. 24, shown in
cross-sectional constant
cross-section.
The
Fig. 24.
It will
be seen
at
the insulated wire
is
once that the cross-sectional area of g^ X .7854, in which g represents
the diameter of the insulated wire, while the actual area
40
consumed
Wliile this
is
THE ELECTROMAGNET.
equal to g^, which
is
the area of each square.
is
only approximately correct on account of

it
the imbedding of the wires,

ciple.
illustrates the
general prininsulated
wire, not-
As a matter
it is
of fact,
wire,
best to
fill
when winding with known volume with the
ing the length of the wire, and then working backwards by
using formula (43), g^

of g', a
V = j-
Then, with the known value

to suit
volume may be calculated

fill
any required
calculated
length of wire, or the length of wire
may be
which
will just
a given bobbin.
As
the resistance of an electrical conductor of constant

its
cross-section varies directly with
length,
it is
evident
in
that the resistance of any wire
which may be contained
any bobbin or winding volume may be readily calculated.

25.
Copper Constants.
Now,
it
has been found by careful experiment that a

soft
commercial
drawn copper
wire, .001" in diameter,
has a resistance of 10.3541 ohms per foot at 68 F.

Therefore, to find the resistance in
ohms per
foot for any
other copper wire, divide 10.3541 by the area of the wire

in circular mils,
a mil being one-thousandth of an inch,

the square of
(5
and a
circular mil
the diameter in mils.
Thus, a wire .005"
mils) in
diameter has a cross-
sectional area of 25 circular mils, expressed 25
CM.
(40 ^^'
Thus, ohms per foot (u'= ^'^t^^
CM.
Now
may be
suppose
it is
required to
know what
resistance p
in-
obtained in a given bobbin of volume V, with
WnVDIXG CALCULATIONS.
sulated copper
4I
that
wire of diameter A.
Assume
in.,
it is
the
available winding volume

eter of the wire
is
V=
1.68 cu.
and the diamcovered

diameter up
.010" (10 mils), and that
with cotton insulation

to .014"
which brings the
total
= g,
since
=A+
(42),
2.
(46)
Then, by using formula

4,
V =^= g''
1.68
-2
1.68
.014
.000196
8>S7o
715
feet.
By formula ohms per foot ^
(45),
= 10.^1:41 = ^l, CM.

is
^V- = ^^^ =
10.^^41
10=
io.^c4i
100
i3S4i^^^
The
resistance p
equal to the product of the

of feet.
ohms
per foot
times the number

.-.
=
is
.103541
X 71S
74 ohms.
the reader thor-
The
foregoing
given merely to
make
oughly familiar with the underlying principle.
The meth-
ods of calculating volumes of various forms of bobbins

will
be given in subsequent pages.

26.
Most
Efficient
Winding.
The most
efficient
is
winding for developing or absorbing

in
magnetic energy
one
which the resistance
is
low and
is
the turns are numerous, since the magnetomotive force

proportional to the ampere-turns.
Since the current
is
equal to the voltage divided by the resistance, the lower

the
resistance
the greater will be
the current flowing
through the turns of wire, thereby increasing the ampereturns,
and where a constant resistance
is
required,
if
the
42
number
THE ELECTROMAGNET.
of turns
may be
increased, the ampere-turns will
increase also, in direct proportion to the
number
of turns.
For this reason, the wire with lowest resistance should be used and as copper fulfills the practical requirements,
;
all
wires will be understood to be of copper unless other-
wise specified.
The ampere-turns depend upon two

make mean
volts.
things only,
viz.,
the voltage and the resistance of the average turn.

the
To
assume the resistance of the or average turn to be one ohm, and the voltage loo According to Ohm's law, the current in amperes
clear,
,
meaning
would be
/
E ==
pi
loo
ICO amperes.
100 amperes and i turn= loo ampere-turns =ÎN. assume lo turns of wire instead of i turn. The ance
will
Now
resist-
increase
directly with
the
therefore the resistance would be lo

rent lo amperes,
number of turns, ohms and the cur-
and consequently there would be loo
ampere-turns as before.
In calculating the above, the average turn must always
be taken, for the resistance of the turns increases directly

as the diameter increases.
27.
Circular Windings.
round core
is
the most economical form, as more
turns of wire
may be wound
same
thereon with a given amount

Since the
of copper, for the
cross-section of core.
leakage from core to core, for equal

is
mean
distances apart,
proportional to the surface of the core, the round core
WIA'D/NG CALCULATIONS.
has a decided advantage, as
for equal sectional areas,
it
43
has the minimum surface

to
and there are no sharp edges

it is
facilitate leakage, therefore
very commonly used.

is
The winding on a round

der,
core
really a hollow cylin-
and
its
volume
is
equal to ttMLT.
Where
M=
Z=
TT
T=
=
(See Fig. 25.)
the average diameter of the winding, length of winding, thickness of winding, ratio between diameter and 3. 1 41 6
circumference of
circle.
Fig. 25.
M==
Where
J?
+d
(47)
T = D-d
+
sleeve.
(48)
D = true diameter of winding,

^
diameter of core
(49)
and the volume
r=7rZ
XP.
-).
(5)

44
THE ELECTROMAGNET.
also
M=D-T=T+d, T=D-M=M-d,
= 2T+d = M+T,
(S5) (56)
(si)
(S2)
(S3)
(54)
d=D-2T=M-T.
(57)
(58)
Since the unit of length throughout these calculations

will
reduced to ohms per

inches,
be the inch, the resistance of conductors must be inch, and circular mils to circular
one circular inch being 1,000,000 circular mils. Thus, the cross-sectional area of a wire .001" in diameter
is
one circular mil or .000001 circular inch.
Formula (45) then reduces
to
ohms
.86284
per inch,
10.3541
12
CM.
CM.
(59)
Substituting circular inches for circular mils,
<u
.00000086284
(60)
Substituting the value of
V in
(42),
^-=
Since the resistance p
is
^^
d")
(60

of inches,
ohms per inch times the number
.00000086284 ttZCZ)^
.0000027107 Z(Z'''
(/")
4FAr
Assigning ^ to .0000027107, and
(62)
^to
iroj",
^=2'
(63)
formula (62) reduces to
'
45
-^7
i?
(64)
Let
then
p
= ^,
(66)
(6s)
= -^-^(^-^)
4
is
= rmLT.
(67)
Hence
^
I,
is
the resistance which will be obtained
when
MLT ^
factor,
and
the ratio between winding volume and
resistance.
Therefore,
K
is
it
is
the resistance factor,
g"^
is
the space
and
the combined resistance and space factor.

follows that
From
(67)
tables on pp. 137147 are calculated on this prinand the proper wire to be used for the required resistance is found opposite the value of R.
ciple,
The
As
in practice the value of
sizes of wire, the smaller wire is taken
foimd for
between two and a new value by the formula deduced from (6g).
R usually falls
RL
is
This gives the actual theoretical diameter of the winding
when a standard
insulated wire
\.
used.
Example.
(a)
Given D =
d=
.43.
Z=
2.
What
What
size of single silk-covered wire
must be used
?
so that a resistance of 500

{b)
ohms may be obtained
will
be the actual value oi
Dl
46
Solution.
THE ELECTROMAGNET.
From
(69),
R-
4P LiEP-d'^
2,000
2(1 -.185)
2,00
1.63
228
In the table, the nearest

is
R value
for silk-covered wire
Ans. {a). 1,480 opposite No. 35 wire. The actual diameter of the winding will be
D
2,000
.185
2,960
V.676-1-.18S
(B).
=
To
V.861
.928.
Ans.
find the internal diameter of the winding, use formula
derived from (70),
''=V^^Âs
previously stated, the factor
<')
R may be better underThe
stood as a combined space and resistance factor.
resistance of a conductor varies inversely as the square
of
fill
its
diameter, and the length of a conductor that will
a given winding volume varies inversely as the square

its
of
diameter also.
factor,
-^,
Therefore, as a combined space and

varies
resistance
R
is
inversely as A^
A^
A*.
Hence,
R=
not considering insulation.

considered, as
it
When
insulation
always should be
in practice,
^Â^(A+77'
The
following
is
*^^^^
a simple method of finding
MT,
as
the diameters do not have to be squared
4/
Thus,
T= ^~ ^ (48), 00 = D
1.
and
M=T+d
X
.6&
(52).
ÎAA
.32
=T .^ = d :68 = ^
.:
MT=. 32
=.2ij6.
The
length of wire in inches in any winding
is
L = -^r28.
(73)
Points to be Observed in Practice.
be exercised
In calculating magnet windings, the utmost care must in measuring the exact dimensions of the
will
winding volume, otherwise the results
vary greatly.
varies
When
insulation,
it
is
considered
that
the
resistance
inversely with the square of the diameter of the wire plus

it
is
readily seen that these factors are
most
important.
The
variation of one-thousandth of an inch in
the thickness of the insulation on a No. 35 wire will cause
be wound
taken,
a variation of 23 per cent in the amount of wire that in a given winding volume.
Also, the diameter of the core
i.e.,
may
to
wind on must be
core,
slight,
the diameter after the paper or mica or other
insulating material has
been wrapped around the

though
for the thickness of the insulating sleeve,
makes a great
difference in the
volume of the winding.

also
The
true outside diameter
must
is
be accurately meas-
ured for the same reason as with the core, but with the
outside diameter the variation
even more marked.
48
THE ELECTROMAGNET.
was calculated
In the case of the bobbin in Fig. 26, if the winding to contain 950 ohms of No. 36 single silkcovered wire, using the actual dimensions of the bobbin
-:=_")
00
^"
Fig. 26.
wound
without considering the insulating sleeve, and then it was apparently full of wire but really to but .95 inch
diameter, the resistance would be but 795
difference of 19.5 per cent.
ohms, or a
Again,
if
the above wire
operation so that
was stretched during the winding the diameter was reduced from .005 " to
.0049
",
a difference almost too small to detect with a
micrometer, there would be an increase in resistance of

7
per cent.
insulation on the wire is 3 mil another error of 4.6 per cent in the increase, there will be
Assuming
that
the
weight of the wire in the winding.
Thus
it
is
seen that too
much
precaution
cannot be
observed.
The proper way

winding
is
to
which
will just
to measure the outside diameter of a measure the length of a piece of paper go around it, and then dividing it by ir.
Ij,
Thus,
D = (74), where =
length of wrap of paper.
WINDING CALCULATION^.
If
40
be found
to
measured with calipers the winding

it
will
be
slightly elliptical, rendering
impossible to measure
the exact diameter accurately.
29.
Formulae for Turns, Resistance, and Ampere-turns.

of turns
is
The number
which may be contained

g"^
;
in
any
bobbin of any form
equal to one-half the longitudinal-
sectional area divided
by
thus,
^=ânce of the average turn, or

p
(75)
The resistance may be easily found when the turns are known, by multiplying the number of turns by the resist-
= TTis/'MN = KMN.
number
of turns
(76) (77)
To
find the insulated wire
and resistance when dimenare given,
sions of the bobbin and the
proceed as follows
Find the value of g^ by formula deduced from
(75).
^^=#;
(78)
then from the table select the next smaller size for g''' and calculate the new diameter for this value of ^^ by the formula
and with by formula
^ = ^V^, the new value of D the resistance

(67),
(79)
is
calculated
RL{iy-cF)
:
50
THE ELECTROMAGNET.
Ampere-turns are calculated by the formula
/^=^.
KM representing the resistance of the average
this
it is
(80)
turn.
In
seen that the length of the winding makes no
differ-
ence in the number of ampere-turns, for as the length and

turns increase, the resistance increases and the current
decreases in the same

constant for
ratio,
thus keeping the ampere-turns
constant voltage.
With constant
current,
however, the ampere-turns and voltage would increase

directly with the turns
and
resistance.
The
length of the bobbin affects the heating of the
winding, and will be explained fully farther on.
In order to find the resistance of a coil of wire having
a definite number of ampere-turns with a certain size of

wire,
it
is
necessary to find the outside diameter of the
winding.
This
is
found by the following formula derived

2
from (80):
p
produce a certain winding space with given
To
find the exact diameter of wire to

in a given
number of ampere-turns
voltage, use formula derived from (63)
and
(80).
^=i,.
(63)
IN = /i\^=-^.(8o)
Î
T=\.
(83)
D^l^^d.
(52)
(79) (84) (S3,
J/ =
r+ ^
=^ +
+ dj,
d.
MLT = L
f(f..)]
Ng' (^'
(86)
the resistance or weight

directly
Therefore, the MLT'vs, found directly by Formula (87) of the winding is then found
;
by
R and w,
by substituting the values as in formulte (88) and
MLT and
(89).
multiplying
Lb
where Lb
= "^^^
(^^^
J-i
'"^
(89)
weight in pounds.
resistance
w
ance.
is
the combined
and space
factor,
and
bears the same relation to weight that
bears to
resist-
Thus, Lb
= wMLT.
(90)
'=y
when
Q =
(91)
=^2
(92)
ohms per pound
for insulated copper wires.
30.
Constant Resistance with Variable Insulation.

the resistance of
When
will
a bobbin
is
constant with a
given size of wire, the outside diameter of the winding
be increased by an increase in the thickness of the
insulation, but the
number
of turns will decrease as the
5^
THE ELECTROMAdNMT.
thickness of the winding increases, because the length of

the average turn increases with the diameter.
In this case
excepting D.
is
all
dimensions of the bobbin are constant

resistance of the wire
The diameter and

of turns for
constant, but the thickness of the insulation changes.
The number
will
any thickness of insulation
be
"=^1
As an
illustration,
V-
tg+^-.
;?
()
^J-
assume a bobbin with

total
d=
.43
and
L=
2,
wire No. 36, and resistance 500 ohms.
If the insu-
lation
on the wire brought the

If
diameter
.009,
g up to .008",
resistance, but
the turns would be 7,030.
the turns would
be 6,600, or a loss of over with a difference of .001"

31.
6%
for the
same
in the value of ^.
Layers and Turns Per Inch.
To find the number of layers of wire that may be wound in thickness of winding T, use formula
T =
I
(94)
The number
of turns of wire per inch,
m=
Thus,
N= mnL,
_
gi
(95) (96) (97)

(98)
N mL N
53
^=
^,
^'
T
n
(99)
= ^,
(100)
(lOl)
Where ^=
gi
vertical value of wire
and
insulation,
= lateral
value of wire and insulation.

are
often found
The above
practice.
formulae
convenient
in
32.
Windings with Wires Other than Copper.

winding with wire of
differ-
To
find the resistance of a
ent specific resistance than copper, multiply the resistance

of copper
by the
coefficent for the other kind of wire.

silver wire
;
Thus,
8 15^
German
has a specific resistance
18.76 times that of copper
hence the ohms per foot equal

18.76
10.3541
194
The
and
simplest
method
is
to find
if
what resistance would

copper wire were used,
be obtained
in the given
bobbin
tlien multiplying
by the
coefiicient for the other wire.
33.
Small Magnets on High-Voltage Circuits.

