ELECTROMAGNETIC FIELDS

Theory, worked examples and problems

ELECTROMAGNETIC FIELDS
Theory, worked examples and problems

Ruth V. Buckley
Principal Lecturer,
School of Electrical Engineering,
Leeds Polytechnic
© Ruth V. Buckley 1981

All rights reserved. No part of this publication may be reproduced or

transmitted, in any form or by any means, without permission.

First published 1981 by

THE MACMILLAN PRESS LTD
London and Basingstoke
Associated companies in Delhi Dublin
Hong Kong Johannesburg Lagos Melbourne
New York Singapore and Tokyo

ISBN 978-0-333-30664-2 ISBN 978-1-349-05792-4 (eBook)

DOI 10.1007/978-1-349-05792-4

This book is sold subject to the standard conditions of the Net Book Agreement.

The paperback edition of this book is sold subject to the condition that it shall
not. by way of trade or otherwise, be lent. resold, hired out. or otherwise
circulated without the publisher's prior consent in any form of binding or cover
other than that in which it is published and without a similar condition including
this condition being imposed on the subsequent purchaser.
CONTENTS

Preface vi

1. ELECTRON BALLISTICS 1

2. ELECTROSTATIC FIELD THEORY 25

3. ELECTROMAGNETIC THEORY 50

4. PRINCIPLES OF ELECTROMAGNETIC INDUCTION 84

5. ELECTROMAGNETIC WAVES 108

6. PRINCIPLES OF GUIDED WAVES 135

Appendix Stokes' Theorem 161

Bibliography 163
PREFACE

Although most electrical engineering curricula include a study of

network theory as well as electric and magnetic field theory, it is
usually recognised that on most courses much of this subject mate-
rial can and is separated out into different syllabus contents.
This book on electromagnetic theory provides a chapter on elec-
tron ballistics separate from the field theory; the following
chapters concentrate on electrostatics and electromagnetics under
steady state conditions, with a chapter on time varying fields,
Maxwell's equations and their relations to each media property and
finally a chapter on waveguide developments.

The problems vary in their level of difficulty and by so doing

the author hopes that the book can be used at several stages of a
degree or diploma course, as well as the Council of Engineering
Institutions examinations.
The author wishes to express her gratitude to the Council, for
permission to use questions from their examination papers. The
answers given here are the entire responsibility of the author.
The author would also like to express her appreciation to her
colleague, Mr D. H. Trevena, for making many constructive comments
and for improving the presentation of the material.

R. V. BUCKLEY
1 ELECTRON BALLISTICS

This chapter is concerned with the dynamics of particle motion in a

vacuum, the electron being visualised as a particle with a negative
electric charge -e (1.60 x 10- 19 coulombs). It is associated with a
mass m which increases as the velocity of the electron increases
towards that of the velocity of light (3 x 10 8 m/s).

From the theory of relativity it can be shown that

m=

where m is the particle mass at a velocity v and c is the velocity of

light.

However, it is usual to assume that the mass is constant,

9.11 x 10- 31 kg for all velocities up to one-third of the velocity of
light.

Note also a ~arameter often found in this work:

elm = 1.76 x 10 1 C/kg.

ACCELERATION OF AN ELECTRON
If the particle is free to move, it will be subject to Newton's laws
of motion. Assume the electron is placed in an electrostatic field
of ~~rength E volts/m. Then the force exerted on it is

l- =- eE N

while the force required to produce an acceleration f is given by

F = mf N

therefore

f = - ~ E m/s 2
m
If the potential at the point under consideration is V volts then
from electrostatic theory it is known that
av
E = - ax vol ts/m

1
where x is the displacement of the electron from some reference point.
Therefore

ENERGY RELATIONSHIPS

Figure 1.1

Consider figure 1.1 which shows two arbitrary points P and Q in an

electrostatic field. Clearly some form of energy will be needed to
move an electron from P to Q, where the potential difference is
(V2 - VI) volts. Therefore
acquired potential energy = -e(V 2 - VI) J or coulomb volts

When (V2 - Vi) is positive, the negative sign is an indication of a

transfer of energy; alternatively it must be a gain in kinetic energy,
Let the velocities at P and Q be vI and v2 then

gain in kinetic energy = ~mv22

therefore
2 2 2e
v2 - VI = Iii (V2 - VI)

In many problems the electron must start with zero velocity so

that vI = 0, v2 = v, and from a point of zero potential VI = 0,
V2 = V. Thus

This equation shows that the electron's speed at any point in the
field depends only on the potential V at that point.

Note that so far the equations do not involve any co-ordinate

system, except the above arbitrary distance x.

2
Uniform Electrostatic Field

Consider an electron entering the field at point 0 in figure 1.2 with

a velocity v making an angle e with the vertical. It is now necessary
to use a co-ordinate system to indicate which planes the electron will
move through.

Ymax

Figure 1.2

E
x
= Ez = 0 • E
'y
- Y... volts/m
d
Consider a point within the field, such as point A with co-ordinates
x, y. The vertical and horizontal components of velocity are v cos e
and v sin e so that from Newton's laws of motion (initial components)

x =v sin e x t
y v cos ex t + 1
2 f t2
Y
Now

f - ~ E
Y m Y

eliminate t between the equations for x and y hence

This is the equation of a parabola and the problem is often to find

the condition for the electron to strike the upper plate or to
continue on its trajectory and re-strike the lower plate.

Uniform Magnetic Field

A charged particle moving in a magnetic field experiences a force only

if it moves at an angle to the field and the force is perpendicular
to both the direction of motion and to the direction of the applied
field, as shown in figure 1.3.

3
 Force = Bev sin n

~~ectron motion
/

Equivalent current

Figure 1.3

Consider an electron injected into a magnetic field with velocity

v m/s as shown in figure 1.4. The field of strength B tesla is into
the plane of the paper as shown.

Field B tesla

Figure 1.4
Now the magnetic force due to the field, the charge on the electron
and its velocity is

F = Bev
It will be held on a circular path of radius R by a centifugal force
equal to mv 2 /R, therefore
2
Bev IR
= IDV when a = 900
R = mv
eB
period of rotation T = 2uR
v
=~
eB
s

Note that the period depends on the field strength only.

Combined Fields

The plane magnetron is developed as a result of the movement of an

electron in perpendicular electric and magnetic fields, when the path
of the electron starting from rest at the plane cathode will follow a
cycloidal path.
4,
 The co-ordinate axes diagram is shown in figure I.Sa while figure
I.Sb shows the path.

--------------------+
• • •
Ely • e
y -e_~..:::;--_ u • • •
o x

(a) (b)

Figure 1.5

The acceleration in the x-direction is given by

and the y-direction by

dv
m --
dt = eEy - Bz eu

From these equations

2
d 2v = Bze du = _ (Bze_~ v
dt 2 m dt m)
The solution of this second-order equation in v is
E
v = ~ sin ~z ~t
therefore

u = ~ [I - cos (B z i) t]
It follows that the x and y co-ordinates for the position of the
electron are obtained from

dx or
u = --
dt x = 1 o
u dt

therefore
eE
x =~ (oot - sin oot)
moo 2
5
where w = eBz/m. Similarly for the y co-ordinate

v = dy
dt
eE
or y = Iv
t
dt

y = mw~ (1 - coswt)

The maximum value of y occurs when t

2mE
y
Ymax = eB z 2

Example 1.1

Two large plane parallel electrodes are placed 0.3 cm apart in a

vacuum and a potential difference of 400 volts is applied between
them. If an electron starts from rest at the negative plate and has
unimpeded motion to the positive electrode, calculate (a) the velocity
of the electron on reaching the positive plate, (b) the time taken, (c)

e:
the kinetic energy and (d) the force acting on it.

v = v )1

(a) v = (2 x 1.76 x 1011x 400)1 m/s

11.87 x 10 6 m/s

(b) Using Newton's law of motion to obtain the acceleration during

the electron's period of travel

Now u = 0 and s 0.3 x 10- 2 m, therefore

(11.87 x 10 6 )2 I 2
f 2 x 0.3 x 10-2 m s

2.35 x 10 16 m/s 2

also v u + ft, therefore

t 11.87 x 10 6 = 5.05 x 10- 10 s

2.35 x 10 16

(c) Kinetic energy = 1mv 2 J

= 1 x 9.1 x 10- 31 x (11.87 x 10 6 )2

= 6.41 x 10- 17 J

(d) Force acting on the electron = mf N

6
 9.1 x 10- 31 x 2.35 x 10 16

Example 1.2

An electron is projected with an initial energy of 600 electron-volts

in a horizontal direction into an electrostatic field of strength 10 5
volt/m which acts vertically upwards. Deduce the equations givin§
the trajectory of the electron and calculate its position after 10- s.

The path of the electron is shown in figure 1.6.

E
Y x

Position after 10- 9 s

-y

Figure 1.6

The electron-volt is the energy acquired by an electron when it

falls freely through a potential difference of 1 volt. Therefore

initial energy of the electron 600 x 1.6 x 10- 19 J

Therefore

initial velocity Vo = (
2 x 9.6 x 10- 17 )! m/s
9.1 x 10- 31

= 14.53 x 10 6 m/s

This velocity Vo is directed along the Ox axis and the position after
t s is

x = 14.53 x 10 6 x 10- 9 = 0.01453 m

Since the field direction is upwards, the force on the electron is

downwards and the trajectory will be as shown in figure 1.6.

Force = -eE N

7
 acceleration f = _ eEm m/s2
From Newton's second law of motion, the y position aftert s is given by
y = ! ft 2 = _ 21 eEm t 2 m

If a substitution for t is made, then

eEx 2
y = - 2mv02

which is the equation of a parabola.

0.0088 m

Example 1.3
An electron is started with a velocity produced by a potential
difference of 20 kV perpendicular to a uniform magnetic field of
10- 2 T.

Calculate the radius of the resulting circular path and the time
for one complete revolution. How would this time be altered if the
initial velocity of the electron were doubled?
!
Entry velocity v (2 x 1.76 x lOll x 2 x 10Q)2

83.9 x 10 6 mls
mv 83.9 x 10 6
radius R 0.0477 m
eB = 1.76 x lOll x 10- 2
2TT x 0.0477
period of rotation T s
83.9 x 10 6

3.57 x 10- 9 s

There would be no change in the time.

Example 1.4

A particle with a charge q moves with a velocity v in a region occupied

by an electric field with intensity E and a magnetic field with a flux
density B. Justify the statement that the force experienced by the
particle is given by the expression F = q(E + x B). v
If the electric field is zero and the magnetic field is uniform,
show that the particle moves on a helical path, with a constant
angular velocity perpendicular to the field and a constant linear
velocity parallel to the field.
[C.E.I. Part 2, E.F.N., 1968]

8
 q

,Figure 1. 7a

By Coulomb's law, the force between two particles is given by

F
s

where E is the permittivity of the medium. The electrical intensity

experienced by the unit positive charge + 1 is

and is a vector quantity, so that the force experienced by the particle

q is

Fs qE N acting in the same direction

_I
magn)'" :i,jd B
()
•
f--\ metreq
F

q
(b)

Figure 1.7b

The magnetic force acts perpendicular to both the field and the
velocity of the particle as shown in figure 1.7b

Fm = 1dl x BN
but I = qv, therefore

Fm = qv x BN
where df = 1 m. Thus

total force = qE + qv x B q (E + v x B)

9
 If an electron enters a magnetic field at some angle a, the force
experienced by the electrons will be due to the normal component of
velocity and the motion along the line of action of the magnetic field
will be unaffected.

helix
..

Figure 1.8

The axial component of velocity is v v cos a and the electrons

x
will move at this rate from left to right.
The normal component of velocity is v
y
=v sin a so that the force
acting on the electron, at right-angles to both the field Band the
velocity vyis

F = Bev sin a
If the electron moved under this force only, its orbit would be
circular and the radius would be
mv sin a
R
eB
The actual movement is a combination of the circular and the axial
velocities, so the path is a helix.

Pitch of helix P = vxT vx2uR

--
v
y

2uv cos a
(e/m)B
Example 1.5

Electrons projected with velocity 4.0 x 10 6 ms- 1 into a uniform

magnetic field of density B travel in a circular path of radius 200 mm.
An electric field, set up by the potential difference V between a pair
of parallel plates 50 mm apart, is now superimposed in a direction at
right-angles to both B and the initial direction of motion of the
electrons. Estimate B and V such that the electron beam passes
through the combined fields without deviation. Make a sketch to show
the relative directions of the two fields and the electron path.
[e.E.I. Part 2, E.F.N., 1977]

10
 f
s

E volts m- 1

B Tesla

Figure 1.9
mv 2
F
m
= Bev T
mv
or R = -m
eB

4 x 10 6
thus B 113.6 IlT
20 x 10- 2 x 1. 76 x lOll
mv 2 _ eV
F T- d
s

16 x 10 12 x 5 x 10- 2
or
20 x 10- 2 x 1.76 x lOll

= 22.73 volts

Figure 1.9 shows the relative direction of the fields and the path of
the electron.

Example 1.6

A stream of electrons, passing through a hole in a large metal plate

A, makes an angle of 60° with the plane of A. Parallel to and 50 mm
above A there is a second plate B, maintained at a potential lower
than that of A. Obtain an expression for the trajectory of the
electron stream between the plates.

If the velocity on passing through the plate A is 20 x 10 6 mis, at

what potential will the electron stream just graze plate B?
[C.E.I. Part 2, E.F.N., Specimen Paper 1974]

- fBO.866U
5 cm

Figure 1.10

11
 The electron stream enters the field at 0 with a velocity u making
60 0 with the plane of plate A; at some point on the trajectory the
co-ordinates are x and y.

x = O.sut

y 0.866ut + 1f t2
Y

where

f ~ E and E
y m y y

Eliminating t yields
x 2
Y = 1.732x -~ x 1.76 x lOll E ------
Y 0.2su2

lOll E x
2
1.732x - 3.52 x
Y u2

For the stream to just graze the upper plate with an entry
velocity of 20 x 10 6 m/s find dy/dt and equate the result to zero,
thus giving the time to reach a maximum vertical displacement-

dy _ 0.866u _ ~ E t = 0
dt - m y

0.866u
t =

therefore
(O.866u)2 (0.866u)2
Ymax 0.05 - 1
;;
1. 76 x lOll x E 1.76 x lOll x E
Y Y
1 x 0.866 2 x 400 x 10 12
thus E
Y 2 x 1.76 x lOll x 0.05

Now E V/d, therefore

Y
0.866 2 x 4000 x 0.05
V = 0.176 852 volts

Example 1.7

A pair of concentric, cylindrical electrodes are in a vacuum. The

outer electrode at zero potential has an inside radius R. The out-
side radius of the inner electrode is r and the positive potential
of this electrode is YD' Derive an expression for the potential at
any radius x between the electrodes, remote from the ends of the
cylinders.

12
 An electron is injected radially outwards through a small hole in
the inner electrode with a velocity

uo = 5.2 x 10 6 mls

If R = 24 mm, r =0.08mm and V = 50 volts,calculate the velocity of

the electron where x = 20 mm and also the velocity of the electron on
impact with the outer electrode.

dx

Figure 1.11

The potential V at any radius x is obtained from V

that is
-fi dx,
V-l R
- -
qdX
21fx£0

where q Cis the charge per unit length of electrode. Therefore

V = ---L log ~
X 21f£0 e x

when x = r, Vx = VO' therefore

Vo R
V
x 10ge(R/r) loge x

when x = 20 mm
V - 50 10 2.4 8 3 1
x - log (2.4/'0:8) g -2- = . vo ts
e
From an energy basis, as the electron decelerates

gain in potential energy = loss in kinetic energy

_ 1 _ v 2]
e x (50 - 8.3) - 2 m [(5.2 x 106) 2
X

v 2 27.04 x 10 12 14.68 x 10 12
x

= 12.36 x 10 12

v = 3.52 x 10 6 mls
x

13
On impact

V 2 27.04 x 10 12 - 17.6 x 10 12
x
9.44 x 10 12

v = 3.07 x 10 6 mls
x

Example 1.B

In figure 1.12, AA and BB represent two horizontal, parallel plates

distance 2y apart in a vacuum, A potential difference V2 exists
between the plates, with the lower plate positive, An electron,
accelerated through a potential difference VI enters the space between
the plates centrally at P in a direction at an angle e to the
horizontal, as shown. Derive an equation for the subsequent path of
the electron. Field fringing at the edges of the plate may be
neglected.

Figure 1.12
If e = 45°, V2 = 200 volts and y = 20 mm, calculate the value of VI
that will cause the electron just to graze the upper electrode and the
total time then taken for the electron to reach the lower plate.

The entry velocity v is found from

while

electric field intensity E

y
=-(-)=
-V2
-
V2
2y) 2y
- eV 2
force acting on the electron = - eEy 2y

The acceleration in the vertical direction f is found from

y

F _ e V2
f
y m iii 2y
e V2 2
vertical displacement Y = v sin e t - - -- t
2m 2y

while the horizontal displacement is found from

x = v cos et
14
eliminating t between the equations for X and Y, yields

Y v sin a X eV 2 X2
v cos a - 4my 2
cos 2 a
v

eV2 X2
X tan a - 4my 2
v cos 2 a
This is the.equation of a parabola. For maximum height, find dY/dt
and equate to zero, to find the corresponding time.
dY . eV2t
-dt = v S1n a-- - =0
2my

thus t = v sin a 2my

eV2
v 2 sin 2 amy
and Y
max eV 2

2 x 10- 2 = V2 sin2 45° x 2 x 10- 2

1.76 x lOll x 200

1.76 x lOll x 200

70.4 x 10 12
0.5

v = 8.39 x 10 6 m/s

Therefore
70.4 x 10 12 200 volts
Vl=
3.52 x 1011 3.52 x 1011

To find the time to travel to the lower plate y -2 x 10- 2 m

measured from the upper plate position.

t 12 = _~1:.:6_ __ 4
0.952 x 10- S s
4.2 x lOS
17.6 x 10 16

Time to reach the upper plate

8.39 x 10 6 x 0.707 x 2 x 20 x 10- 3
t = 1.76 x 1011 x 200 0.675 x 10- 8 s

Total time from starting point = 1.63 x 10- 8 s

15
Example 1.9

The potential difference between two parallel plates A and B in a

vacuum is v = Vo + Vm cos wt with plate A maintained at zero potential.
The distance between the plates is d. At time t = 0 an electron at
rest is released from plate A and attracted towards plate B. Derive
expressions for the subsequent velocity and position of the electron
in relation to wt, and sketch these functions.

If Vo = 3 volts, Vm = 15 volts, w = 5 x 10 7 rad/s and d = 80 mm,

calculate the kinetic energy of the electron and its distance from
plate A (a) when wt = u and (b) when wt. = 2u.

L f v = Vo + V
m
cos wt

ry
d

x
A 0

Figure 1.13

Now for a.c. fields

- V - V cos wt
E o m
Y d
V + Vmd cos wt)
acceleration fy ~ Ey = ~ ( ~o__~~_____

vertical velocity v =It f dt

y y
o
= ~
md
(v t 0
+
V
~sin wt) +
w
k

but when t = 0 v
y
=0 therefore k = 0

v = ~ (V t + Vm sin w~
ymd 0 w ')

vertical displacement y = X v y dt

~
md
(_02___ ..2.1!.
V t 2

w2
V
cos wt) + k'

When t 0, y o therefore k' = eVm/(m x d x w2 ), therefore

16
 vtZ v
y = -e ( -0 - - - m cos wt
md 2 wZ

When V
o = 3'm
V = 15 w = 5 x 10 7 and d
'
0.08 m

11 .
v 1.76 x 10 (3t +
y 0.08

When wt 11

v 2.2 x 10 1Z ( 3 x 11 + 01 = 0.415 x 10 6 m/s

y
5 x 10 7 )

and kinetic energy x 9.11 x 10- 31 x 0.172 x 10 12 J

also y _ _1_5_ _ cos wt +

25 x 10 14
15
25 x 10 14
J
2.2 x 10 1Z (1.5 112 0.6 0.6J
+ --+
,25 x 10 14 10 14 10 14

0.02 x l. 792 39.4 mm

When wt 211

v 2.2 x 101Z( 3 x 211) 8.3 x 10 5 m/s

y 5 x 10 7

and kinetic energy 1

2 x 9.11 x 10- 31 x 68.89 x 10 10 J

3 4112 0.6 0.6)

also y 2.2 x 101Z ( "2 x - - - - --+
25 x 10 14 10 14 10 14

52.1 mm

Example 1.10

The spacing between the parallel plates of a capacitor in a vacuum is

10 mm. There is a steady potential difference of 1.76 kV between the
plates and a uniform magnetic field of 0.01 T directed parallel to the

17
plates. An electron is released from the negative plate with zero
velocity. Calculate: (a) the transit time of the electron; (b) the
velocity with which the electron strikes the positive plate; (c) the
angle of incidence on impact.

Ey!

x
z ~
Figure 1.14

dv x
m dt = Bevy
dv
m --l = eEy - Bev
dt x

1.76 x 10 5 volts/m, therefore

dv
dt X = 1.76 x lOll x 0.01 x v
y
1.76 x 109vy

dv
~ = 1.76 x lOll x 1.76 x 10 5 - 0.01 x 1.76 x 10 11v x

3.1 x 10 16 - 1.76 x 109vx

d 2v dv
~ = - 1.76 x 10 9 ~
dt

Substitute for dvx/dt to give

The solution of this equation is found to be

Vy 1.76 x 10 7 sin 1.76 x 10 9t

thus v • 1.76 x 10 7 (1 - cos 1.76 x 10 9t)

x

18
 t
Y =j(o vy dt = - 0.01 cos 1. 76 x 10 9t + k

When t 0, Y = 0 therefore k = 0.01. Thus

y 0.01 (1 - cos 1. 76 x 10 9t)

Also x =). v, dt=1.76x10 7 (t-

sin 1. 76 x 109~
1. 76 x 109
+ k'

When t 0, x o therefore k'= 0

X=1.76 X 10 7 (t sin 1.76 x 10 9t)

1.76 x 10 9

1. 76 x 10 7 t - 0.01 sin 1.76 x 10 9t

When y 0.01 the solution for t is the required transit time

0.01 0.01 (1 - cos 1.76 x 10 9t)

1.76 x 10 9t TT
2"
TT
or t 0.89 x 10- 9 s
2 x 1. 76 x 10 9

To obtain the strike velocity, it is necessary to find v and v

at the transit time x y

v = 1.76 x 107 sin 1.76 x 10 9 x TT 1. 76 x 10 7

x
1. 76 x 10 9 x 2

Vy 1.76 x 10 7 (1 - cos I) = 1.76 x 10 7

1
thus strike velocity [(1.76 x 10 7)2 + (1.76 x 10 7)2]2

2.49 x 10 7 mls

angle of incidence on impact = 45°

Example 1.11

In figure 1.15, a constant magnetic flux density B = 1.42 mT and a

constant electric field intensity E = 10 kV/m act in an evacuated
space in a direction parallel to the 02 axis. At time t = 0, an
electron passes the origin 0 with a velocity u = 5.0 x 10 6 m/s in the
direction OY, as shown in the figure. At time t = 0.004 ~s, the
electron has reached the point P. Calculate the co-ordinates of the

19
point P and the velocity of the electron at this point. Determine
also the subsequent maximum and minimum values of the co-ordinates
x and y. Ignore relativistic effects.