it is
In practice
often required to place a small electroin a high-voltage circuit,
magnet or solenoid
that the winding
is
and hence
the resistance of the winding must be very great in order
may
not be overheated.
This resistance
given winding volume with the finest insulated copper wire on the market. It would be impracticable to use a wire of greater
often too high to be obtained in the
54
resistivity
THE ELECTROMAGNET.
than copper, as the turns, and consequently the
fall
ampere-turns, would
short of requirements
therefore,
the usual method
is
to
wind part of the magnet with
copper wire, and the balance with a high-resistance wire,

using enough of the copper to give the greatest number of
turns,
and
still
leave enough
wire winding to be
room for the high-resistance wound over the copper wire winding.
Since the resistance
If
we now represent the
outside diameter of the copper
winding, and consequently the internal diameter of the

high-resistance wire winding
total resistance
and the
i?i,
by X, and let p equal the value for the high-resistance
wire winding by
the formula becomes
^=
^=
X'
R{X^-d-^-\- Rîp'
RX''
X^.
Rd""
+ R^cP - R^X\
R^jy.
RX^ - R^X^ =
^ + Rd^-
^ + Rd^- R^D" ^
R
.:
X=
Changing
signs.
.1
Ri-R
\:ff
f.
(102)
^
^
Example.
.43,
55
A bobbin
with dimensions Z>
i,
</
L=
1.5, is to
contain 10,000 ohms, with the greatest

wire,
amount of No. 40 copper
and
just
enough No. 40
30% German
limiting
silver wire to g^ve the resistance within the
O.D. (outside diameter).
Solution.
X
^
R for No. 40 S.S.C. coppier wire = 12,690. R for Xo. 4030% S.S.C. German silver wire =
12,690
/
28.1
=
(
356,590.
'356.590
740,000
+ 2,348
7\
~ =
and there
'
/^;6,i;qo -^3 .av
1 ^.690 356.590 29,015 I ^
343,900
V.9525 ^^
^
.976. ^'
Am.
Therefore, wind to a diameter of .976 " with copper wire,

will
be
just
enough room
left for
the
German
silver wire to bring the total resistance up to 10,000 ohms.
34.
Resistance Wires.
The
resistance wires most
commonly used in connection
with electromagnet windings to increase the resistance
may be used on a circuit of high voltage, are German silver and Climax. The commercial German silver wdres are of two grades The 18% wire has about viz., 18'^ and 30% nickel.
so that they
18.8 times the resistance of copper, 28.1 times the resistance of copper.
and the
30%
wire
The temperature
.00017
P^'"
coefBcient for
German
silver wire is
degree F. for 18%, and .0001185 per degree
56
F. for
THE ELECTROMAGNET.
30%.
Specific gravity approximately 8.5.
of
Climax
Its
its
wire has about 48 times the resistance
copper.
temperature coefficient
specific gravity 8.137.
is
.00042 per degree F. and
Under
similar conditions, the carrying capacity of
two
wires of equal diameter but of different materials varies

inversely as the square root of their specific resistances.
tests
The German silver wire table on p. 148 is based on made by a well-known manufacturer the ohms per
;
pound are based on an average specific gravity of 8.5 for both 18% and 30%. As the resistance of German silver varies in the same specimen, the table is to be considered as commercially
correct, but not absolutely correct.
35.
One
Coil
Wound
Directly Over the Other.
Simply figure for the

formula (79),
total
number
of turns,
using
D = 2Ng' L
'
The
(66).
total resistance
can then be calculated by formula
"
=
i
is
This
is
not absolutely correct, as paper
usually
re-
wrapped between the two windings.

sistance of each coil
Therefore, the
may be
is
calculated, using the
of
the
first coil
plus paper, for the
d of
the outer
coil.
The
latter
method
necessary
when
different sizes of
wires are used on successive windings.
36.
Parallel Windings.
5/
In certain types of electromagnetic apparatus, especially

those used in telephone and telegraph work, the coils consist of
two parallel wires coiled simultaneously, each being
insulated from the other, thus forming two separate and

distinct circuits.
In telegraph work the

poljirized relays,
coils thus
wound
are used on
and
either act separately, in conjunction

;
with each other, or in opposition to each other

the current
is
thus,
when
flowing through one coil only,
it
acts with of current
a certain strength, and
when the same amount
flows through both coils in the same direction, the

will
magnet
have approximately twice the strength.

if
Again,
the
same strength of current flows through both

but in opposite
directions,
coils simultaneously,
there will
effect of
be no magnetic action whatever, as the magnetic one

coil is neutralized
The
latter
by the effect of the other. principle is employed in the making
of noncur-
inductive resistance coils, only in this case the
same
rent passes through both coils, thus really forming but
one
coil.
The
work
differential relay is also
used in telephone switch-
boards, but the winding referred to above for telephone

is
classed under the heading of Repeating Coils.
The telephone
it
repeating coil
is
really
an induction
coil,
but the fact of having the two windings in parallel gives

the condenser effect also, so that the inductive effect
is
electrostatic as well as electromagnetic.
To
find the resistance
of each wire in parallel windsingle
ings, calculate the resistance for a
insulated wire,
58
THE ELECTROMAGNET.
2,
and divide by
since
it
is
really but
one complete
coil
with half the resistance in each branch.
A good method of finding the respective resistances of two or more wires wound simultaneously and in parallel is to calculate the number of turns, and then apply formula = KMN, to each wire separately. Another form of winding which may be classed under Parallel Windings is where two or more insulated wires are wound in parallel and the respective ends electrically
(77) p
connected together so as to act as one conductor.
This
is sometimes necessary where a wire of large cross-section is to be wound upon a very small core, as several finer ones will have the same cross-section and still be
flexible.
Referring to Fig. 8 and Fig.

in
9,
p.
10, the resistance
4 times their resistance when connected in multiple, or the resistance of any number of wires connected in series is equal to the square of the
two wires
in series is just
number
of wires times the resistance in multiple, or

p.
= n^Jr.
number
(103)
of wires in mul-
Likewise, the resistance of any

tiple is
equal to the resistance of the wires in series divided

of the
by the square
number
Jr
of wires, or
= ^.used
is
(104)
of course equal
series.
The
total resistance of the wire
to the resistance of all the wires
connected in
is
Likewise, the resistance of each wire

total resistance divided
equal to the
by the number
of wires.
37. Joint Resistance of Parallel Windings.
Sg
To
find the joint resistance of

coil,
any number of wires of

coil.
equal resistance in a
divide the resistance of each
wire by the number of wires in the
Thus,
(105)
Jr=^.
To
resistance
find the resistance of each wire, multiply the joint
by the number of
wires,
p=/rn.
The
total resistance is
(106)
equal to the joint resistance
multiplied
by the square
of the
number
of wires,
Ps= n^Jr.
38.
(103)
Size of
Relations Holding for
Any
Wire
and Winding Volume.

If the ratio
between wire and insulation was constant
for all sizes of wire, the following laws
would hold
for a
given winding volume

1.
For any
size of wire, the length of wire,
and conse-
quently the
the wire.
2.
number
of turns, varies as the cross-section of
The
resistance varies as the square of the
number
of turns, or inversely as the square of the cross-section of
the wire, or inversely as the fourth power of the diameter

of the wire.
3.
The
current at any given voltage varies inversely as
the square of the

resistance.
4.
number
is
of
turns, or inversely as the
The weight of wire
constant for any size.
6o
5.
THE ELECTROMAGNET.
The magnetic
efiEect
varies as the current muItipUed
by the square root
of the resistance, or as
the square of
the diameter of the wire.
In practice, however, there

large
is
a wide variation between
and small insulated wires, as may be seen by consizes of insulated wire the resistance is
it is
sulting the insulated wire tables.
For every two

half as
approximately doubled, and for each consecutive size
much
again
or, in
other words, the resistance
in-
creases approximately 501^ for each consecutive size of

insulated wire with the
same
insulation,
and increases
approximately
25%
for half-sizes.
39.
The American Wire Gauge.

in the
(B.
& S.)
Wire gauges are arranged

series.
form of a geometrical
The sizes are determined by the diameter of the circles, which may be placed between two lines at a given angle, in such a manner that the circles will just touch one another and the bounding lines, as in Fig. 27.
33
J*
! Ij?ImY3j1W)
Fig. 27.
The American wire gauge

series in
is
based on the geometrical
which No. 0000
.005" diameter. 40".*
The
*
.46" diameter, and No. 36 is angle of divergence is 6-36'is
W.
J. Varley.
As
there are just 39 sizes between No. 0000
6
and No.
36, the factorial diiference (or progression ratio) between
any two sizes
is
.005
= V Q2 =
1.
1220^2.
Thus, to find the diame.ter of any wire,
^=
i.i22932('^3)
'
(i7)
where
This
is
= the
desired gauge number.
most conveniently worked out by means of
logarithms.
Log Log
.-.
.46
1.
= =
122932
log
A=
1.6627578. .0503535.
log .46
[(j-
3) log 1.122932]
(108)
(109)
1.6627578
is
lis
+ 3)
.0503535].
Example. What Solution.
the diameter of No. 30 wire?
.-.
Log A Log A
= = A=
.6627578 1.6627578
.0100252.
(33
1.
.0503535).
6616655
2.0010923.
Ans.
The
exact diameter of half or quarter sizes

rule.
may be
found by the same
To
find the
gauge number corresponding

:
to
any diam-
eter of wire, proceed as follows
Since log
="1.6627578
J
[(j
+ 3) 0503535]3.
(109)
1.6627578
-log
0503535
(no) '
^
62
Example.
THE ELECTROMAGNET.
What
size in the
?
American Wire Gauge
is
a wire .0085" in diameter
Solution.
^
1.6627578
3.9294189
S3S35
or approximately No. 31.4.
Ans.
In the American Wire Gauge the ratio of diameters for

every six sizes
that
is,
is
nearly
2,
the exact ratio being 2.0050

is
+;
;
the diameter of No. 30

is
twice that of No. 36
No. 18
twice the diameter of No. 24, etc.
The
cross-sectional areas of the wires vary in the ratio
of nearly 2 for every three sizes.
Thus, No. 37 has twice

it is
the cross-sectional area of No. 40, etc.
This
is
important to remember, as
found very con-
venient in estimating mentally the size of wire to have
a certain resistance when the resistance and length of another size of wire are known.
The
will
resistance of a bobbin, however, will vary nearly
two sizes of wire, since there heavy wire in a given volume than there would be of finer wire in the same volume. Thus, a bobbin containing 500 ohms of No. 34 wire would contain approximately 1,000 ohms of No. 36 wire.
in the ratio of 2 for every
be
less length of
From
the data given in the " Explanation of Table," on

:
page 136, the following relations hold
Pounds per foot Pounds per ohm
= =
3.0269 A^
(i
1)
292,400
A*.
i}^'^)
63
= "^^^ Feet per ohm = 96,585 A^. Ohms per pound = ^

Feet per pound
'
(113)
(114)
(i is)
,
^, Ohms
per foot
.000010^1:41
tt^
"
("^)
,.
40.
Thickness of Insulation.
To
wire
find the thickness of insulation permissible,
on a
when
the exact resistance has been calculated for

:
heating or other conditions, proceed as follows
From
(63),
(46),
K=
^
i5,
/.
and from
Now
^=
J(65)
=A+ = ^.(^(73).
v/;
-.--
("7)
t=
in
^_^_^_A,
(119)
which the value of
R from (69)
has been substituted.
41.
Ratio of Weight of Copper to Weight of Insulation.

specific gravity of
The The The
copper
is
8.89.
specific gravity of cotton is 1.377.
specific gravity of silk is 1.03.
64
THE ELECTROMAGNET.
values for cotton and silk being taken
The
tightly
when wound
specific
around the wire.

are
Since the weights

gravities,
proportional
to their
8.89 A^-
is
the relative weight of copper, and
1.377
wire,
S is the relative weight of cotton in 2 representing the cross-sectional area

.-.
the insulated
of the cotton.
S=^'8.89
A""
A2.
(120)
The percentage
of copper in a cotton-insulated wire is

,
.
"= 8.89 A^+ 1.377 S'

and
for silk-insulated wire, per cent of copper
^"'^
_
Therefore, to find the
8.89 A"
"'~8.89A2-h 1.035'
ohms per pound
for
(IZ2)
any insulated
wire, first find the percentage of copper in the insulated wire,
and then compare with the weight
for bare wire.
Example.
What
..
is
the
ohms per pound
for
No. 36
?
silk-covered wire, with two-mil increase for insulation
Solution.
g''
A^
8.89 A^
= = S=
.000049 .000025
8.89A^-H.03S
From
the table
^
fi
= "T^HÎif = 9%
36 bare wire
.90
is
.000024 .0002223
copper.
for No.
5,473.
S>473
4,926.
Am.
= ohms per pound for bare wires. ohms per pound for insulated wires.
The weight
of
6$
copper in the insulated wire

copper, and then
may be
found by multiplying the weight of the insulated wire by

the percentage of
by subtracting the
weight of copper from the weight of the insulated wire

get the weight of the insulation
of weight of insulation
is
;
we
or, since
the percentage
the reciprocal of the weight of
copper,
it
may be found
after the
same manner
as finding
the weight of copper.
The
tables
on pp. 138 and 147 are calculated on

is
this
principle.
The above
the only sure
method of computing the
relative weights of
copper and insulation, and the com-
parison of the tables so deduced, with other insulated-wire

tables, will
that the
show a great discrepancy in the latter method of assuming the weight of copper
is
also,
to
be
90 "fo of the gross weight

insulated
very
The percentage of weight of German silver wire will be

8.5 A^
8.5 A=
-I-
much in error. German silver in
a cotton-
1.377
G.S.
(123)
^^^^-^^'^^
8.5A'+t.o3S
==^^-^-
(''^)
The ohms per pound for any grade of cotton or silk may then be found after the manner of solving for copper wire, by consulting the German silver
insulated wire
wire table on p. 148.
The weight
therefore, the
of insulated wire in a
winding
is
obviously
equal to the resistance divided by the
ohms per pound

fac-
combined weight and space factor may be

of the insulated wire.
found by dividing the combined resistance and space

tor
by the ohms per pound
Thus,
66
THE ELECTROMAGNET.
(91)
'^=^J-
Therefore, to find the weight of wire in a winding,
when
the size of wire and the insulation are known, multiply

the
MLT of
the winding
by the combined weight and
space factor w.
Thus,
in
Weight
pounds
= wMLT.
(90)
42.
Weight
of Insulation to Insulate
Any
Wire.
The weight
of cotton that will insulate one

is
pound
of
bare copper wire
equal to
1.377s
8.89 A^
^
_ "
.1549 S
A^
.1159
'
and
for silk,
1-03
8.89 A^
A"
Therefore, the weight of insulation in pounds that will

insulate
is,
any weight of copper wire
to a given mil increase
For
For
cotton,
C^='^-~
i'^
in
(125)
(126)
silk,
= lLl^^,
pounds.
where
= weight of bare wire

size
Example.
Given the
A=
A.
and weight of a bare copper

it
wire, to find the weight of silk necessary to insulate
to
.002" increase.
Let
.005 500.
No. 36 B.
&
S.,
Solution.
67
^ = A + = .005 + -002 = .007. % =^^ A^ = 000049 .000025 =.000024.

.
From From
TFrom
(46),
(120)
/
(126)
^
/-s
S^ =
c^
<"
.1159 SA. 2Z
^^
.0013908
.000025
iH_ =
^ 1-^63 lbs. Ans.

.
43.
General Construction of Electromagnets.
The
Fig. 28.
usual form of
electromagnet
is
that
shown
in
Fig. 28.
The
steel,
cores, yoke,
and armature are made
of
iron or
usually soft Swedish iron, for small electromagnets,

iron,
and cast
Cast
wrought
iron, or cast steel for large ones.
steel is
namo
ally
field
used extensively in the construction of dymagnets. A magnet of cast steel is often actufirst
cheaper as to
cost than one of cast iron, owing

it costs more per pound. be considered the economy
to the saving in weight, although
Besides the
of space
first
cost
is
to
and weight, which

;
is
veiy important in some not so
forms of apparatus
also, the cost of the core is
68
THB ELECTROMAGNET.
important as the decrease in the cost of wire where a
good quaUty
of iron or steel
is
used, especially in fine
wire windings.
The washers
ber, fiber, or
or flanges are usually
made
of
hard rub-
wood.
Where
brass spools are used, insu-
lating material,
such as paper or linen impregnated with
Sterling varnish, is placed over the brass tube
and against
silk
the washers.
The
about
wire
it
is
usually insulated
by winding cotton or
spirally, so that
adjacent turns in the winding
may
not come into electrical contact with one another.

is
Before the wire
wound on
to the bobbin, the core is
insulated with paper, fiber, rnica, or the insulated linen
mentioned above, according to the voltage to be applied

to the winding.
The
wire
is
then
wound on evenly
in layers
by
revolv-
ing the bobbin on a spindle and guiding the wire by hand.
The reason why the wire should be wound evenly in

is,
layers
that
it
is
necessary to distribute the electrical stresses
uniformly throughout the winding, thus avoiding shortcircuits. If the wire is wound on carelessly, or " hap-
hazard," as
it is
sometimes
called,
some
of the first turns

later
may lie
adjacent to others which were
wound on much
in the operation, thus causing a large proportion of the

total voltage to exist
between these turns.
a puncture or " breakdown "
The result is when comparatively high
voltages are used on the coils.

44.
Insulation of Bobbin for
High Voltage.
be thoroughly insulated with paper shellacked to the brass for
When
bobbins are
made
of brass, they should
69
low voltages, but for high voltages special precautions
must be taken.
The tube should

fringed at the ends
first
be covered with several wraps
of the Pittsburgh Insulating

;
Company's insulating
linen,
and the end flanges should

slits
also
be
covered with several layers of the same material, care
being taken to have the

at least
or cuts in the linen washers
90 degrees
apart, so that there

It is better to
can be no leak-
age at these points.
assemble the linen
washers before the brass washers, as then the linen washers do not have to be
cut.
The Unen washers
should be placed over the wrapping of linen on the tube,
with the fringe between the metal washer and the linen
washer.
If the inside terminal is to
be brought out
at the top
more insulating washers between the terminal and the end of the winding.
of the winding, there should be several
The
terminals should consist of flexible rubber-covered
conductor, the size varying with the size of the wire in

the winding.
The
coil
should be thoroughly baked out and dipped in
the Sterling Varnish Company's "Extra Insulating Varnish " until it is thoroughly permeated by it, and then
baked
linen
until the varnish is dry.
The winding
should also be covered with insulating and treated with Sterling varnish.
Large windings consisting of fine wire are usually covered with heavy cotton cord for mechanical protection.
Press board and Fuller board are also used for low
voltages as insulating washers and covers for the winding.
70
TtlE
45.
ELECTROMAGNET.
Theory of Magnet Windings.

of an electromagnet,
The winding
when evenly wound

to say, the direction
in layers, consists of helices, the direction of the turns
being alternately right and

of the turns
left
that
is
on one
layer, instead of
being at right angles
to the core, will incline slightly to the left, whereas, in the
next layer, the inclination will be to the right.
Fig. 29.
Fig. 30.
At the point where adjacent helices cross one another

they appear as in Fig. 29, but diametrically opposite on the winding the turns of the upper layer sink into the
groove between the turns of the layer beneath

Fig. 30, then gradually leave the
it,
as in
groove until they reach
the highest point again, as in Fig. 29.
Fig. 31.
Where
the imbedding
occurs
the following relations
hold (see Fig. 31):
72
THE ELECTR0MAGNE7\
of loss
The percentage
due
to this is equal to the loss in
turns per layer divided by the turns per layer, or
per cent
=
m
The
lateral value of
is
greater than the vertical value

is
as just explained, but there
another variation due to the

to the bobbin, the vertical
compression of the insulation, which has to be considered.
When
tension
the wire
is
wound on
is
much
greater than the lateral tension,
and the
flattening of the insulation vertically

laterally
;
makes
it
spread out
thus there are less turns per layer than calculated,
but more layers than calculated, were this fact not taken
into consideration.
However, the turns and resistance

the same as calculated.
will
be approximately
In practice, the value of g^

of the diameter of the wire
is
equivalent to the square
with a ratchet-stop
to 147 are
and insulation as measured micrometer, and the tables on pp. 138

this principle.
based on
46.
Paper inserted into the Winding.

coil,
In winding a
is
and especially
to
if
fine wire is used,
it
found necessary to insert stout pieces of paper occasion-
ally
between the layers
form a bridge to keep the

grooves appear which are
winding smooth, otherwise

time the winding will lose
in layers.
little
due to the unevenness of the insulation, and 5n a short

all
semblance of being wound
While
paper
it
is
almost absolutely necessary to insert this

it is
in the winding,
disadvantageous, as the available
winding volume
paper inserted.
This
is
73
reduced
in exact ratio to the
volume of
may be
and
if
appreciated
if
the paper be
removed from
the winding and wrapped about the core, thus forming a
new
this
d,
the volume of the winding be calculated with

difference will be noted in
thin, strong
new d value, a very marked

Hence, only very
the volume.
paper should
be used, and then as sparingly as possible. Paper inserted in the winding thus decreases the ampere-turns by increasing the outside diameter, and consequently the resistance of the average turn.
By
increasing the outside diameter of the winding, howradiating surface

is
ever, the
increased for the
same
resistance, although the increased thickness of the
winding
may
offset this in
most cases.
47.
Duplex "Windings.
its
This winding derives

wire
silk,
is
name from
the fact that bare
coiled into the winding together with a strand of
which insulates adjacent turns from each other latThe layers are insulated from each other by As these windings are made by autosuitable paper. matic machinery, they are also called Machine-wound
erally.
Magnets.
Many more
wire, in this
turns are obtained with the
same length
of
form of winding, than with the common form, as the insulating materials occupy less space.
In the covered wire windings, the insulation
ratio of insulation to wire varies.
is
constant
for nearly all sizes of single-covered fine wire, while the
74
THE ELECTROMAGNET.
may be
constant or
In the duplex winding, the ratio

variable, as desired,
inch.
by the adjustment of the turns per
mMzmmmMMMm,
Fig. 3a.
The
the
if
silk lies
between the wires as shown
in Fig.
32
thus the wires
may be much closer than if the silk lay on common center line AB, and very much closer than
the silk were wrapped about the wire.
In the case of the duplex winding,
= C^ + S, ^=A + /',
gi
where
P= paper allowance.
obvious that there
is
^^=(A + 5)(A+/'); S = silk allowance,
(127) (128) (129)
It is
no imbedding of the wires
in
this case,
but
all
other relations hold as given for covered
wire windings.
The winding volume consumed by

Vj,= ttMLPh,
the paper
is
(130)
and the volume consumed by the
silk
space
is
V,= irMLSmn.
The windings
their
(131)
are
wound
in multiple
on a tube of paper
to 12
or other insulating material, and sawed into sections after
removal from the automatic machines. From
sections are
wound
simultaneously, according to the length
of the winding.
The
sections are slipped
on
to cores
and the washers

its
forced on to the bobbin as in the
common method.
high
The
principal
features of this winding are
efficiency
and cheapness of production.

Other Forms of Windings than Round.
is
48.
Since the round form of winding

all
the most
common,
in the
terms are
made
to apply to that type,
and when other
forms are used, the formulae so arranged as to read
same terms
as those applied
in
the calculation of the
round winding.
The
other forms for which formulse are here given are
as follows
Fig. 33.
Fig. 34.
Fig. 35-
windings on
Windings on square or rectangular cores (Fig. 33), elliptical cores (Fig. 34), and windings on
which
is
cores, the cross-section of
square or rectangular
with rounded ends, as in Fig. 35. It is evident that since the wire constants are fixed,
that
is
all
necessary
is
to express the
case, in terms of
It is to
MLT,
winding volume in each for any form of winding.
be observed also that the winding thickness
and the winding length
are constant, no matter what the
form of the winding, the one point to accurately determine
76
THE ELECTROMAGNET.
M, which
is
being the mean perimeter factor

of the
the diameter
circle.
mean perimeter when

will
in the
form of a
This
ter,
be referred
to in all cases as the
mean diame-
regardless of the form of the winding.
49.
Square or Rectangular Windings.
When wire
is
wound upon a square

form
arcs, the radii of
or rectangular core,
the corners of the winding are not sharp like the corners
which are equal to 36 shows this principle. The four sections formed by the corners would therefore form a circle.
of the core, but
the thickness of the winding.
Fig.
77
(132) (133)
and
MT=IT
(A - 'Ô - -8584 T^, _A_
_ (A A :.
'iid^
- -8584^
(134)
M: (A A
-'d^d^
.8584^
(135) (136)
r=
2
^_
From
(67),
.
A4
(137)
p ^
i?
= RMLT. ^ ^Z[(AA-'/i4)--8584 3^]
(^^8^
jrp
Z[(AA-'^4)--8s84^
^[(AA-'î4)--8s84:7^]'
(^39) (140)
z=
78
Therefore
THE ELECTROMAGNET.
-kM = J/ =
>i=
2 {d^
and
also
.637
(4
d^ + ttT, + 4) + 7;
(141) (142) (143) (144)
then
2T+di
(136),
d,)
and
Since
^' r = B,-d,
M=.63j{d,+
or
+ (^^^^y,
use
(i4S)
(146)
For calculations of turns, applied to round windings.
etc.,
same formulse
as
Fig. 39-
Radiating surface
Sr=2Z [{d, +
Substituting value of
M from
4)
1.5708
(A -
'î)]-
(147)
(135) in (80),
^^^^^
^^" X[(AA-44)--8584nOr by
substituting value of
J/ from
(141) in (80),
^^=X[.637(^f+4) +
In practice, nearly
all
^]-
^'''^
cores of square or rectangular
WIXDIKG CALCULA TIONS.

cross-section have
79
in Fig.
more or
is
less
rounded edges, as
39
but this need not be considered unless the radius of

sufficient to
the arc at the edge
make a
noticeable in-
crease in the length of the
mean
perimeter, irM.
By
in-
specting Fig. 39
it
will
2 r)
be seen that
-I-
j^_
cr
Clearing,
i.e.,
2 (a^
(r/i
(4
IT
2 r)
+it{T+
2 r)
M= .637
in
M=
2r)
(<ii
.637
-F 4) + ^ -547
.637
(4
2r)
''
+ T+
2 r.
(^5)
any case under
this heading, subtract .547
r from
the value of
J/ in formula
(141) for the true value of
M.
Fig. 40.
To
prove the foregoing, assume

r.
the extreme value for
d^ zr = o, which is Here we have the case of the

Applying formula (150) to
d^, r,
circular cross-section again.

Fig.
40 and assigning the values for d^, follows: Let 4. = Ij ^4 = 'i ''= -Sj ^^^d
and Tzs
.5.
M 1.274 .274
Now
by formula
di=^
d^,
T=
Then
-h -S
(52),
I-5-
M= T-\- d =
1.5 also.
When
formula (141)
when the cross-section is square, becomes J/"= 1.274^ + T, and formula

i.e.,
8o
(150) becomes
THE ELECTROMAGNET.
M~ 1.274^1+ T .574^.
(52),
The
latter
formula, (150), is a general formula, applicable to both square and round cross-sections, formula (141) being correct
correct
when r = o, and formula when 2 r = d^ = d^.
(M =
T+ d), being
Formula (150) may be considered as the modulus for converting a square or rectangular winding into a round
winding with the same number of turns and resistance, but with different values for d, d^, d^, etc., M&nd 7" re-
maining constant.
50.
Windings with
Elliptical Cross-Sections.
Fig. 41.
Since the area here

I
is
simply
D^D,-d^\ IDJD^ d^d\
then
j/r=-(i33)=:^5A::i^^ T = D^Dt d^d^
also
(A J/=^?l^
2
,
^^D^n^^ _ AA
îî
"37 4)
-
(iS6)
(iS7) (158)
A+
Since
'^'s
0S9)
M=
(67),
m
;
either case.
(160)
^
'
From
P=
Radiating surface
= RMLT. RL {D,B, -
d,d,)
(161)
^ = R{D,Dl-d,d,)Sr
(^^3)
= ^L y/^ZA!
j^ from
(164)
Substituting values of
(156) in (80),
IN= ^/J'f,^
51.
,,.
(165)
Windings Whose Cross-Sections have

and Rounded Ends.
Parallel Sides
From
winding
Fig. 42
it is
evident that the cross-section of this
may be
resolved into four parts, consisting of two

areas, as in Fig. 43.
:
rectangles
and two semicircular

of the areas
is
The sum
as follows
-f 2
A = irMT^
T{H -
d^),
(166)
S2
THE ELECTROMAGNET.
and
.-.
MT=-, MT= T[M, + .637 (JI~
(133)
d,)}.
(167)
prrq
Fig. 42.
Dividing by 7]
now
J/i
=A+
(168)
'^s
(169)
Fig. 43.
J/ =^^^^^
.637
(^-4).
(170)
Winding calculations.
From
(170),
Z>5=
(I
M+
Jf +
.274 d^
1.274
.
H.
71) 72)
^=
Since
.274 .274 //j Z>5
1.274
173) 174)
^r = r r(^^^') + .637 (^ P
'^'5)]
175)
176) 177)
R=
p
= ^zr[(^^^) +
.637
iH-
'
4)]
78) 79)
Z=
r=
z'5=
180)
.2744
1.274 75r+
RLT
181)
182) .274
H-.
1.274
183)
Radiating surface
^/=Z[2(^-rt'5)+,rA].
184)
84
fÊ ELECTROMAGNET.
M from
(170) in (80),
K]^Y.,,,iH-,.^
Problems.
33.
How many
is
feet of wire
1.5
may be
if
contained
in
winding volume of
factor
34.
cubic inches
the cross-sectional
1,852 feet.
availis
.000081
367 feet of wire will just fill a bobbin whose able winding volume is .42 cubic inches. What
cross-sectional factor of the wire?
g'^
the
.0001143.
35.
What must be
is
the winding volume of a bobbin to
contain 1,000 feet of No. 30

sectional factor 36.
S.C.C. wire whose cross-
.000197
V=
<>'
.197.
cir-
How
What
many ohms
is
per foot in a copper wire 123

?
cular mils in cross-section

37.
.0842.
the
2,
mean
or average diameter of a windi ?
ing whose ZJ
38.
and </=
M=
1.5.
What would be
the thickness of the winding in
7^= .5. Problem 37? what would be the volume of the 39. In Problem 37, F= 7.069. winding if the length of the winding L = ^l = 2, {a) what is the value of J/? When .6, 40.
T=
{b)
= s..^, {a) What is the value of </? would be obtained in 41. What resistance
{b)
d=
.8.
a winding
space of 3.1416 cubic inches with a wire .0142" diameter insulated with cotton which increases its total diameter
.004"
42.
?
When
{b)
Z)
3 and
M=
2.5, {a)
what
is
p 40-Sthe value
of
r?
Of dl
(a)
T=
.5, (b)
2.
43.
85
is
When
(J))
T=
//?
.5
and d = and
.43, {a)
what
the value of
D
of
Of M->
2.5
(a)
D=
.5,
1.43, (b)
M=
.93.
44.
When J/=
?
T=
(a)
what
is
the value
D
45.
(J>)
Of
(a) Z>
3, (5)
^= 2.
How many ohms

What
is
per inch in a copper wire .024"

<o"
diameter?
46.
.001498.
the resistance of a winding where the true

is
outside diameter of the winding
inches, the outside

is
diameter of the insulating sleeve over the core

the length of the winding
eter insulated with silk
is
.55",
and
2|", the wire .0065" diamits
which increases
diameter .002"
47.
What
If the
silk,
is
the value of
i.9SP in a copper wire .002"
diameter?
48.
K = .6777.
1.5
above wire was insulated with
mil in-
crease of
what would be the value oi
Rt
R=
49.
55.322.
in
What
?
wire should be used to obtain 250
ohms
a bobbin where
insulation
MLT =
.i^j, using single (2-mil) silk
No. 36 wire.
50. What would be the true outside diameter of the above winding ii d = .43 and Z = 2 ? > = .641. 51. In Problem 50, how many turns of wire would
there be
52.
in
iV =
?
4,306.
What would be
What
is
the resistance of the average turn
Problem 50
53.
0581 ohm.
the resistance of a winding containing
.9
?
2,000 turns of No. 24 wire, the average diameter J/"=
p
54.
if i/
= =
12.08.
In the above, what would be the outside diameter

.49
?
Z?
1.31.
S6
55.
THE ELECTROMAGNET.
What would be
the ampere-turns in a winding of
if
iy
Z'
average diameter
wound with No.
22 wire and
placed in a no-volt circuit?