Initial velocity VOy u = 5 x 10 6 , vOx = vOz = 0;

fz = 1.76 x lOll x 10 4 along the -z axis. The circular motion due to
the magnetic field yields
mv 2 mvO
Bev Oy ~or R = --y
R eB

5 x 10 6
thus R = - - - - - ' - - - - ' - ' - - - - - = 20 mm
1.76 x lOll x 1.42 x 10- 3
Projection of the path in the xy plane

v 0 + 1.76 x 1015t
z

when t = 0.004 x 10- 6

v 7.04 x 10 6 mls
z

z =- 14.08 mm

The angular velocity w = (e/m}B 2.5 x lOS radls and gives a

period

T ~ = 2.52 x 10-S s
w

The given time of 0.004 ~s is less than the period time T, so that
less than one revolution is made by the electron in reaching point P.
Now

wt 2.5 x lOS x 0.004 =1 rad or 57.3°

10 6
20
thus y R sin wt = 2 x sin 57.3 0 16.9 mm

x = -R(l - cos wt) = -9 mm

1
velocity at P = 10 6 (7.04 2 + 5 2 )2 8.63 x 10 6 m/s

20 mm and x = -40 mm
max

PROBLEMS

1. A cathode-ray oscilloscope has a final anode voltage of + 2 kV

with respect to the cathode. Calculate the beam velocity. Parallel
deflecting plates are provided, IS mm long and 5 mm apart, their
centre being 500 mm from the screen. (a) Find the deflection
sensitivity in volts applied to the deflecting plates per millimetre
deflection at the screen. (b) Find the density of a magnetic cross-
field, extending over SO mm of the beam path and distant 400 mm from
the screen, that will give a deflection at the screen of 10 mm.

[26.5 x 10 6 m/s; 2.67 volts/mm; 0.8 x 10- 4 T]

2. The anode and cathode of a diode are coaxial cylinders of radii R

and r respectively, with R > r. A potential difference V is maintained
between them. An electron leaves the cathode radially with negligible
velocity. Derive an expression for the subsequent velocity of the
electron as a function of the radius x from the axis of the cylipders.
Show that the velocity reaches half of its final value x = (Rr 3)4.
The effects of space charge are to be neglected.

3. Electrons are accelerated through a potential difference Vo to a

velocity v. They then form two long thin and parallel pencil beams at
a distance x apart. A current I is carried by each beam. Derive an
expression for the total force per unit length between the beams. If
the accelerating voltage Vo = 20 kV, x = 10 mm and I = 20 rnA, calculate
the force acting on each beam per metre length and indicate the
direction of the force.

[0.094 ~N per metre length]

4. An electron is accelerated through a potential difference V and

then enters a uniform magnetic field of magnetic flux density B in a
direction perpendicular to the direction of the field. Show that the
electron will describe a circular path, and derive an ~~pression for
the radius of the circle. If the flux density B is 10 T, and the
electron leaves the magnetic field 0.005 ~s after entering it, calculate
the angle between the direction of entry of the electron and its
direction of leaving the field.

5. A high-vacuum diode has concentric cylindrical electrodes with a

potential difference Vo between them. The inner electrode of radius r

21
is at zero potential and is the cathode. The radius of the anode is R.
Derive an expression for the potential V at any radius x between the
electrodes.
An electron leaves the cathode in a radial direction towards the
anode with an initial velocity u1 = 6 x 10 6 m/s. If r = 5 rom, R = 20 m
and Vo = 120 volts, calculate the velocity of the electron at radius
x = 15 rom and the velocity of the electron on impact. Sketch the
relationship between velocity and radius and estimate the transit
time of the electron between cathode and anode.

[8.34 x 10 6 m/s; 8.85 x 10 6 m/s; 2 x 10- 9 s]

6. Two large parallel plates are spaced 5 rom apart in a vacuum, and
a sinusoidal potential difference of V sin (oot + ~) is maintained
between the plates. An electron is injected into the interelectrode
space at time t = 0, at an equal distance from the two plates, and
with an initial velocity parallel to the plates. If the kinetic
energy of the electron at t = 0 is 1500 eV and oo/2~ is less than 1 MHz,
calculate the maximum permissible value of V for the electron not to
hit either plate within the first 20 rom of its traverse, whatever the
value of~. (Electron charge = 1.60 x 10- 19 C, charge/mass ratio,
elm = 1.76 x 1011 C/kg.)

[C.E.I. Part 2, E.F.N., 1975] [187.7]

7. The vertical distance between two horizontal flat plates in a
vacuum is x. The potential difference between the plates is v =
V sin oot and the lower plate is maintained at zero potential. At
time t = 0 an electron at rest is released from the lower plate.
Derive expressions for the subsequent velocity and position of the
electron in relation to oot and sketch these functions.
If V = 25 volts, 00 = 6 x 10 7 rad/s and x = 100 rom, calculate the
kinetic energy and the distance of the electron from the lower plate
in each case when oot = ~, 3~/2 a-d 2~.
[9.8 x 10- 19 , 2.46 x 10- 19 ,0 J; 38.4, 70, 76.8 rom]

8. A sinusoidal voltage of 4 Volts r.m.s. and frequency 10 MHz is

applied to the electrodes of a parallel-plate capacitor in an
~vacuated space. The plates are 30 rom apart. An electron leaves one
plate in a direction normal to the plate with a velocity of 2 x 105 m/s
at an instant t = 0 when the potential difference is a maximum and
the opposite plate is positive. Derive expressions for the velocity
v and for the displacement x of the electron at time t and sketch the
relationship between v and t and between x and t.

Calculate the position and velocity of the electron when

t 0.5 x 10- 7 s and estimate the time taken by the electron to reach
the further plate.

[2 x 105 + 5.28 x 105 sin oot m/s; 2 x lOSt + ~

200~
(1- cos oot) m;

22
 v =2 x 10 5 m/s; X = 26.8 mm; t ~ 1.5 x 10- 7 s]

9. The separation between two parallel plates in a vacuum is a. One

plate is maintained at zero potential and the other at + V. There is
a uniform magnetic field of flux density B parallel with the plates.
At time t = 0 an electron is released at zero velocity from the zero-
potential plate. Derive an expression for the velocity and for the
direction of motion of the electron at time t.

If a = 20 mm, V = 12 volts and B = 0.35 mT, determine the point of

impact of the electron on the positive plate, the transit time and
the angle of incidence on impact.

[9.26 mm for x; 0.0209 s; 65.2°]

10. A rectangular crossed-field system consists of a uniform electric

field E perpendicular to a uniform magnetic field B. An electron is
y z
injected into this system at velocity u. Derive an expression for
x
the trajectory of the electron and hence find the value u xO of the
injection velocity for which the trajectory of the electron is a
straight line.

[C.E.I. Part 2, E.F.N., Specimen Paper 1966] [E YIB z ]

11. A charged particle of mass m carrying a positive charge q is
initially at rest in a vacuum in a uniform electromagnetic field
comprising a static electric field of intensity E in the x direction
and a static magnetic flux density B in the y direction. Neglecting
gravitational force, develop the equations of motion of the particle
in terms of E and B. What is the maximum displacement in the x
direction from the rest position?

[C.E.I. Part 2, E.F.N., 1974] [x

max
= 2mE/(qB 2 )]
12. In figure 1.16 a constant magnetic flux density B = 1.5 mT in an
evacuated space acts parallel to the axis OZ and a constant electric
field intensity E = 20 kV/m acts in the negative direction parallel
to OY, as shown. At time t = 0 an electron passes the origin 0 with
a velocity u = 4 x 10 6 mls in the direction OY. Derive expressions

Z
t B
E
u
Y

Figure 1.16

23
for the co-ordinates x and y of the position of the electron at time
t, and calculate the maximum and minimum values of y attained by the
electron in its trajectory.

[Ymax = 1030 mm, Ymin = 0]

13. Derive an expression for the electric field strength in the
annular space bounded by two concentric cylinders when there is a
potential difference between them. An electron is injected with a
certain velocity and at a certain radius into the evacuated space
between the cylinders in a tangential direction. Determine the
relationship that must exist between electron velocit~ cylinder radii
and potential difference if the electron is to follow a concentric
circular orbit. Calculate the potential difference required to give
a circular orbit if the electron velocity is 10 7 m/s and the relevant
cylinder radii are 20 mm and 60 mm respectively.

[(e/m)V = v2 lQge(b/a); 624 volts]

24
2 ELECTROSTATIC FIELD THEORY

In general the term electromagnetism deals with the mutual inter-

action of electrical charges. When the charges are stationary, we
have problems on electrostatic fields and forces. If the charges
are moving uniformly so that a steady current is produced, then we
have problems on magnetic fields and forces (see chapter 3). Lastly,
if the charges are accelerating, so that the current produced is time
varying, then electromagnetic wave theory is involved, from which
radiating energy ideas may develop.

FUNDAMENTAL RELATIONSHIPS OF THE ELECTRIC FIELD

Coulomb's Law

Two small sphere-like charged particles ql, q2 (whose radii are very
small compared with the distance between them) are immersed in an
insulating medium of permittivity E. The magnitude of the force of
reaction between them is given by
ql q2
F = --N (2.1)
41TEd 2

where d is the distance between the particles.

An electric field may be defined as a region of space in which

electrical charges experience forces. Electric intensity. E volts/m,
a vector quantity, is an essential property of such a field and is
defined as the force experienced by a particle bearing unit positive
charge, at the point under consideration. (It must be noted that the
unit charge does not change or affect in any way the distribution of
the charge at the said point.) Suppose the field is due to a charged
particle q. Then from equation 2.1

(] x 1
E =~ volts/m (2.2)
41TEd 4

Superposition Principle

As the electric field is a linear function of the value of the charge,

the fields of more than one point charge are linearly superposition-
able by vector addition.

Electric Flux, ~ coulombs is a property of charge where

flux ~ = q C
2S
Gauss' Theorem

This states that the total electric flux through any closed surface
surrounding an electrical charge equals the amount of charge enclosed
by the surface.

Electric flux density D C/m2 is also a vector quantity, and is

sometimes referred to as electric displacement. The integral of flux
density D over a closed equipotential surface must be equal to the
enclosed charge.

For a charged particle and a sphere of radius d metres

flux density at the surface D = ~ C/m2 (2.3)

41Td 2

On comparing equations 2.2 and 2.3 the ratio of D to E is seen to be

that of the permittivity E of the medium

D (2.4)
E"=E
Electric potential, V volts is a scalar quantity defined as the
work done by extraneous forces against the forces of the electric
field, in moving a particle bearing unit charge from one point to
another.

XI
L
I---ox--j
....
I Xl

Figure 2.1
The difference in potential between points X and Xl in figure 2.1
is given by

oV = - E cos 8 ox

or ~~ = - Ex

where E E cos 8. In the limit as ox + 0, then

x

dV =_E
dx x

or V= -~Ex dx volts (2.S)

where V is the potential at the point X, or the work that must be

carried out in bringing a unit charge from infinity at zero potential

26
to this point. The negative sign is an indication that the work done
is against the electric field E.

It is now possible to evaluate not only the electric field

distribution of most conductor systems but also the potential
distribution and hence capacitance.

Method of Images

This is an extremely useful approach to many problems, when the

electric field due to charged particles or line charges associated
with conductors is required. For example, the simple expression for
the capacitance of a pair of parallel conductors, isolated from the
ground, having radii r and a distance D between centres is not true

TIE
(2.6)
loge (D-r)/r F/m

when applied to a two-wire transmission line above earth, since the

earth presents an easy path for the electric flux and so increases the
capacitance.

Boundary conditions between different dielectric media are of the

utmost importance in the study of electric fields. Reminder: each
coulomb of charge is associated with a coulomb of flux, whose course
may be altered by changes in the disposition of adjoining dielectrics,
but whose total is unalterable.

Energy associated with the electric field: the problem of

calculating the mechanical forces on bodies in electric fields could
be carried out using Coulomb's law for each pair of charges and
applying the superposition principle. However, it is found to be
more expedient to calculate the energy of the system and deduce the
forces from this result.

The energy of a group of charged conductors is potential energy,

recovered when the system is discharged.

In any conductor system, the energy is given by the expression

W= ~DE dv
the 'summation' being made for every volume element in the whole
field. From equation 2.4

W= !~E2 dv

or ~E~D2 dv (2.7)

Example 2.1

Derive an expression for the capacitance per unit length of a single-

core lead sheathed cable with two layers of different dielectrics.
27
Such a cable, designed to work at 66 kV to earth, has a conductor
diameter of 10 mm. The relative permittivities of the two dielectrics
are 4 and 2.5, respectively, with maximum working stress of 60 and
40 kV/cm 2 respectively. Calculate the minimum internal diameter
under the lead sheath.

a Core Radius
d Radius of First
Dielectric
b Radius of Second
Dielectric

Figure 2.2

Consider a charge q coulombs on the core conductor. Gauss' theorem

gives the stress in the inner layer as

while in the outer layer

If V volts is the potential difference between the core conductor and

sheath then
a d

V -~El
dx +~E2 dx
dab d
~(fe:dXx
o rl
+
~
~dXx)
Er2
d b

~ (f- loge ~ + f- loge £)

o rl r2
Capacitance per unit length = q/v farads

28
Maximum values of stress in layer I is

E = 9
maxI 21TEOE a
r1
V
or E
maxI
aE
rl (tTl log
e a
~ +
Er2 loge £)
Similarly
V
E
max2
dE r2 C £rl
loge
d
a +
1
Er 2 loge £)
Combining these two expressions

-E--=
max2 aErl

a = 5 mm therefore

d =5 x 4 x 60 12 mm
2.5 x 40

therefore
1 12 1 b 66
"4 loge 5 + 2.5 loge IT = - - - - ' - - ' - - - - - - 0.55
5 x 4 x 60 x 10- 1

1 b 1
2.5 loge IT - 0.55 - "4 x 0.8755
b
loge 12 - 2.5 x 0.331 0.83

b
IT= 2.29

b = 27.5 mm

internal sheath diameter = 55 mm

Example 2.2

A point charge Q is placed within an earthed hollow conducting sphere

of radius R at a distance d from the centre, as shown in figure 2.3.
Find the direction of the force experienced by the charge, and obtain
an expression for its magnitude in terms of Q, R and d·.
[C.E.!. Part 2, E.F.N., 1976]

29
 Figure 2.3

If the surface of the sphere is taken to be at zero potential, then

the field outside the sphere is due to the charge Q, and an image
point charge -Q' distance x from the centre of the sphere on a line
joining the centre of the sphere to the point charge.

Figure 2.4

potential at P --L _ Q' -

41TEO r 41TEO r ' - 0

then Q' = ~
r

It is necessary that r'/r is constant at all points on the sphere;

two particular points are at A and B on the line joining the two point
charges.

R- x R
rr' = (l.:R = d+R
+ x

solving yields x = R2/d therefore r'/r Rid and

Q' = ~
d

The force acting on the point charge Q is towards the sphere shell
and given by

F = Q Q'
-~.,.---
41TE o (d-x)2

30
(If the sphere were at a potential V, an additional charge
Q" = 4~£ORV must be placed at the centre of the sphere.)

Example 2.3

A capacitor is formed by two concentric spheres. Derive an expression

for the capacitance in terms of the radii of the spheres and the
permittivity of the medium between them.

If the radii are 500 and 100 mm, respectively, and there is a steady
potential difference between the spheres of 100 kV, calculate the
maximum value of the electric stress in the dielectric. Assuming
that the radius of the inner sphere may be varied, while the radius of
the outer sphere and the potential difference are unchanged, determine
the lowest value of the maximum stress which can be achieved and the
radius of the inner sphere for which it occurs.

Figure 2.5

The Gaussian area for the spherical combination shown in figure 2.5
is 4~x2

electric intensity Ex ~ volts/m

4~x2£

where q coulombs is the positive charge on the inner sphere. The

potential difference between the spheres is V volts, determined from

-=-.s.ldx
4~£ x2

.:....s.b [_ 1 ] a
4~£ X b

= -L
4~£ a (.!. - -bl) volts
.
capacltance C - V -
_ .1 _
(1 4~£
1 F
a - b)
31
therefore

E
x

E
v
max
a2 (a- - bl ) volts/m
1

100
12.5 kV/cm
2(.1
10 l.:fo - 501)
From the above equation

Emax (a -ab J = V
2

then for dE/da 0

2a = b

b 500 mm therefore

a = 250 mm

and the minimum stress is

100
E 8 kV/cm
max 2(,1 1~
25 \,.25 - 50)

Example 2.4

A circular loop, of radius r, of thin wire in a vacuum carries a total

charge q which is uniformly distributed. Derive an expression for
the electric field strength at any point along the axis of the loop.

An infinite conducting sheet is now placed parallel to the plane of

the loop, and 50 mm from it. If q = 0.01 ~C and r = 100 mm, calculate
the potential difference between the centre of the loop and the sheet.

dR.

+q
E
x

Figure 2.6

32
 Line density Po
'"
=~
211r

thus E
x

considering P£ d~ as a point charge.

The axis component E = Ex cos e while for the whole loop, the length
d~ becomes 211r therefore

E = P~211r cos e
411h2£0

or E
q cos e qx vo1ts/m
411hL£0 2
411(x + r2)t £0

-q
+q p

I.. 2D ------.!~

Figure 2.7

At the point P in figure 2.7 the total electric field strength is

given by

qx + q(2D - x)
411£0 (x 2 + r 2)'i 411£0 [(2D _ x) 2 + r2]t
by introduction of an image loop - q
o
potential difference V = -~E dx
2D
o
V = 4~£; [(X 1
2- + r2) 1 + [(2D - :)2 + r 2]J

= ~ [.!. ___
411£0 r (4D2
1__
+ r 2)1 (40' ~ r')! :o}]

33
The potential difference between the centre of the loop and the sheet
is a half of V, that is

For the calculation

potential difference

264 volts

Example 2.5

A capacitor comprises two coaxial conductors separated by a uniform

dielectric of relative permittivity E and resistivity p. If the
r
capacitance and the insulation resistance are C and R, respectively,
show that CR = E E p.
r 0

For a particular capacitor, Er = 2.5 and p = 8.7 x 10 13 n m. If

it is charged to a voltage V and then allowed to discharge through
its own insulation resistance, find the time taken for the terminal
voltage to fall to V/2.
[C.E.I. Part 2, E.F.N., 1971]

Figure 2.8

At any radius x

electric intensity E q
x 21TE E X
o r

where q is the charge per unit of conductor. The potential difference

between conductors is

34
Therefore
211£ £
capacitance per unit length
or

The corresponding insulation resistance is

p10g (b/a)
e- - -
R = ---';-2-1I..:. n

Thus CR = p£ £
r 0

Figure 2.9

From the given information

8.7 x 10 13 x 1 x 2.5 0.192

CR x lOt; s
3611 x 10 9

From the circuit in figure 2.9

i = -C(sv - V) and v = Rl

Eliminate 1

v-
CRY V
1 + sCR s + (l/CR)

vet) = V e- t / CR volts

I = V e- t / CR volts

or e t / CR 2

t CR x 0.694 0.192 x 0.694 x lOt;

1.33 x 10 3 5

Example 2.6

A capacitor with air as dielectric consists of a long metal rod of

35
radius a along the axis of a long metal tube of inside radius b. A
steady potential difference V is applied between the electrodes.
Derive expressions for the capacitance per unit length and for the
electric stress in the air at radius r.

Estimate the effective radius of the inner electrode and the

effective capacitance per metre length when V = 60 kV, a = 5 mm,
b = 100 mm, the breakdown strength of air is 30 kV/cm and the effect
of corona is to produce a uniform conducting layer round the rod.

If the same effective capacitance is to be obtained without corona

by wrapping the rod with a dielectric of relative permittivity 2.0,
find the thickness of dielectric required and the corresponding
maximum stress in the air.

From examples 2.5 or 2.1 it is easily proved that

21[£
c o
log (b/a) F
e

V
and E volts/m
max x loge(b/x)

where x is the effective radius of the inner conductor and corona.

60
30
x loge(b/x)

by trial and error, since x log e Cb/x) 2

x = 7.85 mm

Thus effective radius of inner electrode 7.85 mm

Hence

21[ x 8.85 x 10- 12 12

c log e (10/0.785)
21.83 x 10

21.83 pF/m

From example 2.1

V = ~ (_1_ log ~ + _£1 loge -x

b)
21[£0 £r1 e a r2
21[£0
c 21.83 pF
(2' loge "0:""5
x + loge 1xO)

36
Solving for x
211 x 10 12
0.5(loge x - loge 0.5) + (loge 10 - loge x) = 3611 x 109 x 21.83

- 0.5 loge x - 0.5 loge 0.5 + loge 10 2.545

2.303 + 0.3466 - 2.545

0.5

0.2092

x = 12.33 mm

Thickness of the dielectric 12.33 - 5 = 7.33 mm

v
Emax(air)
x(!log 1.233/0.5 + log 10/1.233)
e e

60
19.12 kV/cm
1. 233 x 2.545

Example 2.7

A region in the shape of a long circular cylinder contains a uniform

space charge of density q. Derive a general expression for the
electric field intensity E within the region and remote from the ends.

Two long parallel non-conducting cylinders having the configuration

shown in figure 2.10 contain between them a uniform space-charge
density q. Show that the electric field intensity within the inner
cylinder is everywhere the same, and is directed parallel to the line
of centres.
[C.E.I. Part 2, E.F.N., 1973]

Space Charge
Density q

Figure 2.10

Apply Gauss' law to a circular contour of radius r centred on the

axis of the cylinder, then

D x 211r x 1 = charge enclosed q1lr 2 (for 1 metre 1 ength)

37
thus D = S£ so that E = S£
2 2E

(E acts radially outwards for positive charge.) Using the principle

of superposition, consider a point with co-ordinates x, y within the
inner cylinder so that, due to +q filling the outer cylinder shown in
figure 2.10

E =~ and E 9l-
x 2Eo Y 2Eo

while due to -q filling the inner cylinder

E q(h-x) and E
x 2 EO y

Thus the field intensity due to both cylinders with q between them is

E
X
=~
2Eo
and E
Y
=0
Example 2.8

The lower plate of the parallel-plate capacitor shown in figure 2.11

is fixed, and the upper plate is attached to a spring of stiffness k.
With the capacitor unchanged the plate separation is a. Find an
expression for the maximum voltage Vmax and the corresponding
separation x for which the plate is in stable equilibrium with a
finite separation of the plates.
[C.E.I. Part 2, E.F.N., 1972]

T --_
~ ~
a
-.L ====~I -
-C:! -T-
Figure 2.11
Assume the plate area is 1 m2.
Spring force = k(a - x) N

electrostatic force = qE = CV)( V

x

since C = E/X for an area of 1 m2 . Equilibrium occurs when the two

forces are equal. Figure 2.12 shows the relationship between force
and distance for different voltage values. p and q are stable points

38
 Force

x
distance
Figure 2.12

while ~ and m are unstable.