56.
if
IN =
10,427.
In the above, what would be the internal diameter
4?
d
if
1.
57.
What
should be the diameter of a copper wire to

the average
produce 3,000 ampere-turns with 220 volts diameter of the winding is 4"?
58.
.01216.
What
is
the thickness of a winding
3" long wound
with 3,000 turns of No. 24 S.C.C. wire?

59. In Problem 58, if diameter of the winding?
7^=
is
M=
.58.
1.7, {a)
what
is
the outside
{F)
What
the inner diam-
eter of the
winding
will
(a)
D=
2.28, (F)
d=
1.12.
60.
What
be the number
2, if
?
of turns in a
bobbin
where
d=
.43,
Z=
wound
(6)
wire, (a) with .002" silk
to 500 ohms with No. 35 With .004" cotton ?

(b)
(a)
N= 8,570,
.43,
N= 7,490.
500
61.
winding where
d=
Z=
contains
ohms
of No. 36 D.S.C. wire with 4 mil insulation.
What
per cent more turns could be obtained with the same

size of wire
and the same resistance but by using 2-mil

13.6%.
layers in a winding
silk insulation?
62.
How many
where
T=
.5
and
^=. 00833?
63.
=6.
turns per inch where gi
How many
= .00958 m = 104.4.
?
64.
In the above two problems,

in the
how many
turns would
be contained
65.
bobbin
ifZ=2?
iV=i,253.
30%
and
would be obtained with No. 39 = 1.43, d= .68, G.S.S.S.C. wire in a bobbin where
resistance
What
2.5
Z=
238,740
66.
8/
What
is
the intermediate diameter of a winding
consisting of No. 40 S.S.C. copper wire and No. 39 G.S.S.S.C. wire, in a bobbin where d .43,
D=
30%
%,
Z=
wire?
67.
I, it
being necessary to have a resistance of 4,000

still
ohms and
have the
maximum number
of
turns of
jr=.593.
winding consisting of three parallel wires con-
nected in multiple has a resistance of 20 ohms.
What
would be the resistance of the winding

connected in series
68.
?
if
the wires were
180.
What
size of S.C.C. wire
bin where //=.36,
D = .648,
must be used in a boband Z = if, in order to
have two parallel windings of 19 ohms each? No. 32 S.C.C. 69. What weight of No. 24 S.C.C. wire would be required in a winding consisting of four parallel wires whose
joint resistance is
70.
40 ohms?
32-57
lbs.
What would be
What
?
the diameter of a No. 2\\ wire in
the American wire gauge?

71.
is
A=
number
.01897.
the gauge
of a wire .001" in di-
No. 50 (49.88). be the permissible insulation on a wire .0074" diameter in order to wind to a resistance of 100
72.
ameter
What
will
ohms
73.
in a bobbin
where d,=
.55,
D = f^, and Z = i^
i
.00252.
What
will
be the resistance of 2\ pounds of S.S.C.

p
wire .0081" diameter insulated to a diameter of .0083"?
1,878.
74.
What
will
be the
total
weight of 200 ohms of cop?
per wire .0072" diameter insulated with 4-mil cotton
.192
lb.
88
75.
THE ELECTROMAGNET.
What
is
the
ohms per pound

?
of No.
30
30%
B
G.S.
wire insulated with 2^mil silk

76.
late
How many
9>3S' pounds of silk will be required to insuin-
100 pounds of No. 2i\ copper wire to a 3-mil

5.78
crease?
77.
lbs. of silk.
How many
pounds of No. 24
18%
G.S. wire will
25 pounds of cotton insulate
to a s-mil increase?
264.2
78. {a)
lbs.
What
is
is
the
mean diameter
Z'j
of a rectangular
2,
winding where
Z>i
4,
5,
d^=
and
d^
2>^-
r? (a) Jlf = 4.185, (,5) r= i. (p) In the above, how many ohms of No. 24 S.C.C. 79. wire would the winding contain when Z = 1.5 ?
What
the value of
P
80.
5,
72-S7-
What must be
5.5,
the length of a winding where

d^
D^ =
D^ =
81.
(^=3, and
2)-h
in order to obtain a resist?
ance of 100 ohms with No. 26 S.C.C. wire
L=
.724.
In the above,
?
at
10 volts
(B)
what would be the ampere-turns What would be the number of turns ?

{a) {a)
IN
2,010,
ij))
N= 1,824.
Sr
82.
In Problem 80, what would be the radiating sur-
face
I3-9S-
83.
In a bobbin where
//j
2.5,
//j
radius at cor-
ners of core ^", what will be the value of ZJ^ and
D^
in
order to obtain 1,000 ampere-turns with No. 20 wire at

12 volts?
84.
Z?!
3.822, Z'2
=
to
5.322.
be reand the winding on the square core is to contain the same number of turns and resistance as the round winding. What is the value of d^ and D^, L
2,
A round winding
where
d =
Z>
=5
is
wound on to a square
core,
being constant
in
both cases
<^
1.57, D-^
4.57.
85.
id)
89
How many
I,
turns of No. 27 S.C.C. wire will be
contained in a
2.5,
//j
elliptical
1,
winding where D^
1.5,
-Z?4
d^ =
and
L = 2,1
?
{V)
What
will be the
resistance of the winding
86.
face
(a) 53.156. 2,259, {b) p In Problem 85, (a) what wUl be the radiating sur(6) What will be the amp>ere-tums at 50 volts ?
N=
(a)
Sr=
19.43, (3)
d^
IN^=
2,124.
87.
In a winding where
D^
= 2,
= .75, ir= 3,
A=
Sr
L=
2.5,
what
size wire with 4-mil insulation

?
and would be re.02395.
quired to obtain 2,000 ttuns of wire

88.
In the above, what would be the radiating surface ?
26.958.
89.
no
In Problem 87, what would be the ampere-tums at =^ i)4S3volts with Xo. 30 wire?
IN
go
THE ELECTROMAGNET.
CHAPTER
III.
HEATING OF MAGNET
52.
Effect of Heating.
COILS.
When
a current of electricity flows through the windis
ing of an electromagnet, heat
produced due to the
current acting against the resistance of the winding,
and
may
properly be called the heat produced by electrical
friction.
The amount
flowing through
of heat
produced
is
proportional to the
of the current
resistance of the winding

it,
and the square
or, in
other words, to the watts lost in
the winding.
The heat from
the outside layer
is
radiated rapidly,
but the heat from the inner layers has to pass through to
the outer layer, core, or washers before
sipated, thus heating the entire winding.
it
can be
dis-
The
at
coil,
then, as a whole, radiates the heat gradually

is
first,
but faster and faster as the heat

finally the
conducted
is
through the outside layer, until

as fast as generated,
fore,
it
heat
radiated
and equilibrium established. Thererequires considerably more time for a " thick "
to
winding
one,
reach this point than
it
does for a thinner
and a thick winding will therefore get hotter inside than a thin one for the same reasons, when under similar
electrical conditions.
From
this
it
is
evident that the heating of a winding

its
and the time required to reach
maximum
is
propor-
HEATING OF MAGNET
COILS.
tional to the thickness of the winding, the square of the
current,
and the
resistance,
and
is
inversely proportional
to the radiating surface.
In practice the radiation from the ends of the wind

ing and
core
to
is
not considered, but the total radiation
assumed
be from the top or outer layer of the winding.
In the average winding about 65
of the heat is radiated
from the outer
add the other 35%, and thus shorten the calculation. In practice this method has been found to give as satisfactory results as any, and hence is commonly used.
layer, so
it is
safe to
The
certain
heat generated in the coil has the property of
increasing the electrical resistance of the winding in a

ratio
is
for
each degree of
rise
in
temperature.
This ratio
called the Temperature Coefficient,
and varies
for different metals.
Of course any change

influences.
of
temperature will vary the
resistance of the coil, whether due to internal or external
The temperature
i.e.,
coefficient for
copper wire
is
.0022,
the resistance of a coil of copper wire will vary
.22%
for each degree F. of change in temperature.

jDj
Therefore,
(186)
(i
4-.oo2 2^)p,
where
f=
rise in
temperature in degrees F.
coil of
Example.
copper wire has a resistance of
100 ohms at 75 F. {a) 100 F.? {b) At 32 F. ? Solution.

{a)
What
will
be
its
resistance at
100
75
=
'^
25 rise ;=
t.
(i+(.oo22 X25)) X 100=1.055
= iS-S ohms. Am.
92
(P)
THE ELECTROMAGNET.
75
32
43 drop.
Pi
I
+ .0022 t
100
(.0022
;
(187)
"
~ + =
\
43)
Q1.36 ohms.
Ans.
1.0946
so
The radiation of heat from a winding depends upon many things that in practice it is assumed that the
the winding
is
average rise in temperature in the winding will be 100 F.
when
radiating a certain
number
of watts
per square inch continuously, the rate of radiation de-
pending on the thickness of the winding.

Thus, when an ordinary telephone ringer magnet
radiating
is
approximately
.9
watts per square inch con-
tinuously, the rise in temperature will
be approximately
while a winding 4J" in diameter, 7" long, and lyV' thick will rise in temperature 100 F. when the winding is radiating .33 watts per square inch con-
100
F.,
tinuously.
The
rise in
temperature in a winding
is
directly pro-
portional to the rate of continuous radiation.
Thus, a
winding that
ing
I
.5
temperature 100 F. when radiatwatts continuously will rise 200 F. when radiating
will rise in
watt per square inch.
The permissible rise in temperature depends entirely on the temperature of the place where the winding is to be used. In any case, the temperature of the surrounding air must be deducted from the limiting temperature. When several coils of the same dimensions, but for use with different voltages, are to be made, it is best
HEATING OF MAGNET
to test
COILS.
93
one
the
coil
and ascertain the
rise in
temperature and
the watts per square inch for different periods of time.
From
easily
data thus obtained the proper wire

for
may be
different
calculated
the
other windings
at
voltages, to obtain the
maximum
ampere-turns without
overheating.
100
30
so
94 To make The source

Fig. 45
THE ELECTROMAGNET.
the
of
test,
use a mil-ammeter and voltmeter.
current must be of constant voltage to
give good results.
shows the connections for the

(3)
test.
By Ohm's law
E
rise in
and from formula (186) transposed, the

perature
.002.2
tem-
voir fêreK.
AHL-AMMBreti.
n w| Q
HEATING OF MAGNET
which
for a
COILS.
95
round magnet
is
r^==-^;^-
EI
nBL
(190)
The
is
resistance of a winding to be used
on any voltage
then
Pi-
? W,itDL
(192)
for a round winding.
If the current is to
be kept constant
in
in the
winding the
voltage will vary, but the rise
temperature for any

of the test as ex-
period of time
plained above.
may be found by means

between the
The
also,
ratio of heating
inside, middle,
outside layers
may be determined by
the above
and method
by connecting wires to both ends of each layer to be
tested, as in Fig. 46.
96
53.
THE ELECTROMAGNET.
Relation between Magnetomotive Force and
Heating.
Where
and
constant voltage
is
used, the ampere-turns vary
directly as the watts, since voltage
and turns are constant,
Ampere-turns
= amperes watts = amperes

between
X X
turns,
volts,
therefore, the
ratio
magnetomotive force and

is
heating due to the current in the winding
constant,
since the current varies directly with the resistance.
Hence the watts and ampere-turns will fall off in the same ratio until the heat is radiated as fast as generated, when they will become constant. For this reason magnets which are to work continuously should be calculated to do the work at the limiting temperature.
First determine,
from the dimensions of the winding,

its
how many
watts the coil will radiate from
surface for
the required rise in temperature.
Then
the resistance at
the limiting temperature will be
^
and the resistance
at the air
temperature
Pi
^
I
+
Pi
.0022
(194)
=
for 100 rise.
1.22
Thus the cold
resistance for 150 rise

Pi
I -f-
would be,
Pi
i
.0022
1.33
HEATING OP MAGNET
With constant
COILS.
97
current, ampere-turns
and voltage and

Since the
re-
watts vary directly -with the
resistance.
sistance iricreases as the heat increases, thus increasing
the generation of heat,
it
is
also very important that the
ampere-turns should be calculated at the limiting temperature.
From
ture,
permissible
watts
per
square inch at limiting
temperature calculate the resistance at limiting tempera-
then the cold resistance

.
_
I
Pi
-I-.0022 t
(194)
A winding
number
of fixed dimensions will contain the
same
of ampere-turns at a given rise in temperature,
no matter what the resistance or voltage may be. The above is of course on the assumption that the relation between diameter of wire and thickness of insulation
is
constant.
In practice the ratio between diameter and thickness

of insulation
is
not exactly constant, so the above rule
is
only approximate.
Therefore,
if
a certain coil will contain a certain num-
ber of ampere-turns with a certain wire at a certain

voltage, approximately the.
will
same number
of ampere-turns
if
be obtained with any other wire,
the voltage
varies in the
of the wire.
same
ratio as the square of the diameter
From
which
is
(8.)
^^
the diameter of copper wire for the given num-
ber of ampere-turns.
98
THE ELECTROMAGNET.
This, however, does not take into consideration the
heating effect of the
coil.
In order to control the heating,
the resistance must have a certain value, and the resist-
ance changes with the thickness of the insulation on the

wire.
Therefore, the volume of the bobbin, the radiating
surface,
and the thickness
of
the
insulation
must be
known.
Example.
greatest
Given bobbin,
of
find the exact diameter of
wire to use with given
insulation
that
will
give the
number
that
ampere-turns
without
overheating
the coil.
Assume
inch for
as follows
:
coil
will radiate
F.,
watt per square
a rise of
ioo
and the dimensions are
Core I", d then the area
= .43,
is
>
irDL
of
= L = 2, and the voltage = 50 = 6.2832 square inches.

1,
The
resistance
the winding at the limiting tem-
perature will be
Pi
^
095)
inch.
?Q7.Q ^^' ^ ohms,
'
where
'
Wi,= watts per square KO^ 2. SCO
"=
Pi
6.2832
4!z^-Z.
T-^
=
=
-TT
6.2832
326.5
=
^
'"
and
ohms
air,
at the temperature of the surrounding
Now
MZT
=
'"
From
(68), if
-^ =
8,022.
HEATING OF MAGNET
Assume
Since
that
i?
/
COILS.
99
.002.
(6s),
and
c
K = ^,
(63),
i?
(197)
(See (72).)
AV'
Since
Ri (46),
andA^=y/^.
A'
g=^+
A/
\J^
completing the square,
^ + ^' + 'j-V^S + 7
and
(198)
..
]*=[V^ 802.2 2 -oo^ s = V.OOOOS8I9 + .00001 A = .00826 .001 = .00726. Ans.