For V the requirement is that the
max
curve should be tangential to spring straight line - hence the slopes
are equal. Slope of the spring law is -k while slope of the electro-
static law is dF/dx which is equal to

2£V 2.
max
-k

thus V 2
max

and x =

But the two forces are equal at this point, so

£V 2 kX13 kXl
k(a - Xl)
max
-- -=
X1 2
--- -2-
2X12

xl 3Xl 2
a - xl 2 or a -2- or xl '3 a
therefore

V ( k8a
3 )2 0.54
1

(~:3r
1

max 2£27

Example 2.9

Derive an expression for the capacitance of a two-wire line above

earth using the method of images.

39
 A P -q

-q
+q
Bl

Figure 2.13

The conductors A and B are of equal radii r distance d apart and

q Clm of charge per unit length of conductor is considered. By the
principle of images, the electric field above ground will be the same
as if the earth were absent and ~wo other conductors A' and B', as
shown in figure 2.13, reflections of A and B were present.

The potential difference between A and B is V volts equal to

a d-r
.....:L- log - - ignoring the images
7[£0 e r
At the point P
-q
electric intensity EA,p

Component in the direction A to B = EA,p cos ~A' EA,

or

-qx

A similar component due to conductor B' is

- qed - x)

Total p.d. due to these imaginary conductors is

f
r __
x_d_x_
4h2 + x 2
+ J (d - x)dx
4h2 + Cd _ x) 2
d-r d-r
40
 r
loge [4h 2 + Cd - X)2]]
d - r

2;£0 loge [4h 2 4~2 C: :2r) 2 ]

2~~0 loge [4h2 4: 2 C~ :2r) 2 ]

Total p.d. between conductors A and B is

~ log d
1[£0 e r

_q_ ~log -d 2 - log 4h2 + Cd - r)2]

-
21[£0 e r2 e 4h2 + r2

equivalent capacitance

Example 2.10

Estimate the capacitance of an air capacitor formed from two conduct-

ing plates of area 0.12 m2 spaced 8 mm apart. State the assumptions
made, and roughly sketch the electric field pattern in the region of
the plate edges when the capacitor is charged.

The capacitance is unchanged when the separation of the plates is

increased to 9.5 mm and a sheet of ebonite 2.5 mm thick is introduced
between them. Estimate the capacitance when the ebonite is in
position and the plate spacing is 8 mm.
[C.E.I. Part 2, E.F.N., 1975]

£0 Area
Capacitance of parallel-plate capacitor
plate spacing

1 x 0.12
C 132.6 pF

+Q

T
-L
V

-Q
!: 7mm

2.5 mm

Figure 2.14

41
The capacitance remains unchanged.

flux density D charge Q = EE

plate area A

In the air E = Q/(EoA); in the ebonite E = Q/(EoErA).

total potential V = - fE dx = .JL (d 1

EoA \
+
d
....l.)
Er

thus C

C 132.6 pF, then

d2
d d1 +
E
r

2.5
8 7 +
Er

E 2.5
r
If d 2 2.5 then d 1 5.5 for d 8 mm so that the new capacitance
value is

0,12 x 10 3
C = --~------ pF
36~ x 10 9 x (5.5 + 1)

163.2 pF

Example 2.11

A long straight cylindrical wire, of radius r, in a medium of

permittivity ~ is parallel to a horizontal plane conducting sheet.
The axis of the wire is at a distance h above the sheet. Derive an
expression for the capacitance per unit length between the wire and
the plane, stating any assumptions made.
Real Conductor A

T
10 cm h

Conductor B
Figure 2.15
42
 The potential difference between the wire and the sheet is 5 kV,
with r = 2 mm and h = 100 mm. Calculate the electric stress in the
medium at the upper surface of the sheet: (a) vertically below the
wire; (b) at a point 200 mm from the axis of the wire.
Assume h»r and conducting sheet infinitely large as shown in
figure 2.15. Evaluate the potential difference between A and B.

E -q- - (-q) Vim

x 2rrXE 2rr(2h - X)E

where q C/m is the charge per metre length .

v
2h - r
f Ex dx
.::.5L
2rr£
2h - r
fr dx
x +
dx
2h- x

.:L
2- [lOg
rrE e
x loge (2h x) J
r

2h - r
o 2h - r
= -=>-log -"-"- volts
rrE e r

VAD p.d. to the plane conducting sheet and is equal to

2h - r
VAD = ~ log
2rrE e r

capacitance per unit length 2rrE

-----~-----F/m
loge (2h - r)/r

The charge on the conductor is

5000 x 2rr
q C/m
36rr x 10 9 x log (20 - 0.2)/0.2

Assuming E = 1 then q 0.06 ~C per metre length

r
0.06 x 10- 6 x 36rr x 10 9
~----------~~--~~ volts/m
rr x 10 x 10- 2

21.6 kV/m

qh 0.06 x 10- 6 x 10 x 10- 2 x 10 4 x 36rr x 10 9

n(lOO + 300)

5.4 kV/m

Example 2.12

Show that (a) the normal component of electric flux density and (b)
the tangential component of electric field strength are continuous at
the boundary between two dielectric media.

43
 An electric field is directed at
media such that it makes an angle a
in one of the media. Determine the
and the angle between the field and
second medium.
the boundary between two dielectric
with the normal to the boundary
relationship between the angle a
the normal to the boundary in the

Figure 2.16 shows the configuration relative to the question where

ABCD is an extremely small cylinder enclosing both media.

Gauss' theorem is then applied to the volume ABCD. If there are no

charges on the surface of the cylinder, then the net outward flux
must be zero, and as AB+O, the flux through the curved surface does
not exist. Thus the normal components of the flux vectors must be
equal.

Figure 2,16

The voltage round a closed path, within an electric field is zero.

Thus

As AB+CD+O

Also Dl = tiEl and D2 E2E2 (see figure 2.17). Now apply the above
Normal
Medium 1
a

Figure 2.17

44
boundary conditions

D1 cos a = D2 cos 8

E1 sin a E2 sin 8
therefore

E1
D'l tan a

Example 2.13

A parallel-plate capacitor is formed by two large conducting plates

spaced by distance 2d. A dielectric material of permittivity £ = £ £
or
is deposited on one of the plates to a thickness d. The remainder of
the space is filled with a negative space charge of uniform volume
density p. The two plates are maintained at the same potential.
Derive expressions for (a) the surface density of the free charges on
the plates; (b) the density of the bound surface charge on the dielectric
material. Neglect fringe effects and assume the dielectric to be
isotropic and homogeneous.
[C.E.I. Part 2 E.F.N., 1970]

-p

Figure 2.18
(a)
From figure 2.18

E =D/E£
X X 0 r

VAB = 0 therefore

JE x dx 0
o

f
d
(PB-px)dx
thus
£0
o
o

45
 2d
[P:: - p~x ] a
r d

PB = 2 Pd(~)
1 + E
r

therefore

P = pd _ Ld (2___
+ Er_ ) = pdE r
A 2 1 + Er 2(1 + Er }

(b) the density of the bound surface charge on the dielectric material
is -PA (Er - 1).

PROBLEMS

1. A long cylindrical conductor of radius 20 mm and at a potential

of 250 kV lies with its axis parallel to and 0.1 m above a large plane
earthed conductor. Find the potential and the potential gradient ata
point mid-way between the centre of the conductor and the plane.
[C.E.I. Part 2, E.F.N., 1972] [214 kV; 3000 kV!m]

2. In a geoelectric survey a current of 250 rnA is fed into a spike

thrust into the ground, and taken out from another spike 100 m away.
Taking the current-spikes to be equivalent to hemispherical electrodes
of radius 0.1 m, find the potential difference necessary to drive the
current through the earth path, if the ground has a uniform resistivity
of 80 Q m.

Two potential-spikes are driven into the ground on the line joining
the current-spikes, and set symmetrically between them 20 m apart.
Find the p.d. between the potential-spikes.

Obtain an expression for finding the resistivity of the earth with

such a measuring arrangement, taking the electrode geometry as fixed.
[C.E.I. Part 2, E.F.N., Specimen Paper 1966]
[63.6 volts; 0.053 volts]

3. A standard air capacitor for use with a high voltage Schering

bridge has a capacitance of 80 pF and consists of a vertical
cylindrical working electrode mounted concentrically inside a longer
cylinder. End effects are eliminated by inner guard electrodes. The
length of the working electrode is 1000 mm and the inside radius of
the outer cylinder is 400 mm. The sinusoidal applied voltage is
100 kV r.m.s. Calculate the radius of the working electrode and the
peak electric stress at its surface, proving any formulae used.
If the capacitor has a power factor of 0.0001 at 50 Hz, and its
equivalent circuit is represented by a pure capacitance in series
with a resistance, calculate the series resistance.

46
 Sketch the electric-field distribution at the lower end of the
capacitor and use the diagram to explain the function of the guard
electrodes. [a = 200 mm; 10.19 kV/cm; 4 k~]

4. A parallel-plate air capacitor, of plate area A, and spacing s,

is connected permanently across the terminals of an electrometer, the
capacitance C of which is related to its terminal voltage V by the
expression C = K + aV 2 where K and a are constants. The combination
is charged to a potential difference VI and then isolated. When a
dielectric slab of area A and thickness d (with d< s)is fully inserted
between the plates of the capacitor, the electrometer indication
changes to V2. Neglecting fringing, find an expression for the
relative permittivity of the slab. Will V2 be greater or less than VI?
[C.E.I. Part 2, E.F.N. 1971]

1
[ "£ = V 2
r dL >(C + K+
S. Explain what is meant by space charge. Discuss the effect of
space charge on the conduction in a planar diode.

Show (a) that the average velocity of an electron in a space-charge-

limited diode is one-third of the maximum; and (b) that in the absence
of space charge the average velocity is one-half of the maximum.
Assume in each case that the initial velocity of the electron is zero.
[C.E.I. Part 2, E.F.N., 1972]

6. A long cylindrical single-core cable has a conductor of radius r

and an earthed metal sheath of inner radius R, the space between them
being filled with a homogeneous insulating material of absolute
permittivity E. The conductor potential is raised to a vOltage V.
Derive an expression for the maximum electric field intensity in the
insulating material, and show that (for given values of V and R) this
field intensity has its least value when log (R/r) = 1.
e
Sketch the radial distribution (a) of the electric field intensity,
(b) of the voltage to earth. Comment on the effect on these
distributions of taking into account the resistivity of the insulant.
[C.E.I. Part 2, E.F.N., 1974]
fE - Vir J
[max - loge (R/r) J

7. In figure 2.19 A and B are two long air-insulated conductors of

small diameter in a plane normal to that of the page. The mutual
capacitance of the pair of conductors is 6 pF per metre length and the
distance between centres is 13 m. The steady potential difference
between the conductors is 10 kV and the direct current carried by the
two conductors, in opposite directions, respectively, is 100 A.
Calculate the magnitude and direction of the electric and magnetic
field strengths at point P. Conductor A is positive with respect to
B, and carries current downwards into the page.

47
 Figure 2.19
[935 volts/m; 30° to AP; 13.8 A/m along PB]

8. State, in integral and differential forms, Gauss t law relating the

electric flux density D at a point to the volume charge density P.

Two infinite parallel conducting sheets A and B, separated by a

distance 2d, are maintained at the same potential. Sheet A is coated
to a thickness d by a homogeneous and isotropic dielectric material of
relative permittivity E = 2.0. The space between the dielectric and
r
sheet B has a space charge of uniform density p. Stating clearly any
assumptions made, determine the surface charge densities on A and B.
[e.E.I. Part 2, E.F.N., 1976]

9. A long cylindrical single-core cable has a conductor of radius

5.0 mm and an earthed metal sheath of inner radius 12.5 mm. The
insulant comprises two dielectrics, of equal radial thickness but with
different permittivities. If the maximum electric field strength is
to be the same in the two dielectrics, determine (a) the ratio of the
permittivities, and (b) the maximum electric field strength for a
voltage of 100 kV r.m.s.

Sketch the radial distribution of the electric field and of the

voltage to earth.

On what property of the insulant would the electric field

distribution depend if the cable were operated on a direct instead of
an alternating voltage?
[e.E.!. Part 2, E.F.N., 1977] [(a) 1.75:1; (b) 169 kV/cm]

10. Discuss briefly the analogy between the following field systems.
(a) A current between electrodes immersed in a conducting medium.
(b) An electric field established by a potential difference between
two electrodes embedded in a dielectric medium. (c) A magnetic flux
crossing a gap between two magnetised iron surfaces. Point out the
comparable quantities. and indicate any differences or assumptions.

Find an expression for the capacitance per unit length between two
concentric cylinders.

Using the result, find the resistance between two copper electrodes
comprising a disc of radius 20 mm centrally within an annular ring of

48
inner radius 100 mm, both hard-soldered to the surface of a sheet of
manganin 0.5 mm thick and of resistivity 0.5 pn m.
[0.256 mn]
11. Two parallel sheets of plate glass mounted vertically are
separated by a uniform air-gap between their inner surfaces. The
sheets, suitably sealed round the outer edges, are immersed in oil. A
uniform electric field in a horizontal direction exists in the oil.
The strength of the electric fluid in the oil is 1000 volts/m and the
relative permittivities of the glass and oil, respectively, are 6.0
and 2.5. Calculate from first principles the magnitude and direction
of the electric field strength in the glass and in the enclosed air,
respectively, when (a) the direction of the field in the oil is normal
to the glass surfaces, and (b) the direction of the field in the oil
makes an angle of 60° with a normal to the glass surfaces. Neglect
edge effects.

[(a) 416.7 V/m, 2500 V/m; (b) 892 V/m, 76.5°, 1525 V/m, 34.6°J

12. An air capacitor comprises two concentric thin-walled metal

spheres, respectively of radius R (outer) and r (inner).

(i) Derive an expression for the mutual capacitance of the system,

and use it to show that the capacitance of a single isolated sphere
is proportional to its radius.

(ii) If R is fixed and r can be varied, show that when a voltage V is

applied between the spheres, the maximum electric stress in the air
dielectric is at least 4V/R.

(iii) Determine the total capacitance to earth of the outer sphere

when the inner sphere is earthed.

[C.E.I. Part 2, E.F.N., 1980]

[ (;~]
- t)

49
3 ELECTROMAGNETIC THEORY

It should perhaps be remembered that the study of magnetism began

with the discovery of permanent magnets and that it was Coulomb who
in the late eighteenth century showed that the force between the
poles of two such magnets followed the inverse-square law.

As the forces on electrically charged particles are associated with

an electric field, so the forces on magnets can be ascribed to a
magnetic field.

However, Oersted in the early nineteenth century revealed that

when an electric current flows, it produces a magnetic field in the
area surrounding the current circuit.

It follows therefore that there must be a mechanical force on a

current carrying conductor when it is situated in a magnetic field -
say from a second conductor - and it is this idea which led to the
definition of the ampere.

R I -_ _ E
F ••

Figure 3.1

When the currents flow in opposite directions, the forces repel

one another
g,
F = I'B IdJi, = I'BJi, (3.1)

F = II' )JJi,
21TR

since B = ~I/(21TR) tesla

A unit pole may be defined as that magnetic pole which experiences

a unit force of one newton when situated in a magnetic field of unit
intensity.

50
 Magnetic potential is a useful and valid quantity in the study of
magnetic field theory and can be evaluated from the above unit pole
concept. The space derivative of magnetic potential gives the
magnetic intensity of the field

H =- grad Q =- VQ (3.2)

A difference in magnetic potential between two points a and b may be

related to the line integral of the vector H, then

Qb - Qa = - JH dx A (3. 3)
a
In this line integral, the path must always be taken in such a manner
that it does not link with any current.

If the integration is taken around a closed loop, starting and

ending at the same point, the value of the line integral is not zero.

Magnetic Work Law

The value of the line integral of H taken once around the circuit is
called the magnetomotive force (m.m.f.). Thus

j\H dx = m.m.f. A (3.4)

If the circuit carries conduction current i, then

m.m.£. = i

and if the current is carried by a winding of n turns of wire, then

m.m.£. = fH dx in (3. S)

that is, the line integral of H around any closed path is equal to
the current linked with the path. Particular attention should be
paid to the name m.m.f. which is customary for the line integral of H
around the circuit; m.m.f. is not a force nor is it work, but in fact
the units of magnetomotive force are amperes.

This means that, while magnetic equipotentials exist in a multi-

circuit, there is no implication that any such thing as a magnetic
charge exists in terms of which work in a magnetic field may be
specified.

Biot-Savart Law

The magnetising force at a point P due to current i flowing in a

circuit, is considered the same as if each element ot of the whole
circuit (see figure 3.2) contributed a vector oH given by

i lit sin 8
oH (3.6)

The direction of oH is normal to both the plane of ot and radius

vector r.
Sl
 p

Figure 3.2
The total magnetising force H is obtained from

f i DR. sin e
41fr2
(3.7)

Faraday's Law

Energy can be transferred from one circuit to another, not connected

to the first, through magnetic flux linkage. The two circuits are
then said to be magnetically coupled. The basis of this magnetic

f
action is Faraday's law of electromagnetic induction which states

e,m.f. = E ds = - ~~ (3.8)

that is a voltage is induced in a closed circuit when the magnetic

flux linking with the circuit is changed; the induced e.m.f. is
always in such a direction as to oppose the flux change.

Method of Images

As in the case of the electric field, it is possible to solve certain

types of problem by the creation of an image current carrying
conductor and the satisfying of the boundary conditions at the air-
iron surface.

Boundary Condition
At the boundary between two different media, the magnetic field may
change abruptly both in magnitude and direction. The normal

Medium 1 \11

--t
Medium 2
B "r2
N2

Figure 3.3

S2
components of flux density is continuous across the boundary between
the two media

BNl = BN2 (3.9)

The integral of magnetic intensity (H) around a closed path is equal

to the enclosed current.

H - H
Tl T2
= o or HTl = HT2 (3.10)

Tangential components of magnetic force are continuous across the

boundary between the two media, provided the boundary has no current
sheet.

Energy Associated with the Magnetic Field

A knowledge of the potential energy of a circuit (current x flux)

enables the calculation of the mechanical force acting on each element
of circuit to be made.

If a circuit is moved in a certain direction by a distance ox and

if the resultant force in that direction is Fx the potential energy
diminishes by Fxox. Therefore

F (3.11)
x

Suppose that a current carrying circuit deforms due to the electro-

magnetic forces set up by its own field, then the deformation will
always be such that the flux linkage increases, so that the stored
energy is greater after the deformation.

Assuming the current to be constant

energy supplied work done and increase in stored energy

id~ work done + ! i d~

therefore

work done = ! i d~

When the deformation consists of a displacement dx in a single

direction, the force F acting in that direction will give rise to
x
energy or work done

F dx
x
= ! i d~

d~
F 1
:2 i (3.12)
x dx

If L is the circuit self-inductance, then since iL ~

53
 i dL d~

thus

F 1 l' 2 dL
x 2 dx (3.13)

Example 3.1

The three conductors of a transmission line are supported at points

which form the vertices of an equilateral triangle RYB of side 1 m, RY
being vertical. At a certain instant the currents in conductors R.and
Yare both 120 A, with the return current of 240 A in conductor B.
Find the mechanical force on conductor B, giving both magnitude and
direction.

The conductor arrangement is shown in figure 3.4 with the

appropriate forces.

T-
R

1 m

L_ y

Figure 3.4

In the simplest form the force F acting on a current carrying

conductor of I amperes and conductor length ~, when lying in a
magnetic field of flux density B tesla is given by

F = Btl N

The flux density B associated with a current carrying conductor is

found from

B = ~T
27Td

where ~is the absolute permittivity of the medium and d is the

perpendicular distance from the conductor. Applying this to the
problem

S4
 4rr x 120 x 120 N/m
10 7 x 2rr x 1

Similarly

Both forces act as shown in figure 3.4. Therefore

horizontally away from conductors Rand Y.

Example 3.2

Derive an expression for the loop inductance for a pair of parallel

overhead lines carrying current in opposite directions, a distance D
apart in air, each with radius r. (Neglect the internal conductor
flux.)

-I

Figure 3.S

Although the internal flux linkage is to be neglected, it should

be remembered that the resultant self-inductance is a constant
quantity Yo/(8rr) H/m for each conductor, if the current density is
assumed uniform for non-magnetic materials.

To find the flux linkage due to the field surrounding the

conductors, apply the magnetic work law at point P.

The magnetising force H at P is given by

H x 2rrx
x
=I and H'
x
x 2rr(D-x) = -I
I
Total magnetising force = -2TTX
-+ I
2rr(D-x)
A/m

Total flux density at P

SS
 To evaluate the total flux linkage, consider a small radial strip
width dx around the conductor I of axial length I m. Then

flux linking this strip = ~~: [d~ + ~~x]

Total flux linkage between the conductors per loop metre length is

I
D-r
~oI 27T
(dX
x
+ dx ~
D-x-) =
~oI
-7T-
I
oge
(Dr-r) Wb-turns
-)
r
Self-inductance between the conductor is

L = -~o log (D---r) H per loop metre

7T e r

Example 3.3

An electromagnet has a short adjustable air gap of length g between

truncated conical soft iron poles, as shown in figure 3.6. The flux
can be taken as concentrated uniformly within the circular area of
radius a. Assuming the iron to have a constant relative permeability
~r' estimate the m.m.f. required for that part of the magnetic circuit

between planes X and Y for a gap of flux density B.

X Y
Figure 3.6

The m.m.f. for the part XY of the magnet may not exceed 3500 A t.
If b/a = 3, ~ = 120 mm and the relative permeability of the iron is
240, find the maximum gap length g for a gap flux density of 2.0 T.

From figure 3.6 deduce figure 3.7 g1v1ng a variable distance x to

work to. (m.m.f. = flux x circuit reluctance.)

56
 b
- a....... C-- =-=
y~
g
I

dx -t I-l
__ J
x - - - - - -...
Figure 3.7

A = 71r2
x

b - a
£= -9,-
x

thus A (b - a) 2 2
71 X
x 9,2

~ dx 9,2
m.m.f(x) = -----
].I7ICb - a)2x 2

total m.m.f. for one pole piece

Now Y = a9,/(b - a) therefore m.m.f. = ~9,/(71].1ab)

total m.m.f. for both po 1 es 2~9, = );'i)
= 71].1ab 2B9,a

Bg
m.m.f. for the air gap = ~=
].I07la 2 ].10

2B9,a Bg = -.! (29,a +

m.m.f· XY = --+
].Ib g)
].10 ].10 ].Irb

thus 3500 2 x
471
10 7 (2 x 12 x 10- 2
240 x 3
+ g)
271 x 3500
or g
10 7

1. 87 mm

57
Example 3.4

A long fluid conductor of circular cross-section, radius Rand

magnetic permeability ~ carries an electric current i. Assuming a
uniform current density in the conductor, show from first principles
that the electromagnetically produced hydrostatic pressure at a point
in the conductor distant r from the axis is

~i2[R2 _ r2]
4rr2R4

A gaseous conductor has a radius of 50 mm when the current is 200 kA.

Calculate the pressure at the centre of the conductor.