/
00000271
.00004
002
To
shorten the calculation, substitute
MLT for
i
.ff,
then
A= A=
Now,
since
p^
cMLT
p
i^.
V
V
4J
(199)
40
(192)
(200)
2
W.'kDL
P
for
fr,7rZ)Z(l+-oo22/)
(201)
any
rise in temperature.
too
THE ELECTROMAGNET.
written
-\
The complete formula may be
=[V^
4^
4J
2
'(202)
which gives the exact diameter of bare copper wire, which will give the maximmn number of ampere-turns
within the limiting heating conditions.
Now,
to prove the last formula, take the

;
same example
as above
then
.00000271
3.1416 X
X4X
.815
=[v/^^^ 4 X 2,500 + .00001 .001 ]'
=[v^
=
0000^^86 ''
1,000
.00001
V.000058I9 + .00001 .001. /.A = .00826 .001 = .00726, Ans., which is the same result as obtained before. Substituting value of ^7" from (134) in (199),
A=
r,
Az[(AA-'^4)--8584y'] ^ qi_
MTlxom.
_l_
i_
(^^3)
for square or rectangular windings.
(152) in (199), ^^^^^
'cL{D,B, - d,d^
= [V^
qi_ I
4J
2
4P
for elliptical windings.
MT from (175) in (199);

.637(^-</a)'
I
A=
/.Zr[(^^j +
(205)
for windings with parallel sides
and rounded ends.
HEATING OF MAGNET
To
find the
COILS.
lOI
number
of ampere-turns, use formula (80),
or
IN^-^.
Therefore, the
(206)
of ampere-turns that
maximum number
in
may be
obtained continuously,
any winding volume
with any voltage, with given insulation and at the limiting

temperature,
is
IN=
Here
^SV
(i
~ + .0022 f)cW, SrMLT
i^ "
cM
(207)
= radiating
[(^1
surface,
and
is
equal to irDL for
round windings,
,
+ 4) +
1.5708
(A -
^0] (147)
for square or rectangular windings,
.L^Wl
for elliptical windings,
(.64)
and
(184)
Z[2(^-4) + xA]
for windings with parallel sides
and rounded ends.

resist-
The
general
relations
between watts, voltage,
and radiating surface are expressed by the following formulae, the values being taken at the normal
ance, current,
temperature, or the temperature of the surrounding atmosphere.
Let
W=
W, = watts per square

total watts
;
inch,
I02
THE ELECTROMAGNET.
sn
(S)=
EI
(6),
'
HEATIKG OF ^rAGXET
COILS.
IO3
Since the magnet requires less current to sustain the

it from a distance, the winding is designed to carry the sustaining current continuously, the
weight than to attract
heating due to the attracting current not being considered

unless the attracting current
is
very
is
much
greater than the
sustaining current, as the former
This principle
stats.
is
employed
in
on only momentarily. self-starting motor rheo-
If
E represents
the attracting voltage, and
E^ the
sus-
taining voltage,
F W = SrTLE
TlV>r-
"
-1-
(I
.0022
/)
('222^ ^ '
-^= JKSrTLE {i+. 0022 n

Eî^
If the attracting voltage
*^"^^
and the sustaining voltage are
equal,
orxZ(i
-{-.0022/
54.
If a
Advantage of Thin Insulating Material.
bobbin was wound with bare wire with no insulation,
the ampere-turns would reach a
maximum when
the watts
were at their maximum.

turns are
multiplied
In
this, of course,
the ampere-
meant to be the number of turns in the coil by the current which would pass through the bare wire when suspended in air. In practice the wire is insulated, which increases the total volume considerably, and therefore, in order to obtain
I04
the
THE ELECTROMAGNET.
same
resistance with the insulated wire, in a given
bobbin, as would be obtained with bare wire, a shorter

length of insulated wire with a smaller cross-section of
copper must be used.
When
this is done,
is
however, the
resistance of the average turn
increased,
the ampere-turns will be reduced and will
and therefore not be at their
maximum when
wire.
the watts are
maximum,
as with the bare
The ampere-turns will therefore reach a maximum when the voltage divided by the resistance of the average turn produces a maximum.
It is therefore
obvious that the efficiency of a winding
increases as the thickness of the insulation decreases.
For
this reason, silk-covered wire is

its
much more
cost
is
efficient
than cotton-covered wire, although
greater.
Whatever
is
saved in
is
first
cost of winding with any
fixed insulation,
at exactly the
paid for in the cost of operating and

rate as the saving in first cost
if
same
is
con-
stant voltage
used.
55.
Work
at
End
is
of Circuit.
When an
electromagnet
to
of a line of considerable resistance, the
be connected at the end winding of the
magnet should have slightly less resistance than the line, in order to do the most work, providing, of course, that the winding volume is great enough to prevent the winding from becoming overheated.
The reason
resistance,
it
for this
is,
that
if
the line has the greater

coil,
will
absorb more voltage than the
with
the same current, thus absorbing more watts.
Again,
if
the coil contains
more resistance than the
line,
HEAThVG OF MAGXET
the
resistance
of
COILS.
will
I05
the
average turn
have been so
increased by the use of finer wire that the ampere-turns

will
be greatly decreased, the dimensions of the winding

voltage across the terminals of the electromagnet
is
being the same in both cases.
The
winding
E = -^^
P
+ Pi
(226)
coil,
where
= voltage
_E^
across
= voltage of line, p = resistance of coil, Pi = resistance of line.

that an electromagnet
circuit,
is
As an example, assume
to
operate at the end of a 220-volt
the resistance of
the line being 250 ohms, and the dimensions of the magnet winding as follows
:
MZT=
.2.
M=i.
and
Watts per
single silk-
square inch permissible.
This
is
shown calculated
for both bare
insulated wire.
The
difference in ampere-turns between
the bare and insulated wire will be noted, also the fact
that the ampere-turns are at a
maximum
with the watts
for bare wire, but not for insulated wire.
Bare Wire.
No.
32
.0429
.0682
33
.0541
34
for
679
1,080
1,715
- 135.8 - 77. s 1,805 216 102 1,880 343 127 1,860
44.2 48.1
47
Here the maximum falls between No. 33 and No. 34 both ampere-turns and watts.
Io6
THE ELECTROMAGNET.
Single Silk-Covered Wire.
i
.0025.
W;H
JiT
33
54i
34
35
.0682
.086
- J^ - E - IN -W p 589-117-8- 70.5-1,320 42 880 176 91 I1330 47 1,307 261.4112 1,305-48

are
Here the ampere-tums

while the watts are
maximum
with No. 34 wire,
with No. 35 wire. Therefore, calculate the size of wire to use assuming
maximum
the resistance of the coil to be equal to the resistance of the line and battery, or source of energy, and then try the
next larger size of wire, selecting the wire which gives the
greatest
number
of ampere-tums.
Now,
and
(See 62.)
E=
p
i^, +
ft
(226)
cMLT = _^___.
(327)
Substituting value of p from (227) in (236),
^ = MA-(A+\7N
1^
(^^)
cMLT ) ^
maximum when
The ampere-turns mum.
are
cM
z-^
is
maxi-
^^=
is
/ fx / ft (A
^ +
,-vx iy \
TAir cM
(229)
\
If the
TL
j^
t^
magnet
to carry the current continuously, the
bobbin must be made large enough to radiate the heat.
HEATING OF MAGNET
COILS.
I07
it
magnet should always be so designed that
will
stand the total voltage of the line without overheating.

It is
a mistake to place a resistance in series with a

lost in the
magnet, for the power
dead resistance
is
nearly
in direct proportion to the relative resistances of the
net and the dead-resistance
coil,
same total energy the magnet will when designed to have the full voltage without
ing.
magand therefore for the be much weaker than

overheat-
Consequently
this practice is
wrong.
it
Furthermore,
is
the cost of operating varies as /^, so
seen that the
higher the resistance, the more economical will be the

operating of the magnet.
Problems.
90.
(a)
The resistance of a winding is 87.5 ohms at 70 F. What will be its resistance at 1 60 F. ? {b) At - 1 0 F.
{a) pi
104.815, {b) p
74.4.
91. What would be the temperature coefficient of a wire which changed from 320 ohms at 130 F. to 310 ohms
at 70 F. 92.
is
.000538.
resistance of a copper wire winding at 80 F.
(a)
The
25 ohms, At 100 F. ?
What
will
be
its
resistance at 0 F.
(6)
{d) p
21.26, (b) pi
26.1.
^=
E
93.
What would be
no, /"=
the watts per square inch where
W,= .$. .3, and 6'r= 66? What would be the permissible resistance in a 94. windingwhere D^ = 4, Z>2 = 4.5, d^ = 2.5, (^ = 3, Z = 2, = 500, and fF^ = .6 ? p = 13,260. In the above, (a) what would be the proper size of 95.
wire to use
?
(6)
What would be
(a)
the ampere-turns
No. 35 S.S.C, (-5)7^^=1365.
I08
96.
THE ELECTROMAGNET.
What must be
.5
the resistance of a winding at 68 F.,
to radiate
volts,
watt per square inch at 150 F., with 500
assuming the radiating surface to be 93 square
inches?
97.
What would be
4;5S7P the theoretically exact diameter of
wire to use with 4-mil insulation at 68 F., in order that

the average temperature of the coil will not rise above
150 F. during continuous service on a iio-volt circuit,
assuming radiating surface to be 40 square inches, W,=.'j,

a.nd
MZT= ^?
What would be
winding where
4.5, voltage
A
the
^1
.01036.
^8.
F. in a
maximum
ampere-turns at 150
=
.5,
and
to
Z=
no,
2, //j 4, >i 3.5, D^ 5.5, insulation 4-mil cotton, and the
watts per square inch
assuming normal temperature
be 68 F?
99.
/iV^= 2,410.
watts per square inch at the limiting
How many
temperature (a) where
/=
.5,
=
=
100,
(c)
Where
=
=
110, p
=
JV,
220,
Sr
60}
.Sy=5o? Where =
(<r)
(H)
50,
/=
.5, ^/-
40
W
100. 100,
.5, (S)
W.
.917,
W,=
.5
.625.
What would be
the safe voltage, allowing
watt
per square inch at limiting temperature, (a) where p
Sr
19
(ff)
Where
1=
(a)
.2,
Sr
15
E = 30.82,
(b)
(6)
E=
37.5.
loi.
What would be
the safe current at 68
F., allow-
ing
(a)
.8
watt per square inch at the limiting temperature,
where p 90?
102.
200, 6'r
25
Where
(a)
7=
.316, (^)
E = 500, Sr 7= .144.
attract
its
An
electromagnet
is
designed
to
armature and load on a 220-volt

maintain the load at 140 volts.
circuit,
and then to
The
required ampere-turns
HEATING OF MAGNET
are found to be 5,100.
COILS.
I09
are
as
follows
space factor g^
The dimensions of the winding d = i^, D = 2^, L = 4^, and the What will be the watts per .000323.
?
square inch at the limiting temperature

103.
W,=
.872.
In Problem 102,
2 2 o volts
how many ampere-turns would

if
be obtained at
1
the watts per square inch at

.9,
40 volts were increased to

104.
and the space
factor
^=
.00041?
An
electromagnet with
/^= 4,147. dimensions ^= .55, Z* =

which has
1.03,
Z=
3, is to
be placed
in a 24-volt circuit
a line resistance of 10 ohms.

to give the
What
wire should be used
maximum
?
ampere-turns, assuming the insula-
tion to
be 4-mil cotton
No. 24.
no
THE ELECTROMAGNET.
CHAPTER
IV.

56.
Forms
of Electromagnets.
is
The
as
it
best form of magnetic circuit
has the
minimum
if
magnetic leakage
the wire
the ring, Fig. 47, but it is not the
best form of electromagnet on account of the space lost

in the winding, for
is
wound evenly
ring.
in layers
on the inside of the
ring, there will
be considerable space
between the turns on the outside of the
Fig- 47-
Again, if the turns are wound evenly and close together on the outside of the ring, the turns will be crowded or bunched on the inner side, thus increasing the length of
the
mean turn, and decreasing the ampere-turns. The Bar Electromagnet, Fig. 48, has the same
permanent bar magnet
is
gen-
eral field as the
in Fig. i, but the

coil of wire,
core
of soft iron
and
is
surrounded by a
Fig. 49.

The bar electromagnet
the armature
is is
II
not
efficient,
however, unless
bent into the form of
so as to complete
the magnetic circuit.
Even
theti there will
be much leak-
Fig. 48.
age between the parallel legs of the armature, unless they

are very short.
If
only one pole attracts the armature,
i.e., is
near and
the other pole distant, the magnet will be very
weak on
Fig. 49-
account of the high reluctance of the air through which

the magnetic flux has to pass.
The Horseshoe
efficient
Electromagnet,
Fig.
50,
is
the
most
type for a short range of action, but as this form
112
is
THE ELECTROMAGNET.
make and
therefore expensive,
is
rather inconvenient to
the practical horseshoe electromagnet

pieces besides the armature.
made
of
three
Fig.
513.
In this form, Fig. 51, the wire

iron," as
is
wound
directly
on to
the cores, and they are then fastened to the yoke, or " back
it is
often called.
Fig. 51.
While
this
form
is
very
efficient,
there
is
a loss due to
the joints between the cores and yoke.

The
Iron-clad Electromagnet., Fig. 52,
is
113
really a
form
is
of horseshoe electromagnet with
one of
its
poles in the
form of a
shell
surrounding the other pole.
This type
usually round,
and
consists of a solid piece of iron or steel
with a deep groove turned inside to receive the winding.
The
external
iron-clad
electromagnet
has
but
the
is
it
advantage of
free
being mechanically protected, and

inductive
influences,
little
also
from
has
the
decided
disadvantage of having
or no ventilation, depending
Fig. 52.
upon the heat being conducted through the shell before it can be radiated by the iron.
great deal of the heat
is
air inside the
Of course a
conducted through the core to the

is
outside, but entangled air
a very poor conductor of heat
for the outside of the winding.
This may be largely overof the shell so as to allow
come by
cutting
away a portion
a circulation of air next to the winding, but this would

increase the reluctance of the magnetic circuit
by decreas-
ing the amount of iron.
The magnetic leakage

is
is
also great
when
the armature
a comparatively short distance from the poles.
This type
is
used extensively in telephone switchboard

is
apparatus where the current
of short duration
and the
114
range of action
THE ELECTROMAGNET.
is
also very short.

it is
Its principal feature
in this case is that
not affected by external magnetic
influences.
Another form
53,
is
of
electromagnet
called the Circular Electromagnet, Fig.
used in magnetic clutches.
This consists simply of a ring of

iron or soft steel with a groove cut in
its
periphery to receive the winding.
When
seeking.
excited,
north-seeking
one of its faces is and the other south-
the ones described are

Fig. 53.
There are many other types, but more commonly
used.
Direction of Flux in Core.
57.
In Fig. 10 was shown the relation of direction of current in a wire to direction of lines of force about a wire.
Fig. 54-
Fig. 55.
is
In the electromagnet where the wire
coiled about
the core, the same relation holds as in Fig. 54 and Fig.

55-
There are a great many

this law,
rules to aid in
remembering
but probably the simplest one is the analogue of the corkscrew or ordinary right-hand screw.
IlJ
As a corkscrew is turned to the right it enters the cork. Simply consider the cork as the north-seeking pole, and
the
direction
of
ours/oe.
rotation as the direction of the current.
Intwo-polemagnets of the
type
shown
it is
in Fig. 51,
convenient
the windFig. 56.
to
make
ouTsioe.
ings of both bob-
bins in the
same
direction.
When
they are fastened to

ter-
the yoke and
connected, however, the two inside
minals must be connected, as otherwise the two windings
would be acting
in opposition to
one another.
Fig.
56
shows
this
principle.
In electromagnets and solenoids having but one bobbin, the direction of the winding
is
immaterial.
When
connected
Fig. 57-
the coils are

in
multiple
(to
reduce the time con-
stant,
or for use on less voltage than for which they were

series), the inside terminal of
designed to work in
other for the
each
winding must be connected to the outside terminal of the
same reasons as explained above.

of connecting the
Fig. 57
shows the method
two windings of an
electromagnet in multiple.
1 1
THE ELECTROMA GNE T.

58.
Action of an Electromagnet.
is
When
is called
the armature
it is
separated from the poles of an
electromagnet,
attracted
by the
poles,
and the magnet

armature
is
an Attractive Electromagnet.
If the
at a considerable distance of the air
from the
poles, the reluctance
gap between the armature and the poles may

magnetic flux will pass
equal or exceed the reluctance between the two poles, and

if
equal, only half of the total
through the armature.
As
the armature approaches the poles, the reluctance of
the air gap becomes less

stronger,
and the
attraction
becomes
and there
is
also less leakage
between the poles,
until finally,
is
when
the armature touches the poles, there

reluctance between the poles and
comparatively
little
the armature, and practically no

poles.
leakage between the
The magnet
is
then said to be a Portative Elec-
tromagnet.
The
since
reluctance in air between two surfaces
is
i?
=A -5
>
/i
=
is
in air.
Therefore, as
increases,
de-
creases, / remaining constant

air
If the reluctance of the
gap
very great as compared with the reluctance of
the rest of the magnetic circuit,

reluctance,
we may
neglect the latter
by assuming that the leakage ratio is constant, concentrate our attention on the effect in the air gap alone. Since the flux density is constant for any
also,
and
cross-section of core, with constant M.M.F., increasing the
area of the poles of a magnet also increases the pull of

the magnet, since the pull
is
proportional to the above

is
WA.
nearly correct
While not
strictly correct,
for comparatively long air gaps.

The
field of
liy
is
best form of pole for producing strong fields

aperture,
conical pole piece of 120
which
will
give a
approximately 250,000 lines per square inch over an extent of several square millimeters. The radius of
the pole face should be about \",

in the
making the pole nearly
form of a parabola.
Calculation of Traction.
59..
When
it
is
desired to construct an electromagnet, the

is
principle data given
the Traction, or pull.

is
The formula
for the pull
P=
72,134,000
(230)
The table on p. 150 is calculated from this formula upon the assumption that there is a uniform distribution of lines of force over the area considered, and that there is no magnetic leakage.
practical working densities for different grades of and steel, providing the permeability does not fall below 200300, are approximately as follows
The
iron
Wrought
Cast
steel
iron
Mitis iron
...,,... ......
50,000
85,000
80,000 50,000
Ordinary cast iron
Therefore, to find the size of core to use, select from
the table the pull in pounds per square inch opposite the
working density, and dividing the required pull by the pounds per square inch gives the area of the pole. Likewise, to find the pull
when
the value of
B and
the area of
Il8
the pole are
THE ELECTROMAGNET.
known, find the pull in pounds per square inch from the table and divide by the area of the core.
Example.
tain
steel.
A magnet is to be designed that will susa weight of lo pounds, the core material being cast
What
Solution.
steel is
1 1 2.3,
The
A=
should be the area of the core at the poles

practical
working density for cast
B=
90,000.
In
table,
when
B = 90,000, P =
therefore the area of the core
=
112.3
.o8q square inches,
or a core .336" in diameter.
In order to obtain fair results, the armature and pole must be accurately faced, as a non-uniform distribution of the lines and increased reluctance may diminish the traction.
On
the other hand, a diminished area of contact will
increase the traction providing the total flux passes through

the joint.
two-pole magnet will sustain twice the weight of a

single-pole magnet, for
is
so-called
the
same
intensity of
induction, as the magnetic flux

is
utilized twice
and there
twice the pole area.
The
will
larger the cores of an electromagnet the greater
be the strength, providing the cross-sections of the
armature and yoke vary in the same ratio as the crosssection of the cores, the outside diameter of the winding,
the resistance,
and the length
of
the
magnetic circuit
will fall
remaining constant; for while the ampere-turns

off as
the core increases in diameter, the cross-section of
the core
and
&
increase faster than the ampere-turns
decrease.

60.
Solenoids.
II9
The
solenoid or coil and plunger magnet consists of an
electromagnet with the core free to

as the armature.
move and which

;
acts
This
is
based upon the tendency of
the lines of force to take the shortest path
and as the
Fig.
58.
iron offers less reluctance than the

increases,
air,
the force greatly
and the core

is
of the winding.
drawn or pulled into the center The solenoid is commonly used where a
is
long range of action
required.
is
The
simple solenoid, Fig. 58,

it is
the least efficient form,
although
commonly
used.
Its efficiency
may be
rated
with that of the bar
electromagnet
only one pole
to
attract
when
is
used
the arma-
ture, as
the only re-
turn circuit for the

lines of
force
air.
is
the
surrounding
form, but
The magnetic
is
field of
a solenoid
is
not perfectly uni-
nearly so at the center, decreasing towards

it is
the ends, where
weakest on account of the demagne-
tizing effect of the poles,
which react as in Fig. 59.
120
THE ELECTROMAGNET.
feathered
The
arrows
represent the direction of the
lines of force
produced by the solenoid, and the plain arrows the direction of the lines of force due to the reaction
of the poles.
When
an iron core
is is
placed inside of a solenoid the.

greatly increased, but
it
demagnetizing action
decreases
as the ratio of length to diameter increases.
6i.
Action of Solenoids.
In the case of the simple solenoid, the pole induced at the lower end of the plunger as it approaches and
enters the solenoid
is
attracted
and drawn farther
in,
thus
decreasing the reluctance of the magnetic circuit and in-
Fig. 60.
creasing the flux

attraction being
maximum when
and consequently the attraction; the the end of the plunger

is
reaches the center of the solenoid.
The most efficient and generally useful form dad Solenoid, or " Plunger Electromagnet,^'' Fig.
of sufficient cross-section
low.
to
the Iron-
60.
In this form the magnetic return circuit consists of iron
make
the reluctance
very

The frame
gap
is
is
121
steel
air
usually a wrought-iron forging or
casting, although cast iron serves very well

great, as the reluctance of the air
where the
is
is
gap
so great
that the reluctance of the cast-iron frame
very low by
comparison.
The
spool
or bobbin
is
usually
made
of
brass, the tube of the spool extending clear through the
upper end of the iron frame

too thin to introduce
that point.
(
this
keeps the core or

is
still
plunger from sticking to the iron, while the tube
much
reluctance into the circuit at
<
FiE. 6l.
In some solenoids of this type the plunger passes clear
through the frame at both ends, as in Fig. 6i
but the
is
form with the iron stop
is
the strongest, as there

is
no
extra air gap, and the attraction
between the iron stop
and the plunger, instead of between the walls of the hole

through the frame and the plunger.
As a portative magnet, the one with the stop will hold many times the weight held by the other form, and as a tractive magnet it is also much stronger, especially if
the stop and plunger are V-shaped, as in Fig. 62.*
*
W.
E. Goldsborough, Electrical World, Vol.
XXXVI.,
July 28, 1900.
122
THE ELECTROMAGNET.
Fig.* 63 shows the best condition for the V-shaped gap.
In the case of the iron-clad solenoid, the action
is
combination of the simple solenoid and electromagnet, the
Fig. 62.
attraction reaching a
pletely closes the
maximum when
circuit.
the plunger com-
magnetic
Fig. 63.
62.
Polarized
Magnets.
is
This form of electromagnet
the
its
same as the comarmature

is
mon form
with the exception that
a per-
manent magnet, or else the entire electromagnet is influenced by a permanent magnet. Fig. 64 and Fig. 65 show two forms where the armatures are permanent magnets made from hardened steel.
W.
E. Goldsborough,
Electrical
World, Vol.
XXXVI.,
July 28, 1900.

In Fig. 64 the armature
in
is
123
pivoted at one end, and
Fig. 65 the armature
is
pivoted in the center.

is
One
great advantage of this form of electromagnet
that the
Fig. 64.
direction of the
movement
of
the armature corresponds
to the direction of the current in the winding.
Thus,
if
the current flows through the winding in the direction of the arrows in Fig. 65 the armature will be attracted to
the
left.
Fig. 65.
If
now
the current
be passed through the winding

results will
in
the opposite direction, the armature will be attracted to the right.
The same
be obtained with the
magnet
in Fig. 64.
124
^-^^
ELECTROMAGNET.
one
The
principle lies in the fact that like poles repel
another, while unlike poles are attracted, therefore the

is simultaneously attracted by one pole and by the other. The methods of polarizing the entire magnet, including the armature, which is soft iron in this case, are illustrated in Fig. 66 and Fig. 67, the former being used principally on telegraph instruments, and the latter on telephone signal bells. Both respond to alternating cur-
armature
repelled
Fig. 66.
rents of low frequency, the synchronous action depending
upon the
Still
inertia of the armature.
another form has the winding upon the armature

oscillates
which
between permanent magnets.

are
Polarized
magnets
very sensitive and
may be may be
worked with great
rapidity.
Where
just
direct currents are used, the armature
balanced by means of a spring, so that the least

change
12$
in the strength of the field will disturb the bal-
ance and move the armature.
The reason why

is
polarized magnets are so
much more
sensitive than the non-polarized
magnets
is
because there
a greater change in the flux density in the former than
in the latter.
Consider
receiver,
a polarized electromagnet
flux density to
in
a telephone
lines per
assuming the
be 5,000
Fig. 67.
square inch due to the permanent magnet alone, and that

the current in the winding increased
pull
it
to 5,005.
The
on the diaphragm before the current flowed would be proportional to B^ = 5,000^= 25,000,000, and after the
current flowed,
pull
B^
5,005^ =25,050,025, an increase in
25,000,000 =50,025 due to the current of 5,005 5,000 = 5 lines per square inch; whereas, if the magnet had not been polarized the increase in pull would only
proportional to
25,050,025
for a
change
in flux density
126
have been
ity,
i.e.,
THE ELECTROMAGNET.
5^
o=
25 for the same change in flux densTherefore, the polar-
5 lines per square inch.
ized magnet in this case would be -^^/-^
2,001 times
stronger than the electromagnet alone.
Since the permeability of the permanent magnet
is
very
much
less
than that of a soft iron core, there
is
not so
great a change in the flux in the steel as in the iron, but

nevertheless the polarized
sitive
magnet is many times more sen-
than one that
is
not polarized.
Problems.
105. How many pounds will a two-pole magnet lift whose pole areas are 1.5 square inches each, and the
total useful flux is
140,000 lines
P= 362.3.
be the same
106.
What would be
?
the effect of increasing the area

total useful flux to
of each pole,
assuming the
continually
107.
it
If the magnet in Problem 105 was polarized, i.e., had a continuous flux through its magnetic circuit, and the total useful flux due to the polarization was 100,000 lines, what would be the per cent increase in traction if the total useful flux was increased to 160,000 lines by means of the magnetizing coil ? 156%.
if
12/
CHAPTER
V.
63.
Induction.
When
a current of electricity flows through a wire or a
coil of wire, lines of
magnetic induction are established

the case of the coil of wire, they pass
coil.
about the wire,
or, in
through the center and about the
Now,
an angle
if
a wire be passed through a magnetic
field at
to the lines of force, a current of electricity will
be generated in the wire.
This
is
the principle of
all
dynamo
electric
machines.
if it
current will also be generin the
ated in the wire,
be placed
magnetic
field,
and
the field then disturbed or entirely destroyed, or suddenly
increased from zero to
maximum.
field
This action of a magnetic

Induction,
and
is
utilized in induction
upon a wire is coils and

is
called
trans-
formers.
The
induction between separate coils

64.
Self-induction.
called
Mutual
Induction.
The principle
as induction
of Self-induction or Inductance is the
same
between a wire conducting a current and a wire placed near it, with the exception that the wire conducting the current
is
acted upon by the field produced

in the wiie
by that
changes
current.
Thus, whenever the current

magnetic
field also
in intensity, the
changes, and
thus in turn generates another current in the wire.
128
THE ELECTROMAGNET.
induced E.M.F,
If the current in the wire increases, the
will
be
in the opposite direction,
thus tending to retard or
hold back the original current.
This
efEect only
lasts,
however, while the current
is is
changing
in intensity, so that as
soon as the current
constant the self-induction ceases, and the current reaches

its
maximum.
Thus, when the current
is
switched on to the wire,
it is
retarded by an opposing E.M.F. produced in the wire,
and therefore does not reach a maximum

Again, when the current
is
instantly.
in-
suddenly stopped, the

i.e.,
duced E.M.F. acts
in the opposite direction,
in the
same
direction as the current.
The
effect is greatly in-
creased
when
there
is
iron in the field,
and an enormous
amount
of current flows through the wire producing a
large spark at the point of rupture.
This principle
is
taken advantage of in electric gasis
lighting apparatus
where the self-induction
purposely
made high in order to produce a strong, hot spark. The E.M.F. induced in a coil of wire is one volt
each turn of wire when the
crease at the rate of 100,000,000 per second.
for
lines of force increase or de-
The henry
is
is
the unit of induction or self-induction, and
represented by the symbol L.
An
inductance of one henry gives rise to an E.M.F.
of one volt
when the
is
current varies at the rate of one
ampere per second.
The inductance
the current,
number
of turns of wire times the strength of the field, divided
and
;
also
by
10* to bring
it
all
by under the
practical system
thus.
ELECTROMACNETlC I'HENOMENA.
12$
Z. is
a constant in a coil without iron, but

iron
is
is
not a con-
on account of the variable permeability of the iron under different destant

in the
when
magnetic
circuit,
grees of magnetization.
In the design of quick-acting magnets,

consider the effects due to inductance.
is
it is
necessary to
the circuit
When
closed the current does not immediately reach a maxi-
mum,
law,
but requires a certain length of time, and Ohm's

(i),
/=
P
does not hold in this case.
However,
t,
it
may be
expressed at the end of any short time,
by
Helmholtz's law.
E /= Vi
.
_f\
I?
i</
(232)
where
2.7182818, which
is
the base of the Napierian
logarithms.
The
By
ratio
is
called the time constant of the magnet.
substituting
for / in (232),
= 634-
p
(234)
From
the above
it is
seen that the time constant of a
magnet is the time it of its Ohm's-law value.
takes for the current to reach .634
The time
inductance,
constant
or
may be decreased by
decreasing the
resistance.
by
increasing the
electrical
130
THE ELECTROMAGNET.
unchanged
;
since the E.M.F. will remain
therefore
it is
small percentage of the total obstruction to the passage of

the current, and the current reaches a given percentage
of
its final
value sooner.
65.
Alternating Currents.
With
alternating currents the inductance retards the
current, thus increasing the Effective Resistance.

fective resistance is called the Impedence,
The
ef-
and
is
equal to
(23s)
Z=Vp='-|-(2.rZ/7,
where
is
the inductance in henries and
/ is
the
num-
ber of complete cycles per second, one cycle being two

alterations or reversals of current.
of the current strength can not be easily due to the variable inductance, but if a curve be plotted showing the magnetic flux for each current strength, an accurate value of the current strength may be
calculated,
The exact value
obtained.
On
account of the inductance, an alternating-current
electromagnet should contain less resistance than one for

use with direct currents.
An
alternating-current-electro-
magnet should also be worked with lower densities duction on account of the hysteresis losses.
66.
of in-
Eddy
Currents.
Another loss is due to the currents set up in the iron by induction. These currents are called Eddy Currents, and are largely overcome by subdividing the core in the
direction of the lines of force.
For bar electromagnets, the core
may
consist of fine
wires, but for
131
magnets
of the horseshoe type, the cores
are usually
iron stampings, which should be insulated

of shellac or other insulating
from one another by a coat

material.
In the design of these cores, great care must be exercised in the selection of the proper size of wire or laminse,
for
if
the wire or laminae be too large in cross-section the

will
loss
due to eddy currents

if
be too great
on the other
will
hand,
the wires be too small in cross-section, or the in-
sulation
be too
thick, the
magnetic reluctance
be so
great as to
currents.
more than
offset the evil effects of the
eddy
APPENDIX
STANDARD COPPER WIRE TABLE.