Figure 3.8

From figure 3.8 the current density J i/(rrR 2). therefore at any
radius r

current

magnetising force Hr 2rrr

corresponding flux density Br

change in hydrostatic pressure dP magnetic force

dr = rrR2

since the change in pressure is effectively the same as force per unit
area. For a metre length of conductor

dP Bri
dr = rrR2

dP

58
 R
~
p =jlli2r dr
21[2R4 21T2R4 [~2r
r
p lli 2 CR2- r2) N/m 2
41TR4

Numerically, when r = 0

41T x (200)? x 10 6 x 10 4
P __ ----~--~----------N/m 2
10 7 x 4 X 1T2 x 25

p 4 x 10 7
-=-=---- x _1 _ kg/m2
25 x 1T 9.81

0.519 x 10 5 kg/m2

Example 3.5

Derive an expression for the magnetic field strength at any point on

the axis of a single-turn loop carrying a steady current I: (a) when
the loop is in the form of a circle of radius r, and (b) when it is
square with sides each of length 2a.

Show that if the circle and the square have the same area, the
respective field strengths at the centres of the loops are in the
ratio (1[3/32) L

In this problem start by using the Biot-Savart law to find the

field strength at any point on the axis of a coil, as illustrated in
figure 3.9

~,H
& Figure 3.9

oH I or sin e
41TR,2

but or 21Tr and e = 90° for the whole coil therefore

Ir2
axial component H H sin cp =
c

59
 For a short conductor, obtain the field strength at a point
directly over the centre of the conductor, as shown in figure 3,10al
figures 3.l0b and c show the square of side 2a,

H
~ 2a--f Hs'2t S5

fw
~
,,/ I
/ I
A B ~ a BO
L-- za --I
Figure 3.10

For (a)

r r x 2 a rlz
H
s 4rrx
2 sin a 4rra an .. 4rra

Applying this result to the square and in particular figure 3.l0c

sin f3
rlz a r/2
H
ss
x ;a i = rr£'

Area of the circle rrr2. At the centre, £, r

r
H
cc 2r

Area of the square 4a 2 , At the centre £, = a

121
H
sc rra

H
thus
cc r rra rra
H= Zr i l l = Z}lzr
sc

but rrr2 4a 2 , or r 2a/lrr , therefore

G~/
H
cc rra Irr
H= i l l 2a -
sc

Example 3.6

(a) A small circular loop of radius r carrying a current I is in the

same plane as, and at the centre of, a large circular loop of radius
R, where R»r. Show that the magnetic vector potential at a point on
the circumference of the large loop is A = ~oI(r/2R)2.
(b) Using the result in (a), or otherwise, calculate the e.m.f.
induced in the large loop when the current in the small loop is

60
I cos wt.
[C.E.I. Part 2, E.F.N., 1972]

@ls
Figure 3.11

A -~E at, direction that of E; or E -aA/at

E n -aA oR,
at

thus ~A oR, =~ Wb

or lfA
s
OR, = !s = B tesla

The line integral of A round any closed circuit is equal to the

magnetic flux linked by the circuit. The flux density on the axis
due to the current I at a radius R from the small loop is given by

110 1
flux density ~ T

this statement is possible since R»r.

]..10 1
flux linking small loop = :lR x rrr2

The path round the large loop = 2rrR, therefore

A x 2rrR

(b) From above

E dA = 110lw sin wt (iii.r)2

= - cit

61
 f E d~ =
-d4>.
dt = 1nduced e.m.f.

thus e.m.f. = 2~R~OIw sin wt (2~2

Since w = 2~f
e.m.f. = ~OIf sin wt x R(~rR)V2
~oI£
= --(u) 2 sin wt
R

Example 3.7

Derive an expression for the mutual force between two coaxial circular
coils of radii RI, R2 , turns Nl,N2 and currents II, 12 respectively,
if the distance between their centres is L, as shown in figure 3.12.

Figure 3.12

Assume the smaller coil is so situated that the flux set up by the
larger coil completely envelops it and can be treated as uniform
over the area of the smaller coil, together with the fact that the
turns of each coil are co-planar.

Applying the Biot-Savart law to the larger coil, in order to find

the magnetic strength of the field on the co-axis, it can be seen
that

and since
RI
1
sin ~ =0Rl 2 + L2y

IINIRI2
HI 2(R12 + L2)t

62
 IJoIINIR12
thus flux density Bl
2(R12 + L2l
Hence

flux linkage with coil 2 = ~2N2

The mutual inductance between the coils is

flux linkage with coil 2

current of coil 1

7TR2 2 IJOIINIR12N2

2(R12 + L2)!r 1

IJ07TNIN2R12R22
2 (R12 + L2) t

Since the potential energy of one circuit in the field of another

is given by M12II12

dM12
force between the coils I 1 I 2""""dL

3IJo7TNIN2RI2R22II12L
F
2(R12 + L2)~

Example 3.B

A coaxial cable consists of an outer conductor of soft copper tube

with a mean radius of 10 mm and 1 mm thick. During a short-circuit
the current rises to 2 x 10 5 A in a few microseconds. Given that the
tensile strength of the copper is 2000 kg/cm 2 , estimate whether the
tube will burst.

Figure 3.13

63
 The flux density at the inner boundary of the outer concentric
conductor is B and is equal to ~OH, where

B ---'---=---=-=----T
41T x 2 x 105
10 x 21T x 0.95 x 10- 2
7

4.21 T

while the flux density at the outer boundary is zero. Therefore the
force on each element of area of the outer conductor varies from a
maximum at the inner boundary to zero at the outer edge. In figure
3.13

area of the shaded element =r 08 or

total segment area of outer conductor = 21Tr't = 21T x 10- 5 m2

Assuming current is uniformly distributed (no skin effect), then

, t h e element,
current 1n l' __ r 08 or I
21Tr't

0.95 x 10 8 x 08 x or
1T

radial force on this element --------N

4 x 10 8 x 08 x or
1T

4 x 10 8 x 08 x or sin 8
vertical component
1T

fa
1T
x 10 8 x or sin 8 08
thus 2 of = 4 'II'

4 x 10 8 or N
of
1T

Stress in this thin shell on the inner edge of the conductor is given
by

_o_F_ N/m 2 4 x 10 8
or -'--;:;-'-;;-. kg/ m2
x 1 1T X 9.81

1300 kg/cm 2

The average stress is 650 kg/cm 2 and the tube will not burst whether
the peak or average value is considered.

Example 3.9

Show, by the use of the method of images, how to obtain the complete
magnetic field pattern associated with a current carrying conductor
situated near an air-iron boundary.

64
 The iron is assumed to have a uniform finite permeability~. The
conductor is parallel to the boundary, with current flowing out of
the paper. Figure 3.14 shows an image conductor carrying I' situated
a distance below the boundary as shown, and a second image conductor
I" alongside the real conductor.

IrO~ _.
Image If

Figure 3.14

On the air side the field is created by I and If, both assumed to
be in air. On the iron side the field is given by I and I" , both
assumed to be in iron. Finally the magnitude and direction of the
currents in the image conductors must be such as to satisfy the
boundary air-iron conditions.

At the point P on the media boundary

I I'
H
I
= --.-
21f r
and H ,
I
=---
21f r

On the air side, the tangential component of the magnet ising force
is given by

h
(I - I')

h
cos a =
r

(currents in the same direction, forces oppose). On the iron side,

since

I "
21fr

h
H
T2
= (I + I")
21fr2
(currents assumed in opposite directions). But at the boundary,
HTl = HT2 , therefore

I' = - I"

65
 Now consider the normal components of flux density. On the air
side

(I+I')T

on the iron side

B = ~ (I + I-II ) T
Nl 27TrZ

But across the boundary BNl = BN2 , therefore

therefore

j.! - j.!O)
11 ( --- I for air
j.! + j.!O

while
j.! - j.!O)
III = - ( ---- I for iron
j.! + j.!O

If the iron has high permeability, j.!»j.!O' the field on the air side
is that of two parallel conductors with nearly equal currents flowing
in the same direction, situated in air, while on the iron side, the
field is that due to two coincident conductors (I and Ill) with the
currents flowing in opposite directions, all situated in iron.
Figure 3.15 shows a typical pattern.

Air Angle

Boundary
~~~~~:j==~~~~~Ma-gnetic
flux lines
Iron

Figure 3.15

Example 3.10

Show that the electromagnetic force on any conductor of a circuit of

inductance L carrying a current i may be deduced as an expression of
the form

66
where a motion dx of the conductor in the direction of the force
increases the circuit inductance by an amount dL.

A wire of radius 10 mm is bent into a circle of mean radius 1000 mm

and carries a sinusoidal current of which the r.m.s. value is 10 kA.
The inductance of a circle of wire may be assumed to be given by

where r is the radius of the wire, R is the radius of the circle and
~o is the permeability of free space. Calculate the peak and the
average value of the tensile force in the wire.

The answer to the first part of this question is given at the

beginning of the chapter, under the heading 'Energy associated with
magnetic fields'.

Consider figure 3.16, with a small shaded element of the circular

loop, at an angle 6 to the diameter. The force will act radially on
the loop, hence

F F

Figure 3.16

~~ = ~oR ~R x f + ~O (lOge 8: - 1.75)

~o (lOge 8: - 0.75J

therefore

or the force per unit length of circumference

.2
1 1 ~o (
f = 2" 2rrR loge

therefore radial force on the shaded element of figure 3.16 is fR 06,

67
while the vertical component of this force is f', where

f' fR 08 sin 8

thus 2F =~fR sin 8 d8

o
2F 2fR

F
~Oi2
= fR = ~
(
loge
8R
~
)
- 0.75 N

The peak value of current is 12 x 10 4 A, therefore

Fpeak 4~ x 2 x 10 8 (log 800 _ 0.75)

10 7 x 4~ e

118.6 N

average value IF
2 peak
59.3 N

Example 3.11

In figure 3.17, A and A' are long parallel conductors which are also
parallel to the surface of an infinite plate YOy' of infinitely
permeable iron. The conductors carry equal steady currents in
opposite directions. Derive an expression for the normal component
of magnetic flux density B at any point P on the surface of the plate.
Find values of y for which (a) B is zero and (b) is a maximum.

If a = 40 mm, b = 60 mm, and the current in each conductor is 100 A,

calculate the normal component of flux density at the point C.

X
IA A'-I

b b
a Air
p
y 0 C Y' Iron

Figure 3.17

From example 3.9, the field in the air is due to the two real
conductors A and A' and two imaginary conductors AI' AI' carrying
equal currents in the same direction. Take current A to flow into
the paper, while A' flows outwards, then conductor A and its image AI
give rise to flux density components, at the point P, of

68
 llOI 110 1 '
27fAP and 27fA I P

where I' is the image current in AI. The normal component is

110(1 + I')
(y - b)
BAN 27fAP AP

since AP

Ilo(y - b)21
BAN = ---------------
27f[(Y - b)2 + a 2]

Similarly for conductor A', the normal component is

110(1 + I') (y + b)

27fA'P A'P

Ilo(y + b)21

therefore

normal component of flux density at P = BAN - BA'N

1l0I [ (y - b) (y + b) ]
BNP = -7f- (y _ b) 2 + -(-y-+""--b-)-2--"+'--a-
2 T
a2

For BNP 0, BNA BNA , or

(y b) [(y + b)2 + a 2] (y + b) [(y - b) 2 + a 2]

multiplying out and cancelling where appropriate leads to

or y2 = a 2 + b2. Therefore

For BNP to be a maximum, differentiate the expression for BNP with

respect to y. Thus

°
d [ Y - b
dy (y _ b)2 + a2

[(y - b)2 + a 2] - [(y - b)2(y -b)]

[(y _ b)2 + a 2]2

69
 [(y + b)2 + a 2] - [(y + b)2(y + b)]
[(y + b)2 + a 2]2

a2 _ (y _ b) 2
[a 2 + (y _ b)2]

For convenience, let X = [(y - b)/a]2 and Y [(y + b)/a]2; then the
above equation becomes

1 - X 1 - Y
(1 + X) 2 (1 + Y) 2

so that provided neither 1 + X nor 1 + Y is zero

(1 - X)(l + 2Y + y2) = (1 - Y)(l + 2X + X2)

1 - X + 2Y - 2XY + y2 - Xy2 =1 - Y + 2X - 2XY + X2 - YX2

or (Y - X)(3 + Y + X XY) 0

r
thus Y =X
y +-
or (- a
bJ2 (y ~ b or y= 0
which must be a maximum since neither (1 + X) nor (1 + Y) is zero.
The second term, XY - Y - X = 3, must yield another maximum point

r r - r- r
beyond the position of the zero value.

(~ (y ~ b (Y : b (y ~ b 3

Complete the square to yield

(y2 _ b2) 2 _ (y2 _ b2) 4b 2
2 3 +
a4 a2 a2

and then add 1to each side to give

_ b2
(y2
a2
1r 4 (1 = + ::)

Take the square root of each side

1
y2 - b 2= 1 ± 2 (1 + b2 )2
a2 a2
Now y2 must be greater than b 2 , that is, y>b to give a second maximum
position

70
 1 1
Y = ±[a 2 + b2 + 2a(a 2 + b 2 )2]2

At the point C, y b 60 mrn

4Tr x 12 x 10- 2 x 2 x 100
BN 0.3 mT
10 7 x 2Tr x 160 x 10- 4

Example 3.12

A radial magnetic field of average density 0.1 T is maintained in the

air gap between two concentric vertical iron cylinders, as shown in
figure 3.18. A copper ring, placed symmetrically in the gap, is
released and falls through the magnetic field. Derive the equation of
motion of the ring, assuming a«r.

Calculate (a) the velocity of the ring at time t = 15 ms after its

release, and (b) the terminal velocity. (For copper, resistivity
p = 1.68 x 10- 8 n m) density 0 = 7800 kg/m 3 .)
[C.E.I. Part 2, E.F.N., 1970]

ring
Iron cylinde:ts

Figure 3.18

As the ring falls, an e.m.f. will be induced in it with a value

B x 2 Trr x v. This will generate a current i = e.m.f. t resistance.
The resistance of the copper ring = p~/A

R= P x 2Trr
Tra 2

thus i Bv Tra 2
p

A current carrying conductor in a magnetic field means that the

ring is subjected to a retarding force of B x 2Trr x i and is also
subjected to an accelerating force due to gravity. Thus

dv
~ = mg - B x 2Trr x i

where m is the mass of the ring and is equal to (2Trr)Tra 2 o

From the above equation

71
 dv B x (2nr) x B x v x na 2
-+ g
dt p x 2n 2 ra 2 1i

or

Substituting the numerical data

dv O.Olv
dt + -1-.6-S-x-'-'-1-'-0'::"_':"'8-x-7-8-0-0
9.S1

dv + 76.3v 9.S1
dt

Conversion

9.S1
sv + 76.3v --s

or -
v
9.S1
s(s + 76.3)

vet) = 0.129(1 _ e- 76 • 3t)

At t = IS ms

vet) 0.129(1

(a) vet) 0.088 mls

(b) terminal velocity 0.129 mls

Example 3.13

Develop an expression for the magnetic flux density at point P due to

a steady current I in each of the coplanar conductor configurations
(a) and (b) of figure 3.19. The conductors are situated in air.
[C.E.I. Part 2, E.F.N., 1973]

L7
a exp (-8/n)

~
I"
I I

0.6a
I /a
" I
V ""'- a
P (a) (b) ---.j
Figure 3.19

72
 (a) For a short conductor, with the point under consideration,
perpendicularly opposite the current carrying conductor, that is P

H = _1_ x sin S
41Td

Therefore, for the conductor furthest away from P

1
H '" 41T x 1.6a x 0.8

1
or H = 81Ta
while for the conductor nearest P

1
H = 41T x 0.6a
x 0.8

1.61
or H = 81T x 0.6a

Therefore

H - 1 (1.6 1
P - 81Ta 0.6- 4.81Ta

and flux density B =~ x 1 0.8331 T

P 107 4.81Ta - 107 x a

(b) Apply the Biot-Savart rule to figure 3.19b which is redrawn to

show the method of solution in figure 3.20.

a exp (-e/1T)

'\
1

Figure 3.20
r oe = 0)', sin ~

since x/r ~oe and x/o)', sin <1>. Now

1 0)', sin ~
oH
41Tr2

Ir 08 1 de 1 d8
---= 41Tr - 41Ta exp(- 8/ 1T)
41Tr2
73
 H
2TT
[
o
1m -
- 4TTa
I f
2TT

0
de
exp( - e/TT) '" ~
4TTa [exp(
2TT
+ e/TT)]
0

I (e2 _ 1) = ~
= 4a a

Therefore
4TT
flux density B = - -
10 7

Example 3.14

In figure 3.21, Land M represent long thin parallel conductors,

perpendicular to the plane of the paper and parallel to the surface
of an infinite iron plate of high permeability represented by PRo
The conductors each carry a steady current of 2000 A in the directions
indicated. Calculate: (a) the magnetic field strength, H, in
magnitude and direction at the points P and Q, and (b) the mechanical
force per unit length on conductor L.

I
T +-
Q L
-G)
I I
20 cm 20 cm ~

~ ~ 90~
Figure 3.21
Two image conductors M' and L' must be set up to represent the
effect of the iron on the two conductors situated in air, as shown in
figure 3.22.

M'

L'

Figure 3.22

74
 The magnetising force H for each conductor, including the images,
is shown in figure 3.22. It is evident that the tangential components
cancel out leaving
H =2 x _I_ x cos a
N 21fr

for M and M' at P

2546 Aim

for Land L' at P

H =2 x 2000 x 20 x 10- 2
1592 Aim
N1 2 x 1f X 800 x 10- 4

Overall magnet ising force at P 954 Aim acting vertically upwards.

Figure 3.23

To evaluate the magnetising force at point Q, consider figure 3.23,

which shows the field component due to each conductor and its image.

2000 x 10 2
21f x 22.36 = 1424 Aim
2000 x 10 2
21f x 20 1592 Aim

2000 x 10 2
21f x 36 884 Aim

75
 2000 x 10 2
HL, = 2~ x 44.72 712 A/m

1424 x 20 20
total vertical components 22.36 + 884 x 36 - 1592

20
- 712 x 44.72

1274 + 491 - 1592 - 318 145 A/m

total horizontal components = 1424 x ~ + 884 x 30

22.36 36

40
- 712 x 44.72

637 + 737 - 637 737 A/m

resultant magnetising force at Q = [145 2 + 737 2 ]! = 751 A/m

acting at an angle given by tan~l 145/737 to the line QL, that is,
-11.1° as shown in figure 3.23.

The mechanical force which acts on conductor L is due to the

magnetic field in which L lies, that is, flux density x current x
conductor length, or

=8 + 38 - 2 = 8.66 N upwards per unit length

Example 3.15

In figure 3.24, OA and OB are infinite plates of high magnetic

permeability meeting at 0 to form a 60° corner. A steady current I
flows in, the long conductor C, which is parallel to the plates and
centrally situated as shown. Sketch the system of images which, with
the current in C, will closely represent the magnetic field of the
current.

Figure 3.24

76
 If OC = 200 mrn and I = 800 A: (a) calculate the force per metre
length on the conductor and show the direction of the force, (b)
determine the magnitude and direction of the magnetic field strength
at a point P ju~t above the surface OB.

Figure 3.25

Each 60° segment of iron will be replaced with an image conductor

carrying the same current as in C and in the same direction.

(a) Using the expressions H = I/(2~x) A/m and B = ~oH T, find the
overall field strength at point C.

Hi 800 x 10 2 2000 A/m

2~ x 20 ~

Hii 800 x 10 2 2000

2~ x 20
- - AI,m
~

Hiii 800 x 10 2 1155 A/m

2~ x 34.6 ~

Hiv 800 x 10 2 1000 A/m

2~ x 40 ~

800 x 10 2 1155 A/m

HV 2~ x 34.6 ~

These vectors act at point C in the direction shown in figure 3.26.

Figure 3.26

77
 2000 1155 13 1000 1
Sum of horizontal components -----"-x
7T 7T 2"--7T- x Z

2000
+ _ 1
_ x_
7T 2

_ 5~~0 = _ 796

2000 13 1155 1000 13

sum of vertical components -7T- x 2" + -7T- + -7T- x 2"

1155 1
+-7T- x '2

= ---z:rr-
8657
= 1378
The resultant magnetic field strength is

acting at tan- 1 1378/-796 or 120 o .as shown in figure 3.26.

mechanical force on e = __
47T x 1591 x 800
10 7

1.6N

(b) As in part (a). calculate t~e magnetic field strength at point

P using H I/(27Tx) A/m. He and Hl cancel each other.

= _15_1_5 = HV
7T

800 x 10 2 = 1111 = Hiv

27T x 36 7T

These vectors act at point P in the direction shown in figure 3.27.

Figure 3.27

78
The horizontal components clearly cancel out leaving

sum of vertical components = 2 x 1515 cos 49 0 + 2 x 1111 cos 16.1 0

TI TI

632.7 + 679.5 1312 A/m

PROBLEMS

1. The three conductors A, Band C of a three-phase transmission

line are equilaterally spaced 1 m apart with conductor B vertically
above conductor C. The line carries a balanced load of 10 MVA at
33 kV line voltage. Calculate the mechanical force per metre length,
in magnitude and direction, produced magnetically on each of the
three conductors at the instant of peak current in conductor A.

[0.011 N/m; 0.0055 N/m; 0.0055 N/m]

2. A long concentric cable consists of a solid central cylindrical

conductor of radius a covered with insulation to a radius b over
which is a thick annular conductor of inner radius b and outer radius
c. The cable carries current in opposite directions in the two
conductors, respectively, and the current density in each conductor
is uniform. Obtain expressions for the magnetic field strength H at
any radius: (i) inside the central conductor, (ii) in the insulation,
(iii) in the outer conductor, (iv) outside the cable.

Sketch the variation of H with radius. Obtain an expression for

the self-inductance of the cable, and calculate its value per metre
length if a = 0.5 cm, b = 1.0 cm and c = /2 cm. The relative
permeability of the conductors and insulation is unity.

[0.22 llH/m ]

3. An air-cored solenoid of length 2£ is uniformly wound with fine

wire. There are n turns per unit length, the mean radius of the turns
is r and there is a steady current i in the winding. Derive an
expression for the magnetic field strength at any point along the
axis of the solenoid.

The solenoid is to be used as the primary of a standard mutual

inductor. The secondary consists of a short cylinder, wound with 50
turns of fine wire, placed coaxially and centrally inside the solenoid.
If n = 100 turns per cm and the ratio £/r 2, calculate the required
mean radius of the secondary coil to give a mutual inductance
M = 1 mHo
The secondary winding is now moved from the centre to one end of
the primary solenoid. Estimate the new value of the mutual
inductance.
ni ni£ ni£
[Z-(cos S - cos a); HC -"'-'---.. ; H d I
(r2 + £2)2 en (4£2 + r2)~
r = 23.7 mm; M = 0.547 mH]

79
4. If a constant-current circuit expands owing to the forces set up
by its own current, show that the mechanical work done is equal to
the increase in the magnetic energy stored in the inductance of the
circuit.

A round-section wire, radius r = 3 mm, is bent into a circle of

radius R = 300 mm, and carries a sinusoidal current of 5000 A. The
inductance of the circuit is given by

Assuming the radius of the wire remains constant for any small change
in the radius of the circle, calculate the peak and average values of
the tensile stress in the wire.

5. Show that the magnetic field intensity H at a distance x from an

infinitely long straight conductor carrying a current I is given by
H = 1/21TX.

Two such conductors of negligibly small cross-section, set

parallel at a distance 2d apart, carry equal currents in opposite
directions. Show that the lines of magnetic field intensity form a
system of circles.
[C.E.I. Part 2, E.F.N., 1968]

6. Show that the total energy stored in the magnetic field of two
coils with self-inductance L1 and L2, carrying currents i1 and i2
respectively, is

2"1 11
. 2L
1 +
1. 2L + . . M
2" 12 2 - 1112

where M is the mutual inductance between the coils.

Use this expression to deduce the relationship between the force of
attraction or repulsion between the coils and the rate of change of
mutual inductance with displacement between the coils. With the aid
of this relationship describe briefly the action of an elementary
rotary machine comprising a fixed and a moving coil.
[C.E.I. Part 2, E.F.N., 1968]

7. Show that the vector potential due to a current element can be

defined as the work done in bringing a parallel unit test element
(that is, I di = 1) from infinity to the point considered.

Two long parallel wires carry a low-frequency alternating current

in opposite directions. Derive an expression for the vector
potential at any point and use it to sketch an arrangement that will
measure the difference of vector potential between any two points on
opposite wires, without an e.m.f. being generated in the leads.
[C.E.I. Part 2, E.F.N., 1968]

80
8. Two large iron plates, one in a horizontal plane and the other in
a vertical plane, intersect to form a right-angled corner. A long,
straight conductor of small, circular cross-section is mounted in
the corner parallel with the plates, at a distance a from the vertical
plate and b from the horizontal plate. If the conductor carries a
steady current and the iron is of very high magnetic permeability,
make a sketch to show the magnetic field distribution in the space
round the conductor and between the plate.

Calculate the magnitude and direction of the force on the conductor

per metre length if the conductor carries a current of 1 kA and if
a = 300 mm and b = 400 rum.

[0.61 N]

9. A straight mercury column of circular cross-section, radius R,

carries a direct current I. Explain why the mercury is subjected to
a radial force, and deduce its direction. Obtain expressions (i) for
the force per unit area at a radius r within the column and (ii) for
the total axial force arising from this effect.
[C.E.I. Part 2, E.F.N., 1973]

10. A concentric cable consists of two thin-walled tubes, of mean

radii rand R, respectively. Derive an expression for the inductance
of the cable per unit length. The cable carries a sinusoidal short-
circuit current of 3 kA (r.m.s.). The mean radius of the outer tube
is 20 rum and its radial thickness is 2 rum. Assuming uniform current
density in the conductor, calculate the peak tensile stress in the
material of the outer tube.

11. (a) State the vector equation glvlng the Biot-Savart law, that
is, the magnetic flux density B due to a current element i d~. Using
this expression (or otherwise) find the magnitude of B at a distance
r from an infinitely long straight conductor carrying a current i.
(b) Two long straight parallel conductors, of small cross-section
and spaced a distance 2d, carry equal currents im cos(lOOnt) in
opposite directions. A small circular search-loop of radius a is
placed with its centre distant r from the axis of symmetry, as shown
in figure 3.28. Assuming that a«d, and that r»d (so justifying the
approximation 81 = 82), determine the maximum e.m.f. that can be
induced in the search-loop, stating the position(s) and attitude(s)
of the loop for this condition.
[C.E.I. Part 2, E.F.N., 1976]
[oB = ~iol sin 8; e.m.f. max na2~olOO di
m

small loop 8 = 0 or 180°, plane is

that containing the conductors]

81
 - -I-
I
/
/
Loop
/
Figure 3.28

12. A wire conductor of radius r = 0.015 m, bent into a circle of

mean radius R = 1.5 m, carries a sinusoidal current of 12 kA r.m.s.
Calculate the average and peak tensile forces on the wire. Take the
inductance of a circular loop to be

but prove any other formulae used.

[C.E.I. Part 2, E.F.N., 1975]
[F mean = 85.5 N; Fpea k = 171 N]

13. A magnetic flux crosses the plane interface between two media of
different permeabilities. Show that, on either side of the interface,
the normal compon0nts of flux density B are the same and the
tangential components of field strength H are the same.

A long straight conductor, in air, carrying a steady current I,

lies parallel to the plane surface of an infinite iron block of
relative permeability ~r' Show that the magnetic field strength H at
any point: (a) in the air, is the same as that due to current I in the
conductor and an image conductor II below the surface with the iron
removed, where

II = I(~)
~r + 1

and (b) in the iron, is the same as that due to a current

12 = 1(_2- )
~r + 1

in the conductor with the whole space filled with iron. If the
conductor is 100 mm above the block with I = 1500 A and ~r = 5.3,
calculate the force per metre length on the conductor.

[1.536 N/m]

14. A large circular coil of N concentrated turns and mean radius R

and a similar but small coil of n turns and radius r are set parallel
a distance x apart, with their centres on a common axis perpendicular

82
to the planes of the coils. Determine the mutual inductance between
the coils.

Let the large coil carry a steady current I, and the small coil
move towards it at a velocity u. Obtain an expression for the e.m.f.
induced in the small coil, and show that it has a maximum when x = ~R.
[C.E.I. Part 2, E.F.N., 1974]

[-~o48TINnr2u
----volts
]
25/5R2

15. Draw the system of images which, with the current in a long
straight conductor buried in (but insulated from) an iron mass as
3hown in figure 3.29, will closely represent the magnetic field of
the current. What are the assumptions made?

Estimate the flux density at point P and its direction, taking the
relative permeability of the iron to be constant at 1000.
[C.E.I. Part 2, E.F.N., 1973]
[0.7 T; -42.1°]

200A
---0-
50 I rnm
I
U:-:t
50 rnm
p

I
Air
Figure 3.29

16. The copper tube that forms the outer conductor of a concentric
cable has a mean diameter of 5 rnm and is 1 rnm thick. If the cable is
not to burst, what is the maximum current that it can carry during a
short-circuit test? The ultimate strength of copper is 200 GPa
(200 GN/m 2).

[C.E.I Part 2, E.F.N., 1980] [3.96 MA]

83
4 PRINCIPLES OF ELECTROMAGNETIC
INDUCTION

The basic relationships of the electrostatic and the steady magnetic

field were examined in chapters 2 and 3. Time-varying fields will be
discussed in this chapter, with particular reference to magnetic
induction.

Two new concepts are examined: the electric field produced by a

changing magnetic field and the magnetic field produced by a changing
electric field.

FARADAY'S LAW (Transformer Action)

When the magnetic flux linking an electrical circuit of N turns is

altered, an e.m.f. is induced in the circuit which is, at any instant
of time, equal to the negative rate of change of flux linkage.

d<l>
e.m.f. =- N dt volts (4.1)

The minus sign is an indication that the e.m.f. is in such a direction

as to produce a current whose flux, if added to the original flux,
would reduce the magnitu~e of the e.m.f. This statement that the
induced voltage acts to produce an opposing flux is known as Lenz's
law.

This equation can be applied to current carrying conductors but it

can also be applied to any circuit path in space through which the
flux may be changing.

It is also known that the electromotive force is any closed circuit

anywhere, is defined as the line integral of the electric field about
this circuit

e.m.f. = fE dx (4.2)

dx = _ N d<l>
dt

since

<I> = Is B ds (4.3)

then! E dx = - ~t I B ds (4.4 )

84
for a constant field and moving circuit, or

(4.5)

for a stationary circuit and changing field.

Now the e.m.f. around the closed path (~E dx) is of course the
induced voltage. Also ~ B ds gives the total flux over the surface

s, bounded by the closed circuit moving with respect to the field.

Or, if the circuit is stationary, this is the changing (decreasing)
field which is linking with the circuit. Applying Stokes' theorem
to the left side of equation 4.5

l(V x E)ds -
is
s
-aB ds
at

then curl E = - ~
at (4.6)

The line integral of an electric field about a closed circuit need

not be zero for a time varying field, and so work can be done in
taking a charge around a closed circuit in such a field. The energy
involved comes from the changing magnetic field.

FARADAY'S LAW (Motor Action)

An alternative form in which Faraday's law appears is

e.m.f. = J (v x B) dx

where dx is a differential element of wire length. This is derived

as follows. For a wire moving in a steady magnetic field the force
on an element of current dx is

dF
m
=i dx x B (4.7)

thus F q(v x B) (4.8)

m

but F
s = Eq
so E =v x B (4.9)

but e.m.f. = JE dx

thus e.m.f. = J (v x B) dx (4.10)

This equation therefore holds for charges in a vacuum or moving

through a wire.

85
 Finally a new concept is introduced, namely that of displacement
current.

Consider a voltage applied to a resistor and capacitor in parallel

- the nature of the current flow through the resistor is different
from that through the capacitor. Thus a constant voltage across a
resistor produces a continuous flow of current of constant value,
while the current through a capacitor will be constant only while the
voltage is changing.

Thus the current, ii' through the resistor is given by

V
(4.11)
R

and a current, i z , through the capacitor is given by

dq _ CdV (4.12)
iz dt - dt

The instantaneous charge q on the capacitor is given by q = CV.

The current i i is a conduction current while the current, i 2 ,
through the capacitor may be referred to as displacement current.
Although the current does not flow through the capacitor, the external
effect is as though it did, since as much current flows out of one
plate as flows into the opposite one.

The circuit idea may be extended to three dimensions by supposing

that the resistor and capacitor elements each occupy a volume area A
and length d.

Thus the electric field E equals the potential V across the element
divided by its length d

V
E = d (4.13)

The current density Ji inside the resistor is given by

ii
Eo (4.14 )
A

where 0 is the conductivity of the medium inside the resistor element.

If the capacitor is considered as a parallel-plate type then

C = £A (4.15)
d

where A is the area of the plates and d the spacing between them.

iz = £Ad de = £AdE (4.16)

d dt dt

The current density Jz inside the capacitor equals the permittivity

86
of the non-conducting medium filling the capacitor element multiplied
by the time rate of change of the electric field.

i2 dE
J2 = A = Edt (4.17)