(Copied from tbe "Supplement to Transactions of American Institute of Electrical Engineers," October, 1893.)
20 C. or 68 F., of Giving Weights, Lengths, and Resistances of Wires Matthiessen's Standard Conductivity, for A. W. G. (Brown Sharpe).
&
A.W.G.
136
THE ELECTROMAGNET.
EXPLANATION OF TABLE,
The data from which

lows
:
this table
has been computed are as
fol-
Matthiessen's
% 0 C.
standard
resistivity,
Matthiessen's temper-
ature coefficients, specific gravity of copper =: 8.89.
Resistance in
terms of the international ohm.

Matthiessen's standard
i
metre-gramme of hard drawn copper =
0.1469 B.A.U.
Ratio of resistivity hard to soft copper 1.0226,

i
Matthiessen's standard
metre-gramme of
soft
drawn copper =
0.9866 international ohms. 0.14365 B.A.U. @ 0 C. One B.A.U. Matthiessen's standard i metre-gramme soft drawn copper=
0.141729 international
ohm
o'^
C.
resistance for 20
i
Temperature
metre,
coefficients of
C.,.
50
C, and
8oC., 1.07968, 1.20625, andi.33681 respectively,

i
foot
= 0.3048028
pounds 453.59256 grammes.
Although the entries in the table are carried to the fourth significant digit, the computations have been carried to at least five figures.
The
last digit is therefore correct to within half
unit, repre-
senting an arithmetical degree of accuracy of at least one part in
two
inch
thousand.
The diameters
of the B.
&
S.
or A.
W.
G. wires are ob-
tained from the geometrical series in which No. 0000
= 0.4600
digit
and No. 36 = 0.005 inch, the neaiest fourth significant retained in the areas and diameters so deduced.
It is to
tivity
its
being
maybe permanently
be observed that while Mathiessen's standard of resisrecognized, the temperature coefficient of
variation which he introduced,
and which
is
here used,
may
in
future undergo slight revision.

F. B.
Crocker,
G. A. Hamilton,
Committee on " Units
W.
E. Geyer,
and Standards
A. E. Kennelly, Chairman,
Pounds per foot varies directly as the area. Pounds per ohm varies directly as the area squared. Feet per pound varies inversely as the area.
Feet per ohm varies directly as the area. Ohms per pound varies inversely as the area Ohms per foot varies inversely as the area.
squared.
APPENDIX.
137
BARE COPPER WIRE.

B.&S.
No.
138
THE ELECTROMAGNET.
BARE COPPER WIRE.

COMMERCIAL HALF
B.
SIZES.
&
S.
APPENDIX.
139
WEIGHT OF COPPER IN 100 POUNDS OF COTTON COVERED WIRE.

COMMERCIAL HALF
B.
SIZES.
&
S.
140
THE ELECTROMAGNET. WEIGHT OF COPPER IN

lOO
POUNDS OF
SILK INSULATED WIRE.

B.
&
S.
APPENDIX.
10 MIL.
141
DOUBLE COTTON, INSULATED WIRE.
B.
&
S.
142
THE ELECTROMAGNET.
5-MIL.
SINGLE COTTON.
B.
&
S.
No.
APPENDIX.
143
8-MIL.
DOUBLE COTTON.
SIZES.
COMMERCIAL HALF
No.
144
THE ELECTROMAGNET.
4-MIL.
No.
DOUBLE
SILK.
APPENDIX.
145
2-MIL.
SINGLE SILK.
No.
146
THE ELECTROMAGNET.
4-MIL.
DOUBLE
SILK.
SIZES.
COMMERCIAL HALF
No.
APPENDIX.
147
2-Mn,.
SINGLE SILK.
SIZES.
COMMERCIAL HALF
No.
148
THE ELECTROMAGNET.
TABLE OF RESISTANCES OF GERMAN SILVER WIRE.
SPECIFIC GRAVITY
8.5
(APPROXO-
APPENDIX.
149
PERMEABILITY TABLE.
Density of
I50
THE ELECTROMAGNET.
TRACTION TABLE.
APPENDIX.
ISt
8ths.
152
THE ELECTROMAGNET.
LOGARITHMS OF NUMBERS.
APPENDIX.
LOGARITHMS OF NUMBERS.
153
!S4
THE ELECTROMAGNET.
ANTILOGAHITHMS.
Proportional Fartt.
.01
m
.03 /)i
.05 .OS
.07
1000 1023 1047 1072 1096

1122 1148 1175 1202 1230
1002 1005 1007 1009 1026 1028 1030 1033 1050 1052 1054 1057 1074 1076 1079 1081 1099 1102 1104 1107 1125
1151
1012 1035 1059 1084 1109
1014 1038 1062 1086 1112
1016 1040 1064 1089 1114
1019 1042 1067 1091 1117
1021
1045 1069 1094 1110
.08 .09 .10

.11
1178 1205 1233
1127 1130 1132 1153 use 1159 1180 1183 1186 1208 1211 1213 1239 1242 1265 1294 1324 1355 1387
1135 1138 1161 1164 1189 1191 1216 1219 1245 1247 1274 1303 1334 1365 1396 1276 1306 1337 1368 1400
1140 1167 1194 1222 1250

1279 1309 1340 1371 1403
1143 1146 1169 1172 1197 1199

1225 1227 1253 1256 1282 1312 1343 1374 1406
.12 .13 .14 .15
1259 1262 1288 1291 1318 1321 1349 1352 1380 1384 1413 1445 1479 1514 1549
1585 1622 1660
1268 1297 1327 1358 1390
1271
1300 1330 1361 1393
1285 1315 1346 1377 1409 1442 1476 1510 1545 1581
16
.17
.18 .19
1416 1449 1483 1517 1552 1589 1626 1663 1702 1742
1419 1422 1426 1452 1455 1459 1466 1489 1493 1521 1524 1538 1556 1560 1563
1592 1629 1667 1706 1746 1596 1600 1633 1637 1671 1675 1710 1714
1429 1432 1435 1439 1462 1466 1469 1472 1496 1500 1503 1507 1531 1535 1538 1542 1567 1570 1574 1578
1603 1641 1679 1718 1758 1607 1644 1683 1722 1762
1611
.21
.22
23
.24 .25
1750 1754
1791 1832 1875 1919 1963
1648 1C87 1726 1766
1614 1652 1690 1730 1770
1618 1656 1694 1734 1774
26
.27
1778 1782 1820 1824 1862 1866 1905 1910 1950 1954
1995 2042 2089 2138 2188
1786 1828 1871 1914 1959
1795 1837 1879 1923 1968
1799 1841 1884 1928 1972
1807 1816 1845 1849 1854 1858 1888 1897 1901 1932 1941 1945 1977 1986 1991
.30
.31
32 33
.34
.35 JJ6
.37
2000 2046 2094 2143 2193

2244 2296 2350 2404 2460
2004
2051 2056 2061 2099 2104 2109 2148 2153 2158
2198 2249 2301 2355 2410 2466

2254 2307 2360 2415 2472
2014 2018 2065 2113 2163 2213
2023 2070 2118 2168 2218
2028 2037 2075 2080 2084 2123 2128 2133. 2173 2178 2183 2234
.38
2239 2291 2344 2399 2455

2'5I2
2259 23E5 2270 2275 2286 2339 2312 2317 2323 2328 2393 2366 2371 2377 2382 2421 2427 2432 2438 2443 2449 2477 2495 2500 2506
2547 2606 2667 2729 2793
2858 2924 2992 3062 3133
2518 2570 2576 2630 2693 2754
2523 S535 2541 2582 2594 2600 2642 2649 2655 2661 2704 2710 2716 2723 2767 2773 2780 2786
2553 2612 2673 2735 2799 2864 2931 2999 3069 3141
2559 2564 2624 261 2679 2665 2742 2748 2805 2812
2871 2877 2938 2944 3006 3013 3076 3083 81481 31S5
2818 2825 2838 2844 2851 2884 2891 2904 2911 2917 2951 2958 2965 2972 2979 2985 3027 3034 3041 304813055 8097 3105 3112 81191 3126
APPENDIX.
ISS
_...
INDEX
158
INDEX,
FAGS
Gauss German German

Gilbert
Silver
Silver
Wire Wire Table

.
.
53>55
.
148
Magnetic Magnetic Magnetic Magnetic
Induction
6
29
Leakage
Poles
.
Units
.
Magnetism
Magnetization
Heating of Magnet Coils Henry

.
...
. .
90
11
Magnetizing Force Magneto Generator
.128
4
Magnetomotive Force.
Horseshoe Electromagnet Horseshoe Permanent Magnet

Hysteresis
Mean
Perimeter
76 4^
.
....
Loop
.
Hysteresis
....
...
Hysteresis Loss
34 -35 -35
130
Mil Mutual Induction
127
N.
Neutral Line
Notation
. .
Impedence
Inductance
Induction
Insulating Linen
.127
127
....
.
.
6g, 15a
.
Oersted
Insulating Materials Insulating Varnish

Insulation of Bobbin Insulation of
150
Ohm
.69 ... 68
.
Ohm's Law
P.
Wire
63, 66
.
Intensity of Magnetization
11
Iron Clad Electromagnet
113
.
Iron Clad Solenoid
...
120
J-
Paper in Winding -72 Parallel Sides and Rounded Ends, Windings 82 Parallel Windings 57, 59 Permanent Magnets i
. .
Joint in Magnetic Circuit Joint Resistance
Permeability
...
.
.
.12
149
.
....
. .
Permeability Table
Plunger Electromagnet
120
122
L.
Polarized Electromagnets
.
Leakage
Coefficient
Polarized Bell Polarized Relay Portative Electromagnet
124
124
.
Limits of Magnetization
Linen, Insulating
Lines of Force
.
116
117
Practical
Working Densities
Lines of Induction
Press Board
....
69
117
Lodestone Logarithms
... ...
M.
. . .
Problems
Pull
...
R.
36, 84, 107, 126
Magnet
Magnetic Circuit Magnetic Field
4,
23
3
Radiation Ratio of Wire to Insulation Rectangular Windings

.
90
.
63, 66
.
76
INDEX.
between M.M.F. and Heating Relation between Wire and WindRelation
ing
IS9
FAGB
Time Constant
96
129
117
Traction
Traction Table
150
Volume
3g, 45, 59
12
Reluctance
U. Units, Electric
Units, Magnetic
Repeating Coils
Residual Magnetization
Resistance
...
.
...
. .
57 35 6 55 45
35
....
V.
6, 7
11
Useful Flux
30
Resistance Factor
Resistance Wires
...
Retentiveness
...
...
Varnish, Sterling
Volt
"W.
Rise in Temperature
91
69 6
S.
Watt
33
.
....
of
Saturation Point
Self Induction
Watts per Square Inch
....
.
.
...
94
11
...
. .
127
Weber
Weight
Copper
lie
Silk Insulated
Wire Tables
.
144-147
in
Cotton Cov63, 138, 139
Solenoids
...
Space Factor Square Windings Sterling Varnish
....
...
T.
.119
.
.
ered Vv
45
76
69
Weight of Copper in Silk Covered Wire 63, 139, 140 Winding Calculations 39 Winding Space 39
...
Work
.
at
End
of Circuit
104
Temperature Coefftcient 55,56,91 Theory of Magnet Windings 70

.
T
Yoke
27, 112
LIST OF
ON
WORKS
Electrical Science
PUBLISHED AND FOR SALB BY
D.
VAN NOSTRAND COMPANY

23 Murray
&
27
Warren
Streets
NEW YORK
ABBOTT,
A. T. The
Electrical Transmission of Energy.
A Manual
Design of Electrical Circuits.

8vo, cloth
Second
edition,
revised.
for the Fully illustrated.

84.50
ABUT OI<D, E. Armature Windings of

by Francis
Direct-current
Dynamos.
.
Extension and
. . .
Application of a General Winding Rule. Translated from the original German B. DeGress, M. E. With numerous illustrations, (in Press.)
Electrical
ATKIN sour, PBOF. A. A.

ism.
and Magnetic Calculations.
$1.50
ATKINSOHr, PH.IL.IP. The

12m, cloth,
.
Elements of Dynamic Electricity and Magne^
...
...
$2.00
Elements
Storage,
of Electric Lighting, including Electric Generation,
and
Distribution.
8vo, cloth
Measurement, Ninth edition, fully revised and new matter added.

81.50
Illustrated.
Power Transmitted by Electricity and applied by the Electric Motor, including Electric Railway Construction. Fourth edition, fully revised and new matter added. Illustrated.
12mo, cloth,
82.00
BADT,
F. B.
Dynamo Tender's Hand-book.
70 Illustrations.
16mo, cloth, 81.00 16mo, cloth.

81.00
Electric Transmission Hand-book.
Illustrations Illustrations
and
Tables.
Incandescent Wiring Hand-Book.

Bell Hanger's
and Tables.
12mo, cloth,
$1.00 81.00
Hand-book.
Illustrated.
12mo, cloth,
BIGOS,
C. H. W. First Principles of Electricity and Magnetism. A book for beginners in practical work. With about 350 diagrams and figures. Illus$2.00 trated. 12mo, cloth
BliAKESIiET,
T.
H.
dents and Engineers.
Alternating Currents of Electricity. For the Use of Stu$1.50 Third edition, enlarged. 12mo, cloth,
. .
BOIVITEI', O. E. The Electro-Plater's Hand Book. A Manual for Amateurs and Young Students on Electro-Metallurgy. 60 Illustrations. 12mo, cloth, $1.20
BOTTOJfE,
A Practical S. R. Electrical Instrument-making for Amateurs. Sixth edition. Enlarged by a chapter on "The Telephone." 80.60 With 48 Illustrations. 12mo, cloth
Hand-book.
Electric Bells;
and
all
With over
tion.
100 Illustrations.
:
about them. 12mo, cloth
Practical
Book
for Practical
Men.
$0.50
The Dynamo
How Made and How

12mo, cloth,
:
Used.
A Book for Amateurs.