Recalling that the electric flux density D EE, then

(4.18)

Hence J1 is the conduction current density J d while J 2 is a

con
displacement current density J d . . Also, current density J, electric
1SP
flux density D and the electric field intensity E are actually space
vectors, which all have the same direction in isotropic media, thus

J cond aE (4.19)

and Jd . (4.20)
1SP

If a single element filled with a conducting dielectric is used, so

that both conduction and displacement currents are present, then the
total current density Jtotal is

(4.21)

According to Amp~re's law the line integral of H around a closed

contour is equal to the current enclosed. Where both conduction and
displacement currents are present, this current is the total current.
Thus

f H dx =
Js'f(J cond + Jd "
1SP
)ds (4.22)

or f H dx = [CaE + E~~)ds (4.23)

The line integral of H on the left side of this equation is around

the boundary of the surface s over which the surface integral is taken
on the right side of the equation.

This equation may also be written in the form

(4.24 )

where J refers only to conduction current density.

By an application of Stokes' theorem to equation 4.24 then

87
 v x H (4.25)

or vx H = crE + E~
dt
(4.26 )

It should be noted that when the electric field varies harmonically

with time (E = Eo sin wt), the conduction and displacement currents
are in time phase quadrature; that is

J crE = crEo sin wt (4.27)

aD dE
and at = Eat = wEEo cos wt (4.28)

Thus when wt = 0, the displacement current is a maximum and the

conductor current is zero. On the other hand when wt = ~/2, the
conduction current is a maximum and the displacement current is zero.
Thus the displacement current leads the conduction current by 90°;
that is

E Eoe jwt (4.29)

aE "E jwt
then id" E- jWEE (4.30)
lSP at = eJw oe

thus equation 4.26 becomes

V x H = crE + jWEE = (cr + jWE)E (4.31)

The operator j in the displacement current term and its absence in

the conduction current term signifies that the displacement current
is advanced in phase by 90° with respect to the conduction current.

Example 4.1

Consider the fixed rectangular loop of area A shown in figure 4.1.

The flux density B is normal to the plane of the loop and is uniform
over the area of the loop. However, the magnitude of B varies
harmonically with respect to time as given by B = Bo cos wt. Find
the total e.m.f. induced in the loop.

1 ---~+-- Area A
e.m.f.e. 1~_______0_~________~
Figure 4.1

This is a pure case of B change only, since there is no motion.

Hence, the total e.m.f. induced in the loop is

88
 e = )[~s = AwBo sin wt volts

The e.m.f. appears at the terminals of the loop. Since the velocity
is zero there can be no motional induction.

Example 4.2

Consider the rectangular loop shown in figure 4.2. The width d of

the loop is constant, but its length x is increased uniformly with
time by moving the sliding conductor at a uniform velocity v. The
flux density B is everywhere the same (normal to the plane of the
loop) and is constant with respect to time. Find the total e.m.f.
induced in the loop.

v
c.m.f.c. o d

Figure 4.2

This is a pure case of motion only, the flux density B being

fv
constant

e.m. f. = x B)dx

vB~ volts

The entire e.m.f. ia this case is induced in the moving conductor

of length d. Since aB/at = 0, the e.m.f. by transformer induction is
zero, but if calculated from the flux linkage theory it is

e.m.f. d~ 1B ds
- Bd dx = Bdv volts
dt

This is the same as the first answer except for the sign, which is
arbitrary.

Example 4.3

Consider the same loop with sliding conductor as discussed in example

4.2. The flux density B is normal to the plane of the loop and is
uniform everywhere. The sliding conductor moves with a uniform

89
velocity v. The magnitude of the flux density B is varying
sinusoidally with time and is given by

B = BO cos wt

Find the total e.m.f. induced in the loop.

This is again a simple example but involves both motion and a time-
varying magnetic field B.

The e.m.f. em due to the motion is given by

e m= ~(V x B)dx vBd

= vdB O cos wt

The e.m.f. e t due to the time-varying field B is

et = Jf~s = wxdB o sin wt

s
total e.m.f. e = em + e t

vdBo cos wt + wxdBo sin wt

,
Bod[v 2 + (WX)2]2 sin (wt + 0)

where 0 = tan- 1 (v/wx) and x instantaneous loop length.

Example 4.4

A rectangular rotating loop lies in a sinusoidally varying magnetic

field B = Bo sin wt as shown in figure 4.3. The loop rotates with a
uniform angular velocity w rad/s. If the radius of the loop is R
and its length is ~, find the total e.m.f. induced.

B : ~ w )

:
~
2R

~
Figure 4.3

When t = 0, B o and e = O. Induced e.m.f. e

m
due to the motion
is

em = ~v x Bx d~ = 2wR~Bo sin 2 wt

90
 = wR~BO - wR~BO cos 2wt

The factor 2 is necessary because there are two conductors of length

~ moving through the field, the e.m.f.s in both aiding.
Induced e.m.f. due to the time-varying field B is

e t = - 2wR~Bo cos 2 wt

= - wR~Bo - wR~Bo cos 2wt

total e.m.f. e = em + et 2wR~BO cos 2wt

Note that the e.m.f. is at twice the rotation or magnetic field

frequency; also that the d.c. components cancel one another out.

Example 4.5

A three-phase transmission line carrying balanced three-phase currents

consists of three equilaterally spaced, parallel conductors. Show
that at a point equidistant from all three conductors, there is a
pure rotating magnetic field and find an expression for the field
strength and its velocity of angular rotation. Determine the value
of the magnetic field strength when the spacing between conductors is
1 m and the line is carrying 10 mVA at 33 kV line voltage.

R Space diagram
R

HR- -....- - - - - - Y
p

B y
'--------B

Figure 4.4

It is already known that a current carrying conductor creates a

magnetic field of strength H = I/(2rrd) A/m for a long conductor. In
this problem the distance d is the same for each conductor, but since
it is a balanced three-phase system of conductors, there is a time
phase difference between the currents

iR I cos wt
m

iy I cos (wt 2rr!3)

m

iB I cos (wt + 2rr/3)

m

Therefore the corresponding magnetic field strength modulii are

H cos wt
m

91
 H cos (wt + 2~/3)
m

at the point P, 120 0 spacing and perpendicular to the distance d from

each conductor. Alternatively, expressing the yellow and blue phase
magnetic field in terms of red phase, that is

so that the resultant magnetic intensity at the point P is

_ 2~'e-j1200 2~, +j120 0
Hm cos wt + Hm cos (wt 3j + Hm cos (wt + ~Je

or Hm {cos wt + cos
2~ cos
;r (wt -
2~
;r) 2~
+ cos :3 cos (wt + :3
2~)

_ j [sin 2; cos (wt _ 2;) _ sin 2; cos (wt + 2;~}

2 2~ cos wt - J"2 S1n
Hm( cos wt + 2 cos:3 2~"
"2 :3 S1n wt)

Hm(1.5 cos wt - j1.5 sin wt)

1.5 IH le- jwt with clockwise rotation

m
(The rotating flux direction can be reversed by interchanging any
two connections of the three-phase system.)
The above expression shows evidence that a rotating field is
produced which is constant at 1.51H I at the angular frequency of
w rad/s. m

For the numerical part of the question

10 7
IL = A(r.m.s.) .
13 x 33 x 10 3

while the peak current = 12 x 175 = 247.5 A, therefore

247.5
magnetic field strength -:::-~~-;:O:;"::;-
2~ x 0.577
= 68.37 A/m per conductor

constant field strength 1.5 x 68.37 = 102.4 A/m

92
Example 4.6
Figure 4.5 shows an elementary electrical machine in which flux
fringing may be neglected, the iron parts assumed to have infinite
permeability, and the gap length ~ is short. Show that, when the
g
rotor is displaced axially from the symmetrical position, an axial
force of B2a~g/~O' where B is the flux density under the pole shoes,
tends to restore the rotor to the position of symmetry. Sketch the
relationship of the axial force to the displacement and add the curve
that would be expected in practice.
[C.E.I. Part 2, E.F.N., 1968]

N turns
~
F
-L_I r{JF
r -I

~_
l__ 1__
1

N turns

(a) (b)

Figure 4.5

Insert a force F in figure 4.5b together with a displacement x.

Because the question states that flux fringing may be neglected and
that the air gap length is short, the solution can be simplified by
the use of the basic law of magnetic circuitry

flux = m.m.f.
reluctance

For an axial displacement of x m, the resultant flux is given by

~ =B x a x (~ - x) Wb

Ug
thus gap reluctance S
a(~ - x)~o

magnetomotive force = 2IN

2~g
Bx a x (~ - x) x

2B~
or 2IN --g
~o

93
The corresponding self-inductance L .p2N/I, or

L = B x a x (t - x) x 2N H
I

. 1 f orce F
aXla = 21'2
1
dL
dx

1 d
2 2INBa dx (t - x)

- 21 2IN x B x a

1 2Bt
- - --,gx B x a
2 IlO

B2 x a x t
----"g,g N
IlO

The negative sign indicates that the force is in the opposite

direction to that shown in figure 4.5, that is, position of symmetry
attained.

Force
- Practical case

Figure 4,6

•
•
E] T Field
....- - - t f - - - + -1-"',- B =
~ 1.0 sin
. . .--tL-d-jr--- 314t
•
Figure 4.7

94
Example 4.7

A single-turn square coil of 0.1 m side is placed in a uniformly

distributed time-varying magnetic flux of density B = 1.0 sin 100 ITt T.
It is rotated about an axis through the centres of a pair of opposite
sides, the axis being normal to the direction of the field, at a
speed of 3000 r.p.m. Assuming that the plane of the coil is parallel
to the field direction at t = 0, evaluate the e.m.f. induced in the
coil.
[C.E.I. Part 2, E.F.N., 1972]

3QOO
Rotational speed = -go- = SO revls

linear velocity = IT x 0.1 x SO 15.7 mls

vertical velocity component v

v
v cos ~ = 15.7 cos e
area of coil = 0.1 x 0.1 = 0.01 m2

Induced e.m.f. due to flux cutting e 2Btv v ' or

g

e
g
2 x 0.1 x 1.0 sin 3l4t x 15.7 cos e

- 3.14 cos e sin 3l4t volts

The flux through the coil =~ BA sin ~, or

~ = 0.01 sin e sin 3l4t

since t d. Induced e.m.f. due to changing flux is

d~ d
ec - dt = - dt(O.Ol sin e sin 3l4t)

3.14 sin e cos 314t volts

total induced e.m.f. = e + e
g c

3.14 cos e sin 3l4t -3.14 sin e cos 3l4t

3.14 sin(3l4t + e)

Note also that e = 1001Tt, so that the total induced e.m.f. can be
expressed in the form -3.14 sin 628t. (Note that because the minus
sign is used for e , it must also be used for e. On the other hand,
c g
there is no basic reason why the direction taken cannot be considered
as positive for e as long as a positive sign is also taken for e ,
g c
but the author feels the first approach is more fundamental since it
could take into account the cases where the plane of the coil is not
parallel at t = 0.)

95
Example 4.8

Two coils with self-inductances Ll = 1.5 Hand L2 = 0.03 +

0.005 cos 28 H carry direct currents of II = 7 A and 12 40 A
respectively. If the mutual inductance between them is
M = 0.2 cos 8 H, sketch the torque as a function of rotor position 8.
Calculate the maximum torque and show that the contribution from the
variable self-inductance is 1/15 of the total.

For the geometry of the machine, one must decide whether Ll or L2

is the rotor - figure 4.8 shows L2 as the rotating coil.

I
~Ll = l/'\M __
~ ') 0.2 CO' e

0.03 +
0.005 cos 28
/

Figure 4.8

Use the known torque equation between stator and rotor windings.
1 dLI 1 dL2 " dM
T "2 i 1 2"""""d8 + "2 i 2 2"""""d8 + 1112d8

1 d ) 1 x 40 2 x d8(0.03
d
-x 7 2 d8(1.s +"2 + 0.005 cos 28)
2
d
+ 7 x 40 x d8(0.2 cos 8)

Torque

Figure 4.9

96
 o+ 800(-0.01 sin 26) + 280(-0.2 sin 6)

(8 sin 26 + 56 sin 6)

dT
d6 = - (16 cos 26 + 56 cos 6)

for a maximum torque

16 cos 26 + 56 cos 6 0

or cos 26 + 3.5 cos 6 0

thus (2 cos 2 6 - 1) + 3.5 cos 6 =0

cos 2 6 + 1.75 cos 6 - 0.5 =0
cos 6 =- 1.75 ± 1(1.75 2 + 2)
2
-1.75 ± 2.25
2

thus cos 6 0.25, 6

therefore

T
max 8 sin 151° + 56 sin 75.5° 3.88 + 54.21

= 58.1 Nm

The contribution from the variable self-inductance is

3.88 1
58.1 = IS
Example 4.9

An elementary electrical machine, as shown in figure 4.10 has a

distributed type of single-phase stator winding which produces a
radial air gap flux density b = Bm cos 6 sin wt. The rotor, of axial
length L and diameter D, carries two ten-turn full pitch coils A1A2
and B1B2, the planes of which are at 60°. The coils are connected in
series and rotated at w rad/s.
r
Derive an expression for the total e.m.f. generated in the coils
at the instant when the plane of coil AIAz makes an angle of 6 = a
with the maximum flux density plane.

Using the derived expression, calculate the e.m.f. for the follow-
ing values: supply frequency = 50 Hz, t = 1 mis, a = 20°, L = 100 mm,
D = 50 mm, wr = 150 radls, Bm = 2T.

97
 Figure 4.10

Two kinds of e.m.f. are generated: (1) e R equals the rotational

e.m.f. due to the conductors cutting the flux and (2) e t equals the
transformer action e.m.f. due to alternating flux in the coils.

b B cos a sin wt
m

b = Bm cos (a + 60 0 )sin wt

The tangential velocity is

v = wr ~m/s
The number of conductors per coil = 20, so that by using the same
negative notation as in example 4.7

eR =- 20bLv volts per coil

D
20Lw
r -2 [Bm cos a sin wt
+ B cos (a + 60 0 )sin wt]
m

- 10LDwr[Bm sin wt(1.5 cos a - 0.866 sin a)]

(2) Now the flux through coil A1A2 is

cjl
A
=
J
-err-a)
D
Bm cos 8 sin wt¥8 L LDB
m
sin wt sin a

while the fl ux through coil B1B2 is

cjl = LD sin wt sin(a + 60 0 )
B
98
therefore

d
dt (10<p A + 10<P B) (10 turns)

10LDw[B cos wt(l.s sin a + 0.866 cos a)]

m

Hence the total e.m.f., e, generated is given by e R + e t volts,

that is

e = - 10LDwr[Bm sin wt(l.s cos a - 0.866 sin a)]

10LDw[B cos wt(l.s sin a + 0.866 cos a)]

m

For the numerical section

wt = 2rr x SO 1_
x __ = 0.314 rad 18°
10 3
therefore sin t~t = 0.309 and cos wt 0.951; as a 20° sin a = 0.342
and cos a = 0.94, therefore

10, 5
e = - 10 x --- x --- x 150 x 2 x 0.309(1.5 x 0.94 - 0.866 x 0.342)
10 3 10 2

10 5
- 10 x --- x --- x 314 x 2 x 0.951(1.5 x 0.342 + 0.866 x 0.94)
10 3 10 2

=- 5.16 - 39.62 =- 44.79 volts

Example 4.10

AIm square single-turn loop carrying a current i lies in the (x,y)

plane with its sides on the lines x = ± 0.5 m and y = ± 0.5 m. A
small five-turn search coil of effective area 0.05 m2 is placed in
the (x,y) plane at the origin. (a) Calculate the e.m.f. induced in
the search coil if i = 20 sin 1000t. (b) Calculate the r.m.s. value
of the search coil e.m.f. if i 20 sin 1000t + 5 sin 3000t.
[C.E.I. Part 2, E.F.N., 1971]
y

D C

x
A B

AB = BC = 1 metre
Figure 4.11

99
 From figure 4.11, the magnetic field intensity at the centre of
the square loop ABeD is given by

4i A t
72X1r
Il04i
thus flux density = ~ T

so that the flux ~ linking the search coil is given by

Il04i
~ = 72X1r x O. 05 Wb

induced e.m.f., e = - N~~ volts

d~
5dt volts

(a) ~ = ;~7x ~ x 0.~5 x 20 sin 1000t

d~ 16 cos 1000t
dt = 72 10 4

therefore

5 x 16
e = cos 1000t - 5.66 cos 1000t mV
12 x 10 4

(b) ~ ~ x ~ x 0.~5(20 sin 1000t + 5 sin 3000t)

10 7

d~ _ 0.566(2 x 104 cos 1000t + 1.5 x 10 4 cos 3000t)

dt - 107

e = (-5.66 cos 1000t - 4.25 cos 3000t) mV

therefore
1
e = (5.66 2 ; 4.25 212 mV
r.m.s. )

5 mV

Example 4.11

As the rotor is turned at angular speed wr radian per second, the

inductance of the exciting windings of the electromagnetic device in
figure 4.12 varies sinusoidally between a maximum of 2L and a
minimum of zero. A current i = I cos wt is maintained in the
m

100
winding. Show that a sustained non-zero mean torque can be developed
only if wr = w. Derive an expression for this torque in terms of the
position angle 0 of the rotor axis at the instant of an exciting
current zero. Derive also an expression for the voltage across the
terminals of the exciting winding, the resistance of which is
negligible.
[C.E.I. Part 2, E.F.N., 1969]

Figure 4.12

Inductance L(I + cos 28)

1 .2 dL
torque T "21 d8

} i 2 x [-2L sin 28]

Tinst I 2 cos 2 wtL sin 28

m

LI 2
~(sin 28 + sin 28 cos 2wt)

LI 2
~[sin 28 + } sin(2wt + 28) + } sin(28 - 2wt)]

Now from the diagram, 8 = wrt - 0, or wr d8/dt; also

I
r.m.s.
I 112. Therefore
m

Tinst LI2[sin(2w t 20) + } sin(2w r t - 20 + 2wt)

r

+ } sin(2w r t - 20 - 2wt)]

The average value of each of these sine functions is zero except when
w = w because then Isin(2w t - 20 - 2wt) becomes -! sin 20, when for
r r
a given torque and with 0 constant

101
 2rr
T =W(!LI2 sin 20 dt
average J;
2rr/w

2rr
sin 20[t] w
2rr/w a

The voltage across the winding terminals is given by

v = ~(Li)
dt
d
dt[L(l + cos 26)Imax cos wt]

= ~t {L[l + cos(2wrt - 20)I max cos wt]}

= ~[LI cos wt + LI cos wt cos(2wrt - 20)]

dt max max

wL1max sin wt wL1max sin wt cos(2wrt - 20)

wL1max sin wt[l + cos(2wrt 20)]

- LI cos wt 2w r sin(2w r t - 20)

For the condition when w = wand i

r
I 0 when 6 0

v = - w LI sin w t[l + cos(2w t - 20)]

r m r r

- 2w LI cos w t sin(2w t - 20)

r m r r

As 6 = wrt - 0 and 0 is constant, this voltage expression can be

expressed as

2w LI cos w t sin 26
r m r

Example 4.12

Find the displacement current in a conductor through which a conduction

current of 1 A is flowing. The frequency is 60 Hz and the conductor
102
is made of copper with a conductivity of S.S x 10 7 S/m while the
relative permeability and permittivity are both unity. The diameter
of the wire is 1 mm.

v x H

but E = Eo cos wt

so that

J oE oEo cos wt
c

aD d
and at= dt(e:oE) =- e:uwEo sin wt

the ratio of displacement current to conduction current is

aD
at 211 x 60 x 1 x 10- 17
Jc = -0- 9 A:\ 7
= -s-.-S;:"x"--1-O--=7'-"--3;:"6-1I-x-l-0-
1 A.
c
If the wire diameter is 1 mm, the cross-sectional area is
0.79 x 10- 6 m and the current density J = 1.27 x 10 6 A/m2, from which

PROBLEMS

1. Comment 011 the difference implied in the following alternative

statements of the Faraday law of electromagnetic induction
(a) the e.m.f. is equal to the rate of change of flux turns;
(b) the e.m.f. is equal to the turns times the rate of change of flux.

Figure 4.13 shows an electromagnetic transducer with a soft iron

armature pivoted between the poles of an electromagnet excited by a
direct current I in a coil of N turns. In the symmetrical position,
each of the four working air gaps has an effective length g; vibratory
motion of the stylus produces a small displacement 0 of the armature,
and an e.m.f. e is consequently induced in the n-turn pickup coil.
Prove that at any instant

e = (On:IA) (~~)
g

where A is the effective cross-sectional area of each gap. The

reluctances of the iron paths are negligible.
[C.E.I. Part 2, E.F.N., 1970]

103
 g - 0 g + 0

Vibratory
motion

Figure 4.13

2. Explain the generalised machine theory approach to the prediction

of performance of a practical electrical machine. Derive, in terms
of the usual quantities, the torque relationship for an electrical
machine having one stator winding and one rotor winding.

The self-inductances of the stator and rotor windings of an

electrical machine are 1.6 Hand (0.03 + 0.005 cos 8)H, respectively,
and the mutual inductance between them is (0.2 cos 8)H. The angle 8
measures the axis position of the rotor winding with respect to the
axis position of the stator winding. Calculate the maximum torque
of the machine when the stator current is 8 A and the current current
40 A.

3. A two-pole machine has a salient-pole rotor, with an exciting

winding carrying a direct current Ix rotating within a cylindrical
stator at an angular speed w rad/s. A single-phase winding on the
r
stator carries an alternating current i = 1m sin wt. The self-
inductance of the rotor winding is L11, and of the stator is
L22 = Lo - L2 sin 28, where 8 is the angular position of the rotor
m.m.f. axis relative to that of the stator winding. The mutual
inductance between stator and rotor windings is L12 cos 8.

Explaining the physical meaning of the parameters and of the steps

in the analysis, develop an expression for the torque. Show that the
torque averages zero except when w = w, and determine its value for
this condition. r
[C.E.I. Part E.E.C., 1978]

104
 [~ L2ImL cos 26 - i L12Ixlm cos 6;
8 = (wrt - 6~
4. A torque motor comprises cylindrical stator and rotor members;
each carries a two-pole single-phase winding. The respective winding
inductances are 3.0 and 0.9 H, and when the stator and rotor windings
are coaxial, their mutual inductance is 1.8 H. Resistances are
negligible. Stating any simplifying assumptions, develop an
expression for the mean torque on the rotor; evaluate this for a
stator current of 7.07 A (r.m.s.), with the rotor winding short-
circuited and an angle of 30° between the winding axes.
[C.E.I., Part 2, E.E.C., 1976]

[77.88 N m]

5. Figure 4.14 shows an elementary electrical machine in which the

ferromagnetic parts have infinite permeability. A winding (not shown)
on the inner surface of the stator establishes a radial gap flux
density B(8, t) = (B cos 8) cos wt. The rotor, of effective axial
m
length~ and diameter D, carries a single-turn coil PQ. Obtain an
expression for the e.m.f. induced in the coil when the rotor is
driven at a speed w rad/s and the position of coil side P is at
r
angle 8 =a when t = O.
What does the expression become (a) if wr .w, (b) if wr o and
the coil is fixed with coil side P at 8 = a?
[C.E.I. Part 2, E.F.N., 1969]

(b) ~DwBm sin a sin wt]

Rotor
Figure 4.14

6. The iron core of a d.c, solenoid coil (figure 4,1~ comprises two
parts: one fixed, the other a plunger of mass m movable under the
restraint of a spring of stiffness k. The position of the gap face
of the plunger can vary between x = 0 (open-gap) and x = c (closed-

105
gap). In the open-gap position the spring imposes zero restraint.
The inductance of the coil is given by

where a and b are constants and b>l/c.

(a) Find an expression for the maximum coil current and the
corresponding displacement x for which the plunger is at rest in
stable equilibrium with the gap partly closed.

(b) A constant current I, greater than that in (a), is maintained

in the coil. The plunger is suddenly released from the open position
x = O. Find (i) the velocity with which the plunger hits the fixed
core, and (ii) the energy supplied by or delivered to the source
during the motion, neglecting coil resistance.
[C.E.I. Part 2, E.F.N., 1971]

m:~
[I'-------=-....IW" K
I I
I l

Figure 4.15

[ca) r
max
(0 . 735K,
~J'
! x = b1 i (b) Ci) [a r2( e bcm- 1) _Kc2.]~

(ii) ar 2 (e bc - l)J
7. The inductance of a single-phase two-pole reluctance motor is
given by L = (0.05 + 0.03 cos 2S)H, where S is the rotor position
angle. When the motor is running at synchronous speed from a 50 volt
50 Hz, sinusoidal-voltage supply, its load angle for a certain load
is 45° and its power factor is 0.5 lagging. Derive an expression for
instantaneous torque of the motor using the basic relationship

Using the derived expression, calculate Tinst when the voltage has
reached its maximum value.
[-0.114 N m]
106
8. A copper disc of ISO mm diameter is located at the centre of a
long solenoid of length 1 m, having a diameter of 160 mm. The
solenoid is wound with 1200 turns. The disc is arranged to rotate
about an axis coincident with the axis of the solenoid at 1800 rev/min.
Brush take-offs are provided at the centre and at the edge of the disc.
If the current in the solenoid is 2A, what is the e.m.f. between the
brushes?

[1.6 mY]

9. A long straight non-magnetic cylindrical conductor of resistivity

p and radius r lies in a uniform magnetic field that alternates sinu-
soidally at a low angular frequency wand has an r.m.s. value B.
Derive expressions for the power loss per unit length of the conductor
when its axis lies (i) in the direction of B, (ii) at right-angles to
B.

[Note: f sin 2 8 cos 2 8 d8 = ~ (8 - 1 sin 48)]

[C.E.I. Part 2, E.F.N., 1980]

107
5 ELECTROMAGNETIC WAVES

In conjunction with some of the theory given at the start of chapter

4, this chapter is mainly devoted to a study of Maxwell's equations
and their application to various problems (excluding wave-guides - see
chapter 6).

From Arnp~re's work law, m.m.f. =~H dx = i, we can say

V x H = J (5.1)

which states that in a space where current is flowing the magnetic

field has curl. If the divergence of both sides is taken, then

v • Vx H =V • J =0 (5.2)

But for alternating currents, V • J = 0 as in the case of a capacitor

where no charge flows across the air space yet current flows,
signifying an apparent source.

From the equation of continuity which holds for the flow of

charges

V'J=-~
at
(5.3)

while from Gauss' law

div D = p (5.4)

a
or 3T(V' D} (5.5)

(5.6)

combining equations 5.3 and 5.6

(5.7)

Finally

(5.B)

where aD/at is called the displacement current. For Maxwell's first

lOB
equation then

(5.9)

The second of Maxwell's equations is obtained from Faraday's law

(equation 4.5)

_ (dB. ds (5.10)
Js dt

From Stokes' theorem

~A • dx =~ x A • ds (5.11)

then~E • dx =~ x E • ds (5.12)

therefore from equations 5.10 and 5.12

VXE=-~
at
(see equation 4.6). Since div curl A 0

V'VXE=O=-V.~ (5.13)
at
from which it can be concluded that
v • B o (5.14 )

The magnetic flux which issues from any closed surface is always zero.

Finally the Lorentz equation, from

and Fm q x v x B

so that the field vectors E and B can be expressed in the form

F = q(E + v x B) (5.15)

The following relationships will also occur, and are usually

obtained experimentally

D = EE, B ~H and J = aE (5.16)

where E, ~ and a are the permittivity, permeability and conductivity

of the medium with which the problem may be concerned.

109
 For conductors and dielectrics, the ratio a/wE is that of the
conduction current density to the displacement current density; it
can be used as the criterion for conductors and dielectrics (w rad/s
is the sinusoidal angular frequency). For good conductors

~»l at 30 000 MHz

WE

for example, for copper the value is 3.5 x 108 while for good
dielectrics

~«1
WE

for example, for mica the value is 2 x 10-~.

Example 5.1

With Maxwell's equations as a starting point, develop the wave

equation pertaining to a plane electromagnetic wave propagating in a
medium of conductivity a relative permittivity Er and relative
permeability ~ = 1. Discuss how a, E and the angular frequency W
r r
affect two of the following concepts: (a) the intrinsic impedance of
the medium to wave propagation, (b) the depth of penetration, ec) the
phase velocity.
[e.E.I. Part 2, E.F.N., 1968]

Maxwell's equations in differential form for a plane wave are

aH z aE
- ax- E--2::.
at
+ aE
y
(i)

aH aE
y - z
ax- E--
at
+ aE z (ii)

aE z _ aH y
ax - ~at
(iii)

and (iv)

from these equations

(v)

(vi)

110
therefore

If these quantities vary sinusoidal with time then the general form
is

or + jw(cr + jw£hiE
ax:"

where y2 = jw~(cr + jw£) and y is referred to as the propogation

coefficient for the medium and equal to (a + j6). The solution for
E then becomes
E -- E0 e jwt e -yx

in the positive x direction, therefore

E jwt -ax -j6x -ax j(wt - 6x)

= Eo e e e = EO e e

This represents a wave travelling in the positive x direction with a

velocity w/6 and which is being attenuated by a factor e- ax .

(a) For a plane wave in a pure dielectric of permittivity £0£ and

r
travelling in the x direction, then from equation iv above differ-
entiated with respect to distance

and from equation vi with cr 0

Equating these last two equations

111
This equation relates the space and time variations of the scalar
magnitude E of the electric field intensity and is called a wave
y
equation in E .
Y

If equations iv and vi are differentiated in reverse order, then

This has the same form as the equation for E .

Y

If a quantity v is introduced as being given by

1
lle:

a2E a2 E
then --y = v 2---1'-
at 2 ax 2
The dimensions of v are seen to be metres per second, that is, a
velocity. For free space (vacuum) v is approximately equal to
3 x 10 8 m/s.

The solution for the above second-order equations can be

E Eo sin(c'lt 8x)
y

and Hz = Ho sin(wt - sx)

both representing a wave travelling in the positive x direction.

To find how Hand E are related, recall

z y

aE aH
~ = - llat Z

Substituting for E into this relationship by the indicated differ-

y
entiation, and then integrating with respect to time yields

H
Z
=~
llW
Eo sin(wt - Sx)

Taking the ratio of E to H for a single travelling wave, as given

y Z
by the equations above, we obtain

112
For comparison purposes consider

E
y
= Eo sin(wt - ex) volts/m

Hz = ~(~) Eo sin(wt - ex) A/m

It is apparent that E and H are identical functions of x and t, but

y z
their magnitudes differ by a factor I(~/~) or its reciprocal.

The dimensions of the ratio EO/Ho are

volts/m volts
ohms
Aim amperes

Thus I(~/~) has the dimensions of impedance and it may be written as

where n is called the intrinsic impedance of the medium.

For free space (vacuum)

n no = -1(411 x 3611 X

" 10 7 1
If E and H are in time phase, n is a pure resistance.
All dielectric materials, however, have some conductivity a and
while it may be neglected in many cases, it is necessary to establish
a criterion for the cases when it cannot be ignored. From equation
vi a must be included. The mathematic procedure will be as before
and is found to yield

that is, the intrinsic impedance is now a complex quantity. Hence

the electric and magnetic fields are no longer in time phase.

(b) The depth of penetration 0 is that depth at which the wave

travelling in the conductor has been attenuated to lie of its
original value. The amplitude of the wave decreases by a factor of
e- ax , therefore when ax = 1 the amplitude is lie of its value at
x = O. Hence when x = 1, it means that by definition the depth of
penetration is 0 = l/a. In general

113
For a good conductor

_ (wllal ~
a. - l-z--)
z ) 21
thus 0 -
- ~wlla

(c) To develop the phase velocity, consider the propagation

coefficient

and let a/(we»>l for a good conductor; then

1 1
y (jwlla)2-= (wlla)2L45°

thus a. S = (w~a)~
where a. = attenuation coefficient, nepers per unit length, S phase
change coefficient, radians per unit length.

The phase velocity of a wave in a good conductor is

If a/(we)«l for a good dielectric, then we can use the binomial

theorem on the expression for y, that is

y= jwl(ll e ) [1 - jz~e + ~ '~er + . ' -]

Hence

( _j-2-1
Zwe)
Similarly

or, in many cases, where a = 0

Thus the effect of a small amount of loss in the dielectric is to add

a small correction factor to S.

114
 The phase velocity is given by

1
v = !!!.=
B
(lid ~(l ~)
+ 8w 2 £2

Example 5.2

A metal rod, of radius r, resistivity p and relative permeability II ,

r
is arranged coaxially within a long magnetising coil of N turns per
unit length carrying an r.m.s. current I of angular frequency w.
Show that the magnetic field strength H at radius x within the rod is
given by

where

Estimate the power dissipation per metre length in a rod of radius

SO mm within a coil of 40 turns/m carrying 20 A r.m.s. at 250 kHz.
The rod has a resistivity 0.05 n m and a relative permeability of 1.0.
Ignore end effects, and assume that the currents in the rod do not
appreciably affect the magnetic field strength due to the exciting
current.
[C.E.I. Part 2, E.F.N., 1975]

I
I
I
t--- 2 r--l
Figure 5.1

At radius x within the rod, let the field quantities induced by

the exciting coil be H ,E and J . (H in the axial direction since
x x x x
it is a cylindrical rod.) Due to time-varying field H , there will
x
be an induced electric field E which, since there is no physical
x
movement in the system, is given by

E x 2'1fx
X -I~
at ds =-
s

115
where \ is a dummy variable representing the radius.

J cyE
x x

thus J
x
jWCY
x
IB \ \
d\
o

or

Differentiate with respect to x

aJ
x----E. + J
ax x
= - jwcyxB
x
= - jW]JCYxH
x

But J = curl H, that is, J

x
-aH lax,
x
then

a 2H aH
x x
x--+ ax - jW]JCYxH
x
0
ax 2

or o

since n 2 =- jW]J/p lip) .

Now JH
6£ = enclosed current = EN! = H x 1 as H is essentially
constant and N turns given per unit length, therefore
x
E = - jWJJH!\ d\ = _ jwH]Jx
x x 2
o

As J
x
CYE x ' power density IJ x Ex 1 CY 1E 12
X

r
power dissipated per metre length = j(2nxpx dx
o

116
 = 122 W per metre length

Students should check that the assumptions are reasonable

II
r
o
aE dxl«NI if ~<l
w].Ir2
p

For the given figure

0.025 which is «1

Example 5.3

Derive expressions for the electric and magnetic fields within along
cylindrical isolated conductor carrying a sinusoidal current of
angular frequency w. Assume the field penetration to be small
compared to the radius of the conductor, and make what further
simplifying assumptions are appropriate. Show that for calculating
power loss the resistance per unit length may be taken as
1
(l/a) (!w].I/a) 2 , where a is the perimeter of the conductor, and].l and cr
the permeability and conductivity of its material.
[C.E.I. Part 2, E.F.N., 1974]

6zT
I Total
2b cur:ent
~I ,J"

Figure 5.2

Vectors J and E are axial while H is circumferential. Consider a

thin cylindrical shell of radius r and thickness or.

curved surface area = 2rrror

therefore

current enclosed = 2rrrJor =J"Ho~

2rr(r + or)(H + ~~)
dr
- 2rrrH = 2rrrJor

117
 dH d
H + rat:" = rJ dr CrH )

I d(rH)
J
r~

E + dE x cSr
I I
..
dr
T
18 H cSr
_-:-1__._---:-- -L
l-- I m-----J
Figure 5.3

From Faraday's law (figure 5.3)

E x I - (E + ~; cSr) x I = - jw~Hor x I

dE
dr = jw~H
also J = aE
This last equation ignores the displacement current for a good
conductor, hence

I dJ
H = jw~a dr

ld(r dJ}
or J = r 'drl) w~a drJ

.
thus JWjJaJ = -I (dJ
r r
-d + rd 2-J)
dr 2
I dJ
--+
r dr

A general solution for this last equation involves the use of

Bessel functions. However, the wording of the question states that
the field penetration is small, that is, J is only significant for
r ~ b so that the second-order equation in J above can be approx-
imated to

jw~aJ

The solution to this equation is of the form

118
where y2 = jw~a; also

[1 + j]2 w~a, y = (1 + j) (l
1
where (l (w~a/2)2, therefore
1 dJ J o yr
H 2- dr = ye
y
at r b

JO b I jwt
H= - eY 2rrb e
y

therefore

yI -yb jwt
JO = 2rrb e e

I y(r-b) jwt
thus H 2rrb e e

and E ~ eyer-b) e jwt

2rrba

since

yI y(r-b) jwt
J 2rrb e e

At r b

Ie jwt
H ~and E

as expected.

The Poynting vector p EH is radially acting inwards, or

Pr real part of EH*

EH*

Integrate this expression over a unit length of surface when the real
part of the integration is equal to the power loss due to heating,
that is, I2 R. Therefore

x 27Tb x 1

where 27Tb = a = perimeter. Therefore

119
 1
R =~
ao
=.!.a (WIl)
20
2: n

where R is the resistance per unit length.

Example 5.4

An annular-shaped resistor is made with inner radius r and outer

radius R from resistive sheet material of resistance p per square.
Find the resistance between the inner and outer circular edges.

Such a resistor is used to provide a matching termination for a

loss-free coaxial cable of radii rand R. The cable is insulated
with po1ythene (relative permittivity e r = 2.30). Find the necessary
value of p. Derive any formulae used.
[C.E.I. Part 2. E.F.N •• 1968]

Now current. I = 1 J ds.

resistance _ ~ x 1en~th dr
area Pds

dr
p 2TTr x 1

= Lldr
2TT r
r

L2 [log r]R
TT r

p
2TT log rR n
For a loss-free cable. the characteristic impedance Zo is given by

Zo =
fL)~ n
lc
the analogue of inductance and capacitance from the above expression
for resistance is
llO R
L = 2iT log r H

2TTeoer
and C F
10g(R/r)
2 R llO log (R/r)
thus ...£:. 10g2 2TT x 2TTeo x 2.3/10g (R/r)
4TT2 r

120
 ].10

1
10 9)2
p ( 4~ X 36~ X

10 7 X 1 X 2.3

248 n/square

Example 5.5

A semi-infinite slab of semiconductor material shown in figure 5.4,

has two identical hemispherical ohmic contacts of radius 1 mm
embedded in its surface and set a relatively large distance apart.
The resistance and capacitance between the contacts are, respectively,
1 nand 1 pF. Calculate the resistivity and relative permittivity of
the semiconductor material.
[C.E.I. Part 2, E.F.N., 1969]

Figure 5.4

Let the distance apart of the hemispheres be D m and let each be

of radius r. From the work in chapter 3, the capacitance between two
hemispheres is given by

Again using the analogue relationship. The resistance between the

hemispheres is

R £.
~
(.!. - ~)n
r D-r
As C

1 X E: r
~ x ----'--
367T X 10 9
---~-~~-- D»r
(10 3 - 0)

thus E: r 36

121
Similarly R 1 (l, therefore

1 x 1f
p 0.00314 (l m

Example 5.6

Explain what is meant by space charge. Discuss the effect of space

charge on the conduction in a planar diode.

Show (a) that the average velocity of an electron in a space-

charge-limited diode is one-third of the maximum; and (b) that in the
absence of space charge the average velocity is one-half of the
maximum. Assume in each case that the initial velocity of the
electron is zero.
[C.E.I. Part 2, E.F.N., 1972]

In a planar diode, electrons emitted by the cathode will be

attracted to the anode if this is at positive potential, say, Vo. If
Vo is large, all the emitted electrons will reach the anode, (the
actual number depends on the cathode temperature) and the resultant
diode current is said to be 'temperature saturated' with the potential
Vo linearly distributed between the anode and the cathode.
If Vo is a lower value than above, the slower moving electrons
form a negative space charge, particularly in the region of the
cathode and this results in a modification of the electric field
distribution and so controls the current flow. Assuming that the
emitted electrons have zero velocity, the electric field is adjusted
by the space charge until the limiting distribution is reached when
the field is zero at the cathode and increases to reach its maximum
value at the anode. This will give rise to the 'three-halves' power
law for the planar diode characteristic.

(a) Space-charge-limited case. Current density J will be constant,

or dJ/dx = 0; alternatively div J = 0; but div J + ap/at = 0,
therefore the charge distribution is constant with time. If the
electron velocity at any distance x from the cathode, is v mis, then
J = pv = constant. Now

e:dE J
div D =p or dx =p v
-e:

or

Consider the kinetic energy available equated to the actual potential

energy, then

eV or v

122
or

Suggest a solution of the form V Vo(x/D)n

2 Vo n- 2 k
-n/ 2
-d V = --n(n-l) (x~
- 1 x (~
dx2 D2 D (VaP
Equate indices to yield n - 2 = n/2 or n = 4/3.
This now satisfies dV/dx = V = 0 atx = 0 and gives ka:V a 3 / 2 , that is,
Ja:V o 3 / 2 . Therefore

v =
but v = dx/dt, therefore

tI a
v
max
_z..
D 3

where t is the transit time of the electron. Therefore

m

thus vmax 3 x average velocity

(b) No space charge

dV Vo
dx = 0

so V =
Va(~)
From the velocity equation

v = ce:o)!(~)! = vmai~)!
but v = dx/dt, therefore

123
 v t 1
~ = 2(0)2
o 2

20
or v 2 x average velocity
max t
m
Example 5.7

A dielectric material of relative permittivity £r and loss tangent

tan 0 is subjected to a uniform electric stress of r.m.s. value E
alternating at angular frequency w. Show that the specific dielectric
loss (power per unit volume) is given by p = E2£Q£ w tan 0
A concentric cable of inner and outer radii r =~0.01 m and
R = 0.02 m has an insulant of relative permittivity 3.0. The loss
tangent is affected by temperature gradient, and for a given service
condition is related to the radius x by tan 0 = [1 + (O.OOS/x)] x 10- 3 .
Estimate the dielectric loss for a 1 km length of cable supplied at
150 kV, 50 Hz.
[C.E.I. Part 2, E.F.N., 1978]

curl H dO
J + dt = (a .)E
+ ]W£

J and E are in phase and produce power loss per unit volume equal to
JE = aE 2
r.m.s.

also loss tangent, tan 0 = a/w£

so that the loss per unit volume due to the conduction current is
obtained by substituting for a, that is

For the numerical calculation, the insulant has a conductivity a

given by

If I is the current at radius x, then

J x 2nx x 1 = (a + jw£)E x 2nx = I

I
thus E
2nx(a + jw£)

potential V I
E dx '" -.--
]2nw£
1
R

r
-dxx

_1_ 10 ~ '" I 1 2
j2nw£ ~r j2nw£ oge

124
 R
power loss p per unit length = Io'E,22rrx dx
r

2rrV2
__ RI
~r~.~m~.~s~._ WE(l + O.005/x) x

(10ge2)2 r x

2.32 kW

Example 5.8

A long conductor, of conductivity a and carrying a current I of

angular frequency w, lies in a slot in a semi-infinite ferromagnetic
block of infinite permeability and zero conductivity (see figure 5.S).
Assuming the magnetic field strength, H, in the conductor to be y-
directed and the electric field strength, E, to be z-directed, apply
Maxwell's equations to determine expressions for: (a) E and H, and
(b) the impedance of the conductor per unit length when the depth h
is much greater than the depth of penetration 5 = 1(2/WlluO).
[C.E.I. Part 2, E.F.N., 1978]

z
Block

'--""----__t_ y

Conductor
Figure 5.5

Assume the material is a good conductor so that O»WEp or the

current is essentially one of conduction. In the ferromagnetic block
II = 00 so that" magnetic intensity H is zero, though the flux density B
has a finite non-zero value, as shown in figure 5.6.

125
 ].I = 00 a].l 0 EO
a = 0 E
z
Hy
y H 0
I
I I
I--b ---I
Figure 5.6

Using Maxwell's equations

ux uy uz

ajax ajay ajaz u [aE + jWEE ] ~ u aE

z z z z z

o H 0
Y

u u u
x y z

ajax ajay ajaz -jW].IoH u

Y y

o o E
z

therefore
aH aH
--L = 0, --L = aE
az ax z
aE aE
z __ z = jW].IoH
ay-= 0, ax y
Div B = 0, therefore aH jay = 0 and div D 0, therefore aEzjaz O.
y

Thus Hy' E vary only with x and time.

z

y = ± (1 + j )~(w~oa)

depth of penetration 15 = _/(_2_)

" w].Ioa

thus y ± (1 + j) assuming h»Q

15

126
The general solution is of the form

At the bottom of the slot H must be continuous (x = 0) then at

y
x = 0 H = 0 unless there is a current sheet at the bottom of the
y
slot. If there is no current sheet then H
y
=0 at x =0 so that the
coefficients A and B are equal. Therefore

Ez 2A cosh yx

H :yA sinh yx
y ]WllO

NOW~H d£ = current enclosed to the boundary of the conductor, then

:yAb sinh yh = I
]WllO
(a)
I sinh yx
thus H
y b sinh yh
jWlloI cosh yx
and E
z bY sinh yh
h
Note:~baEzdx = I. For iYhl »1
o jWlloI
Ez (on the surface)
~
The potential difference per unit length E x 1 V volts per unit
length, therefore z

v
(b) impedance, Z y=
bel + j)v{WllOa/2)

(1 + j) WllO
b To

1 + j
or Z bCi8 Q/m

Example 5.9

A sinusoidal current of angular frequency W flows in a conductor of

127
resistivity p and absolute permeability~. The conductor cross-
section is rectangular, of width much greater than its thickness 2b.
Show that the current density J at a distance y from the central plane
of the conductor (where J = J o) is given by

where n 2 -jw~ /p. Determine J in terms of J o , w, b and the material

constants, and hence derive an expression for the power loss per unit
surface area of the conductor.
[C.E.!. Part 2, E.F.N., 1979]

aH
~ay J ay
ay

aHx _
thus ay-- - J +~
p z

aE
and __ z ay jw~H ay
ay x

jp dJ
thus H +--
x w~ dy

therefore

Now y2 + n2 = 0, or

.wll 2w~
y2 = J - = (1 + j) -
p 2p

thus y = ± (1 + j) f~~ji = ± a (1 + j)

where a = wll/(2p). General solution will be of the form

E
B e -yy) e jwt
Z
J (A e YY +
p
when y 0, J o =A + B

H jyp (A e YY - B e-YY)ejwt
x Wll

128
But Hx(Y) =- Hx(-y) so that Hx(O) o and A B Jo/2. Therefore

E J~P (eYY + e-YY)ejwt

z
. J
H
JYP 0 YY _ e-YY)ejwt
x 2Wil(e

Reminder: Y = a(l + j) and Y* a(l - j) is the quadrature relation-

ship between E and H .
z x

a(l + j) J 02p2
(Sinh 2ay - j sin a y )
WI! 2

If Pr = real part of EH*

aJ02p2
pr =- """"2Wil (sinh 2ay + sin 2ay)

Power loss per unit surface area is

aJ02p2
P = 12p r (y = b) = ---(sinh 2ab
WI!
I + sin 2ab) W/m2

Example 5.10

A concentric cable of inner radius r and outer radius R has an

insulant of relative permittivity E. Derive an expression for the
r
capacitances C of the cable per unit length.

Evaluate C for r 0.005 m, R = 0.020 m and Er = 4.0.

Carefully draw a flux/equipotential curvilinear square plot for the
region of the insulant and use it to obtain a comparable numerical
value for C.

If there are n radial lines in such a plot, what should be the

ratio r2/rl of the radii of successive equipotential lines forming
the outer and inner boundaries of any curvilinear square?
[C.E.I. Part 2, E.F.N., 1979]

Let q coulombs be the charge on the inner conductor per metre

length of cable. Then flux density at any radius x within the
insulant is

D ....5L C/m 2
21TX

129
Corresponding electric intensity is

E = ~2 vol ts/m
7TX€

potential difference, V
-] R
-L-2
7TX€
dx vol ts

= -L log ~
27T€ r

.
t hus capac1tance C= ~
V = log(R!r)
27T€ F! m

For the numerical calculation

27T x 1 x 4
C
36 x 7T x 10 9 x log 4

160 pF

The curvilinear square plot is shown in figure 5.7. Tubes of flux

f = 32; potential steps = 7.

~+-~+-~__+-~Potential V

Figure 5.7

<: x f 4 x _32
capacitance
7

162 pF compared with 160 pF

130
 For the marked curvilinear square for n radial lines

rr(rl + r2)
n r2 - rl

rl(rr + n) rz(n - rr)

r2 n + rr
rl n - IT

PROBLEMS

1. Starting from the Maxwell equations, obtain the wave equation for
a plane electromagnetic wave in free space. Explain briefly what is
meant by the Poynting vector, and show for the plane wave that it
leads to a result that could be deduced directly from energy
considerations.
[C.E.I. Part 2, E.F.N., 1975]

2. From the Maxwell equations derive expressions for the intrinsic

impedance and propagation coefficient appropriate to the propagation
of electromagnetic waves of angular frequency w in a medium of
conductivity 0, permittivity £ and permeability~. Discuss the
characteristic differences in propagation in good conductors and good
insulators.

From the intrinsic impedances and attenuation coefficients,

compare over the frequency range f from 10 kHz to 10 000 kHz the
relative effectiveness of copper and permalloy sheets, 1 mm thick, in
shielding an enclosure from incident electromagnetic waves. The
relevant data are: copper: 0 = 58 MS/m, £ = I, ~ = 1; permalloy:
r r
o = 1.2 MS/m, £
r
= I, ~
r
= 104/f(with f in kHz).
[C.E.I. Part 2, E.F.N., 1977]
[Cu: 8.6 x 10- 8 , 1.8 x 10- 26 ; Permalloy 0.086]

3. An electromagnetic wave with an electric vector E = 200 mV/m in

free space falls normally on to the flat surface of a loss-free solid
dielectric material of relative permittivity 4. Proving any formulae
used, find the magnitudes of the electric and magnetic field vectors
in the incident, reflected and transmitted waves, the corresponding
rates of energy flow and the velocity of propagation in the dielectric
material.
[E = 200, -66.7, 133.3 mV/m; H = 0.53, 0.177, 0.707 mAIm;
energy flow 106, -11.8, 94.2 ~W/m2 velocity 1.5 x 10 8 m/s]

4. The shape shown in figure 5.8 is cut from a uniform sheet of

conducting material. Each marked square has a conductance of 1 unit
between opposite sides. Edge AB is maintained at a potential of 100
units; edge CD is maintained at zero potential. The 75-unit equi-

lTI
potential line is shown dashed. Estimate, by any suitable method,
the potentials at E, F and G, and estimate the conductance of the
sheet between AB and CD. Sketch in roughly six flow lines.
[C.E.1. Part 2, E.F.N. Specimen Paper, 1966] [66, 52, 50; 1.22]
A B

- -- ..........
'\
G F E

C D

Figure 5.8

5. A wire of circular cross-section is long and straight, and the

return path is remote. It carries an alternating current of angular
frequency w, and cr, p and E are its conductivity absolute
permeability and absolute permittivity. At the operating frequency
O»WE.
Derive the following expression for the current density (comp1exor
J) in the conductor at any radius r

Obtain this expression by applying Maxwell's equations. The

following is the Laplace operator on a vector W in cylindrical
co-ordinates r, e, x, where there is no variation with x or e

[C.E.I. Part ,E.F.N. Specimen Paper, 1966]

6. State Maxwell's electromagnetic field equations in differential

form. Use them to obtain the solution for a plane electromagnetic
wave of angular frequency W rad/s in an infinite medium of
permeability p, permittivity E and conductivity cr.

Derive expressions for the attenuation coefficient a and the phase

change coefficient S. Show that, if the medium is a good conductor,
the electric and magnetic field vectors differ in time phase by 45°.

Define skin depth and briefly discuss how it affects the electro-
magnetic wave shielding property of a conducting sheet.
[C.E.I. Part 2, E.F.N., 1968]

132
7. (a) State Maxwell's equations in differential form. With these
as a starting point, develop the wave equation for a plane electro-
magnetic wave propagating in a medium of conductivity 0 permittivity
E and permeability ~.

(b) A single-frequency plane wave is propagating in a medium for

which 0 = 10.0 S/m, E = 1000Eo and ~ = ~o. For what frequency will
the conduction current and displacement current densities be equal?
[C.E.I. Part 2, E.F.N., 1976] [0 = WE; f = 180 MHz]

8. State how Maxwell's electromagnetic field equations are related

to the basic experimental laws of electromagnetism.

Evaluate the conduction and displacement current densities J C and

J D, set up in a uniform isotropic material of conductivity 0 and
absolute permittivity E = EOE r by an electromagnetic field of electric
field intensity E.

A material of conductivity 10 S/m and relative permittivity 1000

(both independent of frequency) lies in a electromagnetic field of
frequency f. Calculate the ratio JD/J C for frequencies f of 100Hz and