A Hand-book
for
Sixth edi$0.90
100 Illustrations.
Electro-motors How Made and How Used. Practical Men. Illustrated. 12mo, cloth
Amateurs and
$0.75
BCBIER,
E. T.
Beginners.
Questions and Answers about Electricity. 12mo, cloth,
First
Book
for
$0.50
;
CARTER, E. T.
Motive Power and Gearing for Electrical Machinery a treatise on the theory and practice of the mechanical equipment of power stations for electric supply and for electric traction. Illustrated. 8vo, cloth $5.00
CROCKER,
F. B., and S. S.
Motors.
WIIEEI.ER. The
Vol.
I,
Practical
Management
.
.
of
Dynamos and
Eleventh thousand.
Illustrated.
12mo, cloth,
$1.00
CROCKER,
cloth.
F. B.
II.
Electric Lighting.
The Generating
8vo, cloth.
Plant.
8vo,
$3.00
Illustrated,
Volume
Distributing Systems
Its
and Lamps.
Illustrated,
$3.00
DTJTAlHdC EEECTRICITY.
Modern Use and Measurement, chiefly in its application to Electric Lighting and Telegraphy, including: 1. Some Points in Electric Lighting, by Dr. John Hopkinsoii. 2. On the Treatment of Electricity for Commercial Purposes, by J. N. Schoolbred. 3. Electric Light Arithmetic, by
R. E. Day, M. E. Science Series.)
Fourth edition.
18mo, boards.
(No. 71
Van
Nostrand's
$0.50
....
Second
EWIlf G,
J. A.
Magnetic Induction in Iron and Other Metals.
issue.
$4.00
Illustrated.
8vo, cloth,
FEEMIHrO, Prof.
Practice.
Vol.
J. A. The Alternate-Current Transformer in Theory and The Induction of Electric Currents. 500 pp. New edition.
.
Illustrated.
8vo, cloth,
...
Induced Currents.
594 pp.
j5.oo
Vol. II:
cloth,
.
The
.
Utilization of
Illustrated.
8vo, $5.00
FOSTER, HORATIO
diagrams.
A. (with the
collaboration
flap
of
eminent
specialists).
Electrical Engineers' Pocket-book.
1000 pp.
Illustrated with
many
cuts
and
$5.00
Pocket
size,
limp leather, with
GORE, Br. GEORGE.

retical
The Art
Illustrated.
of Electrolytic Separation of Metals (Theo-
and
Practical).
8vo, cloth
J3.60
GUT,
ARTHUR F.
C.
Electric Light
and Power, giving the Result

8vo, cloth.
its
of Practical
$2.50
Experience in Central-station Work.
Illustrated,
HASKIirS,
H. The Galvanometer and

Fourth edition, revised.
;
Uses.
A Manual
for Electricians
.
.
and Students.
12mo, morocco,
...
$1.50
Illus-
Transformers Their Theory, Construction, and Application Simplified. trated. 12mo, cloth
$1.25
HAWKIjrs,
Dynamo
;
C. C, M. A., A. I. E. E., and WAEEIS, F., A. I. E. E. The its Theory, Design, and Manufacture. 190 Illustrations. 8vo, cloth,
$3.00
Hay, ALFRED.
trated,
Principles of Alternate-Current Working. 12mo, cloth.
Illus-
J2.00
HOBBS, W.
R. P. The Arithmetic
E.
of Electrical
ous examples, fully worked.
New edition.
Measurements. 12mo, cloth,

.
With numer.
... $0.60
8vo,
$1.40
HOSPITAI.IER,
cloth,
Polyphased Alternating Currents.
Illustrated.
I jrnuCTIOBr COIl,S : How Made and (No. 53 Van Nostrand's Science Series.)
How Used.
Tenth
edition.
16mo, cloth.
80.50
INGATrDESCEBTT EL,ECTRIC I/ieHTIBTG: A

;
Practical Description of the Edison System, by H. Latimer. To which is added The Design and Operation of Incandescent Stations, by C. J. Field a Description of the Edison Electrolyte Meter, by A. E. Kennelly and a Paper on the Maximum Efficiency of Incandescent Lamps, by T. W. Howells. Fifth edition. Illustrated. 16mo, cloth. (No. 57 Van Nostrand's Series.), 80.50
;
JEHI,, FRASrciS. Member A.

tric
I. E. E. The manufacture of Carbons for ElecLighting and other purposes. A Practical Hand-book, giving a complete description of the art of making carbons, electros, etc. The various gas generators and furnaces used in carbonising, witii a plan for a model factory. Illustrated with numerous diagrams, tables, and folding plates. 8vo, cloth. Illus-
trated
$4.00
KAPP, OISBERT, C. E.
tion, Subdivision,
Electric Transmission of
and
Distribution.
Energy and Its TransformaPractical Hand-book. Foiu:th edition,

$3.50
revised.
12mo, cloth
190 pages.
Alternate-Current Machinery. trand's Science Series.)
Illustrated.
(No. 96
Van
.
. .
Nos$0.50
Dynamos, Alternators, and Transformers.
Illustrated.
;
8vo, cloth,
$4.00
KEMPE, H. R.
Tables,
,
The Electrical Engineer's Pocket>book Modern Rules, Formulse,

32mo, leather
$1.75
and Data.
KEnrif EIiIiY A. E.
I.
Theoretical Elements of Electro-dynamic Machinery. Vol.

$1.60
Illustrated.
8vo, cloth
KII^GOVR, M. H., and fSWAST, H., and BIGGS,

tribution
:
C.
H. W.
Electrical Dis$4.00
Its
Theory and Practice.

Electric Light Primer.
Illustrated.
8vo, cloth
liEVY,
comprehensive digest of all the most important facts connected with the running of the dynamo and electric lights, with precautions for safety. For the use of persons whose duty it is
C.
li.
A simple and
to look after the plant.
8vo, paper,
$0.50
I/OCKWOOD,
trated,
T. D. Electricity, Magnetism, and Electro-telegraphy. A Practical Guide and Hand-book of General Information for Electrical Students, Revised edition. 8vo, cloth. Profusely IllusOperators, and Inspectors.
2.50
EODGE, PROF.
a,
OHVEB J.
and
Signalling Across Space Without Wires
being
8vo,
S2.00
description of Hertz
his successors.
Third edition.
Illustrated.
cloth
l^0j[[]f Cr, A. E.
tion, revised.
A Hand-book of
16mo, cloth.
(No. 39
the Electro-magnetic Telegraph. Fourth ediVan Nostrand's Science Series,), . $0.50

. .
MAjr SFIEIiD, A.
cloth.
jr.
(No. 64
Van
Electromagnets Their Design and Construction. Nostrand's Science Series.)

;
16mo,
$0.50
mORROW, 3. T., and REIS,

12mo, cloth,
T.
Arithmetic of Magnetism and Electricity.

81 00
MVirRO, JOHUr,
C. E., and JTAMlESOir, AITDRE'W, C. E. A book of Electrical Kules and Tables, for the use of Electricians and Engineers. Fifteenth edition, revised. With niunerous diagrams. Pocket size, leather, $2.50
.
Pocket
UriPHER, FRAKGIS
Appendix on the Metbod
E., A. M. Theory of Magnetic Measurements, with an of Least Squares. 12mo, cloth, $1.00
OHM,
S. The Galvanic Circuit Investigated Mathematically. Berlin, Translated by William Erancis. With Preface and Notes by the Editor, Thos. D. Lockwood. 16mo, cloth. (No. 102 Van Nostrand's Science Series.), .$0.50
1827.
Dr. e.
OVDIir, MA1TRICE A.
revised.
taining numerous photo-reproductions, diagrams,

8vo, cloth, illustrated
Standard Polyphase Apparatus ajid Systems, conand tables. Third edition,

$3.0C
PAIiAZ,
A., Sc. D.
Treatise
cation to Electric Lighting. W. Patterson, Jr. Second edition, revised.
on Industrial Photometry, with special appliAuthorized translation from the French by George
8vo, cloth.
Illustrated
$4.00'
PARSHAIili, H.
tric
F., and
HOBART, H.
M. Armature Windings
and
of Elec-
Machines.
With
140 full-page plates, 65 tables,
165 pages of descriptive

$7.50
letter-press.
4to, cloth
PERRINE, F. A. C, A.
their Materials
struction,
Press.)
M., B. Sc. Conductors
for Electrical Distribution
and Manufacture, the Calculation of Circuits, Pole Line ConUnderground Working and other uses. 8vo, cloth. Illustrated. {In
PliANTE, OASTOBT. The

Effects created
Storage of Electrical Energy, and Researches In the by Currents combining Quantity with High Tension. Translated from the French by Paul B. Elwell. 89 Illustrations. 8vo, cloth $4.00
POPE,
F.
I..
Modem
and
Electricians
Operators.
Practice of the Electric Telegraph. An entirely new work, revised

Illustrated.
A Hand-book
for
and enlarged, and.

$1.60
.
brought up to date throughout.
8vo, cloth
Illustrated. 8vo, cloth,
POOI.E, J. The Practical Telephone Hand-book.
$1.50
PRITCBABD, O. G.
8vo, paper,
The Manufacture of
Electric Light Carbons.
Illustrated.
80.60
PREECE, W.
12mo, doth
H., and
S.
STUBB8, A.
J.
Manual
of
Telephony.
Illustrated.
$4.50
RAM, GII.BERT
cloth
The Incandescent Lamp and
its
Manufacture.
8vo, $3 OO
RANDAIiI<,
trated.
J. E. A Practical Treatise on the Incandescent Lamp. 16mo, cloth
Illus-
$0.50
RVSSEEE, STUART
tricity.
A.
Electric-light Cables
Illustrated.
and the Distribution
of Elec$3.00
Second
edition, revised.
8vo, cloth
SAIiOMOirS, Sir AVI1>, M. A.

agement of Accumulators. and enlarged. 12mo, cloth.
Electric-light Installations.
Vol.
I.
Man$1.50
Practical Hand-book.
Eighth edition, revised
Illustrated
SAIÔMOirS, sir DATID, H. A.

Vol. III.
:
7ol. II.
Apparatus
S2.25 $1.50
Application
SGHEIiIiEN, Dr.
II. Magneto-electric and Dynamo-electric Machines: Their Construction and Practical Applieaiion to Electric Lighting and the Transmission of -Power, Translated from the third Uerman edition by N S. Keith and Percy Neymann. Ph. D. With very large Additions and Notes relating to American Machines, by N. Second edition
S.
Keith.
Vol.
I,
with 353 Ulnstrations.

$5.00
SHELDOIir, PROF.
Electric
;
S.,
Ph.
D., and
IIOBART, MASON,
B. S.
Dynamo
net, $2.60
Machines.
Machinery Its Construction, Design and Operation. Second edition. 8vo, cloth. Illustrated,
Direct Cnri ent
SliOANE, Prof. T. O'COirOR.

trations.
Standard Electrical Dictionai-y.
300 Illus$3.00
8vo, cloth,
:
The Transmission and DistriSBTEIiEi, AI.BION T. Electric Motive Power bution of Electric Power by Coutinuous and Alternate Currents. With a Section on the Applications of Electricity to Mining Work. Second edition. 8vo,
cloth. Illustrated,
. .
.
S4.00
:
SWOOPE,
C. W. Lessons in Practical Electricity and Arithmetical Problems. 8vo, cloth. Illustrated,
Principles,
.
Experiments
.
net, $2.00
THOM, C,
and JONES, W. H. Telegraphic Connections, embracing recent methods in Quadruplex Telegraphy. 8vo, cloth. Twenty colored plates, $1 50 P.
THOMPSON, EDWARD
Science and an Art.
How
to
Make
Inveiitions
or.
Inventing as a
.
An
Inventor's Guide.
Illustrated.
8vo, paper,
$1,00
THOMPSOJT.
Prof. S. P. Dynamo-electric Machinery. With an Introduction and Notes by Frank L. Pope and H. R. Butler. Fourth edition. Fully Illustrated. (No. 66 Van Nostrand's Science Series.), .... $0.50
.
Recent Progress in Dynamo-electric Machines. Being a Supplement to " Dynamo-electric Machinery." Illustrated. 16mo, cloth. (No. 75 Van Nostrand's
Science
Series.),
. . .
$0.50
TREVERT,
E.
Practical Directions for

Illustrated,
.
Armature and Field-magnet Winding.

.
12mo, cloth.
$1.60
TIJIHIiIRK, Dr. Potential, and its application to the explanation of Electrical Phenomena. Translated by D. Robertson, M. D. 12mo, cloth, .... $1.25
TUNZEI/MANN,
cloth,
.
G. "W. de.
Electricity in
Modern
. .
Life.
.
Illustrated.
12mo,
.... $1.25
$3.00
URftUHART,
cloth,
,'
J. W. Dynamo Construction. A Practical Hand-book for the Use of Engineer Constructors and Electricians in Charge. Illustrated. 12mo,
.
WATT, AI.EXANDER.
of Voltaic Batteries,
Electro-Deposition.
Practical
Treatise
on the
and other Metals, with Descriptions Magneto and Dynamo-Electric Machines, Thermopiles, and of the Materials and Processes used in every Department of the Art, and several chapters on Eleotro-Metallnrgy. With numerous Illustrations. Third ... ^^..wt edition, revised and corrected. Crown, 8vo, 668 pages,
Electrolysis of Gold, Silver, Copper. Nickel,
...
Riontro-MetallurEV Practically Treated.
Eleventh edition, considerably en$1.00
WALKER, FREDERICK.
to
Wind
for
any Output.
Illustrated.
Practical Dynamo-building for Amateurs. How 16mo, cloth. (Mo. 98 Van Nostrand's
$0.50
Science Series.)
WAL.KER, SYDWEY
tical Treatise
T.
Electricity in our
Homte and Workshops.
A Prac.
on Auxiliary Electrical Apparatus.
Illustrated. 12mo, cloth,
$2.00
WEBB. H.
1,.
Illustrated.
A Practical Guide to the Testing of Insulated Wires and Cables. 12mo, cloth, $1.00
. .
....
WEEKES,
R.
W. The
Design of Alternate Current Transformers.
Illustrated.
$1.00
12mo, cloth,
WEYMOUTH,
and
F. MARTEST. Drum Armatures and Commutators. (Theory Practice.) A complete treatise on tlie theory and construction of drumwinduig, and of commutators for closed-coil armatures, together with a full resume of some of the principal points involved in their design and an exposition of armature reactions and sparking. 8vo, cloth. Illustrated $3.00
;
WILKINSOIV, H.
cloth
I>.
Submarine Cable-Laying, Repairing, and Testing.
8vo,
$5.00
WORDI?rIIAI, HEJTRY
Organization,
cloth. Illustrated,
C.
and Management. With
Central Electrical Stations; their Design, 136 diagrams and folding plates. 8vo, net $7.50
WORMEIiIj, R.
Treatise
Electricity in the Service of

of Electricity in
on the Application
and
J.
edited, with copious additions,
by E.
Man. A Popular and Practical Modern Life. From the German, Wormell, and an Introduction by Prof.
$3.00
Perry.
.T.
With nearly
850 Illustrations.
8vo, cloth,
YOUSTG,
EliTOW.
Electrical Testing for Telegraph Engineers.

8vo, cloth.
With Ap$4.00
pendices consisting of Tables.
Illustrated
***
A General Catalogue 80 pagrosof Worhs in all branches of Electrical Science furnished g^ratis on application.

The Electromagnet

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

The Electromagnet

Transféré par

Droits d'auteur :

Formats disponibles

fyxmll

BOUGHT WITH THE INCOME FROM THE

Van Nostrand Company

intend this booh to be sold to the Public

the Trade on terms which will not allow

Cornell University Library.

There are no known copyright

the United States on the use of the

3 1924 031 243 367

Chief Electrical Engineer

Varley Duplex Magnet

VAN NOSTRAND COMPANY

VARLEY DUPLEX MAGNET

author has endeavored to give the facts connectedly,

so that the reader

easily follow the reasoning without

which are placed

of the data, especially that concerning windings,

has been obtained from actual practice.

ELECTRIC AND MAGNETIC CIRCUIT CALCULATIONS.

Measure General Relations between Magnetic Units

Effect of Joint in Magnetic Circuit

Simple Principle of Calculating Windings Copper Constants

Points to be Observed in Practice

31. 32. 33. 34.

37. 38. 39. 40.

General Construction of Electromagnets

Relation between Magnetomotive Force and Heating

Advantage of Thin Insulating Material

ELECTROMAGNETS AND SOLENOIDS.

Direction of Flux in Core

Action of an Electromagnet 59. Calculation of Traction

Standard Copper Wire Table

Bare Copper Wire

(Commercial Half Sizes)

mercial Half Sizes)

142 142 143 I4j

2-Mil. Single Silk

2-Mil. Single Silk

Single Silk (Commercial Half Sizes)

Table of Resistances of German Silver Wire Permeability Table

Decimal Equivalents of Fractional Parts of an Inch

as in Fig. 41, p. 80.

as in Fig. 42, p. 82.

electromotive force in volts.

base of Naperian logarithms

magnetomotive force (English system). M.M.F. = magnetomotive force in gilberts.

combined resistance and space factor

watts lost per cubic centimeter of iron.

cross-section of insulation in circular inches.

ohms per pound ohms per inch. ohms per foot.

for bare wires.

ELBCTRIC AND MAGNETIC CIRCUIT CALCULATIONS.

that peculiar property occasionally pos-

course, pointed south.

Navigators took advantage of this

principle to steer their ships,

magnets which retain their magnetism for a

long time are called Permanent Magftets.

other and unlike poles attract each other.

In this book the term North-seeking Pole will be used

confused with the geographical North Pole of the earth.

north-seeking pole of a magnet