1000 MHz. Hence indicate the frequency ranges over which the
material can be considered to behave (a) as a conductor, (b) as a
leaky dielectric.
[C.E.I. Part 2, E.F.N., 1973] WE ~ x 10-6. SO.
9 '9'
(a) 0 - 18 MHz, (b) 18 x 10 9 -18 x 10 8 Hz]

9. State Maxwell's equations and briefly explain their physical

significance.

Two long and perfectly conducting strips in air form a

transmission line as in figure 5.9. Disregarding fringing fields at
the strip edges, write down the Maxwell equations for the region
between the strips, and use them to show that

ay
3V
=- ~o(s/b)(31/3t)

31
and 3y =- Eo(b/s) (3V/3t)

(C.E.I. Part 2, E.F.N., 1979]

Figure 5.9

133
10. Define the Poynting vector and briefly discuss the field concept
of energy flow.

A long transmission line is formed from two parallel conducting

strips in air, of width b much greater than the spacing d. Determine
the power transmitted along the line for a given applied voltage V,
of frequency such that d is small compared with the free space wave-
length.
[C.E.I. Part 2, E.F.N., 1974]

11. A plane electromagnetic wave of frequency 10 kHz passes normally

into a large flat slab of brass (conductivity 15 MS/m) coated with a
0.01 mm layer of silver (62 MS/m). Calculate the surface impedance
of the silver. Describe the effects of increasing the frequency (i)
on the wave propagation through the silver and brass, (ii) on the
surface impedance of the silver.

[C.E.I. Part 2, E.F.N., 1980]

134
6 PRINCIPLES OF GUIDED WAVES

It is beyond the scope of this type of textbook to go into the

extended principles of wave-guide theory. However, the basic theory
can be stated so that readers should be able to follow and perform
the derivations and computations of the problems.

TE-type waves: electric vector perpendicular to the propagation

direction. These are transverse electric waves. If z is the
direction of propagation, Ez = O.

TM-type waves: magnetic vector perpendicular to propagation

direction. These are transverse magnetic waves. If z is the
direction of propagation, H = O.
z
TEM-type waves: electric and magnetic vectors are perpendicular to
the direction of propagation. These are transverse electromagnetic
waves. If z is the direction of propagation, both E = 0 and H = O.
z z
The behaviour of waves between guides can be predicted using
Maxwell's equations since the wave-guide surfaces are boundaries at
which the boundary conditions for both electric and magnetic fields
can be applied and so the arbitrary constants in the differential
equations can be evaluated.

The figure configurations in the wave-guides have characteristic

quantities that will be integers such as m and n which can take on
all integral values from zero to infinity. Only certain combinations
of m and n are allowed, thus giving rise to limited modes of
propagation down the guide. Those modes that can exist do so by
virtue of the dimensions of the guide and the frequency at which the
guide is excited.
Transverse Electromagnetic Waves
Since there exists a lower cut-off frequency for every mode below
which propagation cannot take place, the wave-guide is found to
behave like a high-pass filter.

Assuming propagation in the z direction, together with plane waves

varying sinusoidally with time and with the space enclosed by two
conducting planes which are infinite in extent and distance a apart.
E is uniform in the x-y plane, or
x

aE aE
-axx -_ -ayx = 0 everywhere
With these assumptions
135
 1J2 E + W211Q£Q£ E 0
r
a2Ex
or - - + w211Q£Q£ E 0 (6.1)
dX 2
r x

the solution of which is

E EIe -jl3z +
E jl3z
2e (6.2)
x

where 13 = W/(IlQ£O£ ) and EI and E2 are constants having dimensions

r
of volts per metre. Note that the electric field is formed from two
waves travelling in the positive and negative z directions respect-
ively with a phase change coefficient 13. If the source of these
waves happens to be in the region Z<O then in the absence of
Doundaries in z>O, there is only one wave EI e- j13z propagating
parallel to the z-axis. In free space £r = 1 so that 13 = W/(llQ£O)'

phase velocity v (6.3)

p

where c = 3 x 10 8 mls is the velocity of light.

~ __ 21TC c
wavelength A = 1.0 = 13 w (6.4)
I
where f is the frequency. In a dielectric,s w/(llO£O£ ), then
r
c
v
p
7£ and A
r

Thus in a dielectric the phase velocity and the wavelength are both
reduced from the free space value
. by 1£ r .

Associated with E is a magnetic field, that is

or H (6.5)
Y

There are no other components of H. Since there are no axial (z axis)

field components, this wave is a TEM wave.

For both waves, E and H are in the same time phase, but are in
x y
space quadrature. Because E and H are in time phase in the x-y
x y
plane (that is, the plane transverse to the direction of propagation),
it is called a plane wave.

136
Transverse Electric Waves

Setting Ez = 0 and taking into account the sinusoidal time variation,

it can be shown that for waves between perfectly conducting planes

E . (m1TX~
E1 Sl.n - - e (jwt - yz) (6.6)
y a

where y is the propagation coefficient of the waves.

H ~E
jW)Ja cos ~m:j e (jwt - yz) (6.7)
z 1

while

fix ....L.. E sin (m:je(jwt - yz) (6.8)

jW)J 1

where a is the separation between the plates and the relationship

between m and a is obtained from k = mrr/a, m = 1, 2, 3, etc.

Transverse Magnetic Waves

This time set Hz = 0 so that for a sinusoidal time variation signal

H Ez cos (m~j e (jwt - yz) (6.9)

Y

E
x ~2 cos (-m~Je(jwt - yz) (6.10)
Jw£

E
z
jmrr E
£wa 2
sin (-m:j Jjwt - yz) (6.11)

for m 0, 1, 2, 3, the number of field variations in the x direction.

For both these latter wave types

(6.12)

Above a critical frequency where the negative term prevails, the

radical will be imaginary, signifying a = 0 and

8 = I,w2 )J£
L - (mrr)2J
~ 1
2 (propagation) (6.13)

At the frequency below the one at which the two terms under the
radical are equal and determined by

4rr2f 2 = ~ (mrr~2 (6.14)

c )J£ ~a )

the radical is real, 8 o (no phase change) and only attenuation

137
takes place. Of course, no phase change means no wave motion,
therefore no propagation. Thus the cut-off frequency for the guide
is

m mv
f
c 1(1l£)2a 2a (6.15)

where v is the velocity of light in the medium. Also

A
g
=~
S
and v
f
= Ag f
where Ag is the wavelength in the guide and v f is the phase velocity.
Therefore

2lT
A 1 (6.16)
g
[ w2 1l£ - (mlT/a)2]2

where AC = 2a/m. Also

v c A
f=Il£=7£ (6.17)
r r

If w21l£»(mlT/a)2, v f approaches vp 1/1(11£), and if the space

between the conducting planes is air then v = c = 3 x lOB m/s.
p

~
-'1-
z ~ a--i

Figure 6.1

The main emphasis is on the solution of problems and hence

starting with Maxwell's equations and assuming a sinusoidal
excitation in time, then for the rectangular wave-guide of figure 6.1
with a>b, a solution of the wave equations must be found, which
satisfies the boundary conditions of the conducting walls, with
transmission in the z direction. Note that

(6.18)

138
 y aH aE
z jw£ z (6.19)
H -- +
x y2 + w2jJ£ ax y2 + w2jJ£ Ty

y aH aE
z jw£ z
H
Y
ay - ax (6.20)
y2 + w2jJ£ y2 + w2jJ£
aE aH
y z jWjJ z
E
x y2 + 2
w jJ£
ax - y2 + w2jJ£
Ty (6.21)

y aE aH
z jWjJ z
E
y y2 + w2jJ£
-- +
ay y2 + wL]J£ ax (6.22)

For the TE modes, E

y
0 at x = 0 and x a and E
x
o at y 0
and y = b.

HjS m1T tn~Yjej (wt - Sz)

H
x y2
- sin
a a
tm1TX~ cos (6.23)
+ wL]J£

HjS n1T . ~n1T~ - Sz)

H
b cos (m1Tx)
- - Sln - - e j(wt (6.24)
Y y2 + w2jJ£
a b

H H cos (m1Tx) - - e j (wt - Sz)

- - cos (n1TY~ (6.25)
z a b

H(jwjJ) n1T - Sz)

E
x b cos (m:x) sin (n~y)ej (wt (6.26)
y2 + w2jJ£

(-jwjJ)H -m1T Sln--

. t1TX) cos (n1T1 Sz)
E
y b e j (wt - (6.27)
y2 + wLjJ£ a a

These are the propagation equations obtained by setting y = js

(above cut-off, y must be imaginary for propagation); yL + w2jJ£

r
is also found to be equal to (mrr/a)2 + (n1T/b)2 so that

y [(~1Tr + (~rrr - w2jJ£ (6.28)

so S [w2 _ (:rrr b
(n1TrJ l for air (6.29)
c2

f
c
1
2rrl(jJ£) [(:~r (b1T + rJ l
(6.30)

A
c rc c = ' 2
[(m/a)2
2
+ (n/b)2]
1
(6.31)

while the wavelength in the guide is

139
 A
g (6.32)

where E is the relative permittivity of the dielectric filling the

r
interior of the guide, while A is the free space wavelength.

Similarly the phase velocity is

2~f w
v =--= 1 (6.33)
f a {w21lE _ [(m~/a)2 + (n~/b)2]}
For TM modes, the propagation equations are

H (6.34)
x

EjWE m~ cos
C-m~x)
a-
.
SIn
(n~Yj
-- e
b
j (wt - az) (6.35)
y2 + W 21lE a

E _ .....J!..;:·E:;:a__ m~ cos [m~x:J sin (n~Y"ej (wt - az) (6.36)

x y2 + W21lE a l- J \"")
E jEa ~~ sin (m:x1cos (n~Ylej (wt - az) (6.37)
Y y2 + W21lE \"") \-)

E E sin (m:x) sin (n~~ej (wt - az) (6.38)

z

Similar equations for the propagation coefficient, for the phase

coefficient, for the cut-off frequency and for the various wave-
lengths as with the TE modes.
It will be noted that m is an indication of the number of field
varlations in the x direction while n indicates the number of field
variations in the y direction, so that TMll means m = I, n = 1 while
TElO means m = I, n = O.
Several other useful relationships include

A = 2a tan
g
e (6.39)

where e is the angle that the component wave fronts make with the
walls of the guide - see figure 6.2.

Also v
g
=v cos e

from the vector diagram in figure 6.2, where v is the velocity of the
component waves and v is the group velocity
g

v
f
= the phase velocity = ~e
cos
(6.40)

140
 / Direction of
/ propagation
vf I •
I

Figure 6.2
so that

c Z for air (6.41)

Example 6.1

(a) Assume a parallel plane wave-guide consisting of two sheets of

good conductor separated by 0.1 m and operated in the TEl mode. Find
the propagation coefficient y at frequencies of 100 MHz, 1000 MHz and
10 GHz. Does propagation take place?

(b) Repeat the question if the guide is excited in the TE z mode.

(a) m =1
1

y
[(iJ 4112 f ZI1 £J2

11 [~r - ~~ ZJ!
(i) at 100 MHz
1

y 11 [(o\y _ (200 x 106)Zr

3 x 10 8
1
= 11(100 - 4/9)2 31.35 nepers/m

(ii) at 1000 MHz

y = 11 ~OO _ (20 x 10 8 )ZJ!

3 x 10 8
1
11(100 - 44.44)2 23.42 nepers/m

Again as in the case (i) y is real, so there is no propagation.

(iii) at 10 GHz

y 11 ~OO e~Orr 1
11(100 4444)~ = j207 rad/m

141
Here y is imaginary, so that 8 207 rad/m and propagation takes
place.

(b) m = 2

(i) y = 211 ~OO 62.73 nepers/m

here y is real, therefore there is no propagation.

(ii) y 211 ~OO

[
- ( 10 9~2J!
3 x lOS-;
= 59.24 nepers/m

This is still below cut-off, therefore there is no propagation.

(iii) y = 21! ~OO _( 10 10 ~2Jl = j200

L 3 x lOS)

Propagation does take place since 8 200 rad/m

Example 6.2

A wave-guide consists of two extensive and perfectly conducting flat

sheets set parallel and 0.1 m apart in air. For how many TEn modes
is unattenuated propagation possible at a frequency of 5 GHz, and
what are the velocities of propagation in these modes? Justify any
expressions used.
[C.E.I. Part 2, E.F.N., 1973]

Figure 6.3 shows the general arrangements.

=a
xlL:
x

E = 0
z
o y
x =0
Figure 6.3

Assume that the fields are independent of y but vary with time t and
z, in the form e (jwt - yz)
Consider E has an Ex component only. Maxwell's curl equations are
u u u
x y z

a/ax a/ay a/az

E
x
o o
142
therefore Hx 0, Hy +y/(jwp)E x and Hz 0, then

u u u
x y z

a/ax a/3y a/az

o H 0
Y

O.

For TEM mode with Ex' Hy only, y2

therefore

s = wl(po£o) and ~
S
= velocity

Consider E has an E component only, then

y

u u u
x y z

a/ax a/ay a/az

o E 0
Y

yE , H - aE lax. Hence
Y Y y

u u u
x y z

a/ax a/ay a/az

H 0 H
x z

therefore

aH aH aH
z x
0 + - - - _z = jw£oE
Ty= az ax y

aH
and -2. = 0
ay

a 2E aH
z
Now --L = jwpo - - = + jwpo(aH + jw£oE )
ax x y
ax 2

+ (jw) 2pO £oE y2E

Y Y
a 2E
thus --L + K2E 0
y
ax 2

143
 E = A cos Kx + B sin Kx
y

Boundary conditions are that E .

y
=0 at x =0 and x a therefore A o
and Ka = nw for n ~ O. For non-attenuating mode y j 13, therefore

13 2 = W21l0 £0 - K2 = w21l0EO _ &wj 2

For 13 to be real, w2/c 2>(nw/a)2, that is, w/c>nw/a, thus

2af
n<
c

where c is the velocity of propagation.

For f = 5 GHz, a = 0.1 m and c = 3 x 10 8 m/s

2 x 0.1 x 5 x 10 9 10
n< 3 x 10 8 <;3

This means that the only propagating modes are TEM, TEl, TE2 and TE3.

The phase velocities are given by v

p
= w/l3, or

w c
V T \
P [w 2 /c 2 - (nw/a)2f2 [1 - (nc/2fa)2]2

c 3 x 108
2fa = 0.3
2 x 5 x 109 x 0.1

TEl v 3.14 x 108 m/s

p

TE2 v 3.75 x lOt! m/s

p

TE3 v = 6.88 x 108 m/s

p

Example 6.3

Consider a guide consisting of a pair of highly conducting plates.

For a plane parallel wave moving in the z direction and having a
frequency of 100 MHz, determine the power absorbed fer square metre
in the plates. Take E = 1 volt/m and a = 5.8 x 10 S/m.
x

For this problem, the wave must have two components, E and H
x y
together with a component E (very small) due to the currents in the
walls of the guide that arezinduced by H .
Y

Now in air E /H = n = 377 ~, so that

x y
144
 1
H
y 377 = 2.65 mAIm

For the metal

1

11
m ~~r ;45°
1

en x 10 x 4n x 10-7j2~
8
~5°
5.8 x 10 7

3.69 )( 10- 3 k rI./m2

Therefore

2.65 3.69
E
z
11 H
my A5°
103 103

9.78 x 10-6 ~

power P }z 1:/1 = 18.33 x 10-9 W/rrf

m

or P

9.78 x 2.65 x 1.41 14 = 18.33 x 10-9 W/m2

106 10 3

Example 6.4

Derive an expression for the power transmitted between two parallel

conducting planes in the TEl mode.

This problem requires the use of the Poynting vector. For

propagation y jS. otherwise if y is real. only attenuation occurs.

Taking the real parts of equations 6.6 and 6.8

Ey E1 cos CUlt - Sz) sin (m:x) and m = 1

SEl
H - - cos CUlt - Sz) sin (m:x)
x Ul]J

therefore

P = Ey x H
x

145
therefore

dx = 1__
a

o
SE
Wll
2
1- cos 2 (wt - sz) sin 2 ~ dx
a

SE12a
or p -- cos 2 (wt - sz)
z 2Wll

and P averaged over one cycle is

z
SE12a
P --W/m
av 4wll

Example 6.5

Derive an expression for the attenuation of TE waves in a wave-guide

consisting of two parallel conducting plates separated by a distance
a.

It is known that the magnitude of current density is equal to the

tangential component of the field intensity at the surface of both
plates, that is, H at x = 0 and x = a. Thus IJ I = IH I. From
z x z
equation 6.7, the magnitude of Hz is
m1TE l
H
z jWlla

m1TE l
or JH z I IJ x I
wlla

The power loss per plate is lJ 2R where R is the resistive

x r r
component of the wave impedance, Z = (1 + j)/(wll/2a). Therefore
w

so that

! ----

The power transmitted per unit area along the z direction is

where H* is the complex conjugate of Hand E and H are the real

y x

146
parts of the appropriate field components. Now

E
y El cos' (wt - Sz) sin (n:X)
SEl
H cos (wt Sz) sin (m:x)
x Wll

E 2S
power 1 (m1TX]
unit area 2wll sin 2 "a)dx dy

so that the total power transmitted for I m width of wave-guide is

So that the attenuation is obtained from

or

nepers/m

Example 6.6

Show that the magnetic field strength in a flat sheet of large

surface area is given by

if the applied field is uniform and parallel to the surface.

Show that the eddy current in such a sheet will reduce the
alternating magnetic field at the centre so that the ratio of the
magnitudes of surface and central field is given by

\
HH S[2 = ~ (COSh ~a + cos ~a)
c
1
where 2a is the thickness of the sheet, 0 (2/allw)2 the depth of

147
penetration and W is the angular frequency of the magnetic field
(e jwt ).

Figure 6.4

From Maxwell's first two equations

'V x H = J = oE
'V x E = _ ~ = aH
at - Ilar
and
aH
'V x 'V x H = o'V x E =- ollar

From figure 6.4 the magnetic field operates in the y direction,

therefore

. t
which for an exponential time dependent signal e Jw reduces to

d 2H
- - = jWlloH
dy2

The solution to this type of equation which has exponential time

dependence is

H = A eel + j)y/6 + B e-(l + j)y/6

where
1

(ir~llor
Let Hs be the surface value of H, that is, at y D, Y 2a, therefore

H A + B
s

or H A eel + j)2s + B e-(l + j)2s

s

148
letting 5 = a/c. Hence

A eel + j)s (e(l + j) 5 e-(l + j) 5) B e-(l + j)s

(e (1 + j) 5 e-(l + j) ~

B A eel + j)2s H A(l + eel + j)2S)

5

H
5
thus A
1 + e (1 + j) 25

Hs ( (1 + j)y/c (1 + j)2s (1 + j)y/cl

H (1 + j)2s ~e + e e - )
1 + e

When y = a

H
H5 ( (1 + j) 5 (1 + j) 51 5
(1 + j)2s te + e J cosh (s + js)
1 + e

The ratio of H /Ho becomes cosh (s + js), or

s

cosh (5 + js) = cosh 5 cos s + j sinh s sin 5

therefore

cosh 2 S cos 2 5 + sinh2 s sin 2 5 !(cosh 2s + cos 25)

Example 6.7

A wave-guide with rectangular cross-section has dimensions 20 rnm high

by 40 rnm wide. Assuming TE operation, good conductivity of the walls,
and the dielectric to be air, calculate (a) the mode, (b) the guide
wavelength, (c) the phase velocity, and (d) the group velocity, if
the guide is operating at a frequency 20 per cent higher than the
cut-off level.

From equation 6.31, for the TE 10 mode

2
(a) A
c
= 2a = 0.08 m = 80 rnm
[(1/a)2]!

3 x 10 8 3 x 10 8
thus f 3750 MHz
c A 0.08
c

20 per cent above this value is 4500 MHz, therefore

149
 6/6.7 mm

For the TEol mode

2
A 2b 0.04 m
c

f 7500 MHz, so the performance must be restricted to TE IO mode.

c

(b) From equation 6.32

A
g

6.67
120 mm

(c) from equation 6,39

A = 2a tan
g
e
12
or tan e = -S = 1.5

cos e 0,555
3 x,108
thus v f = 0.555 = 5,41 x 10 8 mls

(d) From equation 6.40

v = v cos e = 3 x 10 8 x 0,555 1,67 x 10 8 mls

g

Example 6.8

The propagation mode in an air-filled rectangular wave-guide (figure

6.5) is such that the amplitudes are

E o H A e- Yz cos 11
bY
z z

Show that for propagation to be possible without attenuation, the

angular frequency must be w>wo, where Wo 2 =(1I/b)2/~oEo. Derive the
other components of E and H, and sketch the field pattern in the
guide.

150
 y

Figure 6.S

Assume time variations to be of the form e jwt . Maxwell's curl

equations give

u u u u u u
x y z x y Z

a/ax a/ay a/az - jW]JoH a/ax a/ay a/az

E
x
E o H
x
H
Y
H
z
Y

Now assume all z variations to be of the form e- Yz so that

E B e- Yz , E C e- Yz , H D e- Yz , H = F e- Yz . Then
x y x y

yC jW]JoD (i)

yB jW]Ju F (ii)

ac aB 1Ty
-ax- ay =- jW]JoA cos (iii)
b

Similarly

1TA . 1Ty
b Slll b + yF (iv)

thus - yD = jw£uC (v)

(vi)

From equations (i) and (v)

C - -Y-
D Y jw£o

or C = D= 0

From equations (ii) and (iv)

lSI
thus F rnA sin(1Iy/b)
b(y2 + w21l0E:o)

1IjwIl oA sin(1Iy/b)
B
b(y2 + W21l0E:o)

If C and D F 0, that is, y2 + W21l0E:O = 0 then F = B = 00, therefore

C = D = O. Thus equation (vi) is satisfied. From equation (iii)

- - - - - - - = jWlloA
b 2 (y2 + W21l0E:O)

For propagation without attenuation y j6, 6 real, that is

(vii)

or 62>0 so that

2 112 1 2
W >- X -- = Wo
b2 lloE:o
jWllobA . 1Iy -j6z
thus E = _ _ _ S1n b e
x 11

These satisfy the boundary conditions at y 0 and y b.

x
E fi eld
in x y plane

o b y
(a)

Figure 6.6(a)

152
 z
A
g

[cc?~ .-
]
H field
in xz plane
Az

[®]
2

y
(b)

Figure 6.6(b)
The waveforms are shown in figure 6.6a and b.

Example 6.9

A rectangular guide has dimensions a = 30 mm and b = 20 mm through

which 5 kW is transmitted at 7.5 GHz. Calculate the characteristic
impedance and the maximum voltage across the 20 mm dimension of the
guide.

Now for a guide in the TE 10 mode

Wll 2b
l3a
211 x 7.5 x 10 9 X 411
__ x 4
503 rl
157
10 7 3"

Since

S [ 4112 x 7.5
2x 10 18 _ li)2]!
9 x 10 16

157 rad/m

4WllWT _ 4 x 211 x 7.5 x 10 9 x 411

x 5000
SCab) -
157 x 0.03 x 0.02 x 10 7

1. 257 x 1010

1.12 x 10 5 volts/m

153
 1.12 x 10 5 x 0.02 2242 volts

Example 6.10

Derive an equation for the propagation of power in a wave-guide and

for the electric field intensity across the b dimension of the guide
in the TElo mode.

The Poynting vector P = E x H W/m 2 .

(a) For the dominant mode, it is only necessary to consider E and H ,
that is y x

P - EH
z Y x

but it is also known that

E
p
z -(if}/
x

where E /H n, so that
Y x
P nH 2
z x

11"X
and H H sin - a sin (~t _ 8z)ej(~t -6z)
x m

El8
H
= jWlJo peak value of the field intensity
m

Then P nH 2 sin2 2!!. sin 2 (wt - 8z)

z m a

For one period, the average of the Poynting vector is

1
211"
W
z ~1r
nH 2 sin 2 1rX sin 2 (wt - 8z)d(wt)
m
0
a

!nHm2 sin2(1r~)

The total power W

T
=fwz ds over the guide cross-section
dx dy

-1 nH 2ab
4 m

154
 (b) The peak electric field is found from

E nH
y m

But H
m
2(~)!
n ab

thus E
Y

E zab
1 y
or 4-n-

Example 6.11

(a) Find an expression for attenuation due to conductor losses

between two planes, when excited in the TMI mode. (b) Show that the
frequency of minimum attenuation is at 13f .
c
power absorbed
(a) Attenuation coefficient a
2 x power transmitted

In the TMl mode

P = IJ zR
z
= 2IIH YIZR = 2IE ZZ (W\lJ!
\20; per plate
abs 2

H = E2 s cos Z m x
2
Ptrans = 21 Real part of E x H = 2E
1
x Y 2WE
7l
a
(from equations 6.9 and 6.10). Total power transmitted is obtained
from

E Z 2 Sa b
total Ptrans ~

while
EZZ 1

total Pabs = (W2\l0~

2 x --2- ~ ~ 2

for both plates, therefore

1
EZZ(W\l/20)2 !
(w\ll 20)
a = abS/2wE
2Ez z Sab/(4w£)
1 1 1
But S (WZ\lE - W Z\lE) 2 (\lE) 2 (w Z W 2)2
C c
ISS
Substituting in the equation for a
1
(Wll/2cr) 22wE

(b) Find da/dw and equate to zero to find the minimum value.

da
dw = K [(W 2
1 3 1
- wc 2)-i -:::tIl
2
1 i(W
2 - w2 1 2 - W 2 ) J 2w
c
~
1 ~
(3/2)w 2 w2
1 0
(w 2 - w/H 2
(w - w 2)2
C

3 3 1
thus '2(w 2 - wC 2)2 =w2 (w 2 - W 2)2
C

w2 = ~2 (w 2 _ wC 2)

w2 3w 2c or w 13w
c
therefore

f.
ml.n = fc x 13

Example 6.12

Two sections of rectangular wave-guide each have perfectly conducting

walls, of inner dimensions a and O.5a, enclosing a solid dielectric
of relative permittivity 2.25. The two wave-guide sections are
coupled by a third section of the same wall dimensions but which is
air-filled. This wave-guide system transmits energy from a source
operating at a free space wavelength of 2.5a into a matched load.
Explain why the air-filled central section causes the transmitted
signal to be attenuated.

Source Z1

Dielectric Air Dielectric

A B

Figure 6.7

156
 Determine the value of attenuation of the signal power. in dB. if
the air-filled section has a length of 1.5a.

For the dielectric wave-guide

1 2.25 ____1___ =9 - 6.25

Ag'2 = (2.sa/ (2a)2 (sa)2

Sa
A
g 1.658 = 3.015a (real)

For the air wave-guide

1 1 1 4 - 6.25
:f2 = 2 - ----z = 2
g (2.sa) (2a) (Sa)

A jsa -_ J·3 • 33 a (imaginary)

1.5
g

In the dielectric section. the propcgation wave is a normal TEOl

mode with B = 2~/(3.015a) rad/m while in the air section. the wave
mode is evanescent with an attenuation factor a = 2~/(3.33a) nepers/m.
-aL
Over the length l.sa. e = e -2.83 = 5.9 x 10 -2 By s1m11ar
..
reason-
ing Zl = E/H is real for the dielectric and Z2 = E/H is imaginary for
the air. There are three factors causing signal attenuation: (i) due
to the mismatch between Zl and Z2. only a fraction of the incident
energy crosses the junction to the air-filled section; (ii) due to the
evanescent mode, there is significant attenuation across the air-
filled section; (iii) the mismatch between Z2 and Zl causes a further
loss of energy at the second air dielectric-interface.

Let the incident E field at junction A in the dielectric be lEI.

Then the electric field transmitted through the interface at A is
given by
2Z2
due to the effective match.

Due to the evanescent mode. the incident field at junction B in the

air section is
IE'I = I E I e -aL
t

Thus the field transmitted through the interface at B is

IE' I
t
= IE' I I Zl
2Zl
+ Z2
I
Thus the attenuation (power) factor is
2
IE'I
t
2aL
~=

157
Now Z2/Z1 = j3.33/3.0l5 = jl.l, therefore

attenuation = 14 jl.l 12 5.9 2 x 10-"

.(1+jl.l)2 x

0.0138 nepers/m

PROBLEMS

1. Derive an expression for the field configurations in the TMI mode

of propagation between plane parallel conductors.
[ Ex
~
dx sa
= dy = - ---; cot (wt - sz) cot ~J
2. Calculate the average power transmitted and the peak electric
field across the b dimension of a wave-guide that has a width of
0.07 m and height of 0.035 m. The excitation is carried out in the
TE IO mode at a frequency of 3 GHz with a peak magnetic field intensity
of 10 A/m at the guide centre. Assume negligible attenuation.
[33 W; 5390 volts/m

3. The fields of the TE IO mode in a rectangular wave-guide with broad

dimension a and nar=ow dimension b are given by

E E . TIX -jsz
Y I Sln 11 e

The characteristic impedance of a transmission line can usually be

determined from a calculation of: (a) power and longitudinal current
or (b) power and transverse voltage or (c) transverse voltage and
longitudinal current. Show that for the TE IO wave-guide, three
different answers are obtained, but for TEM transmission lines these
three possibilities lead to the same answer. (The transverse voltage
can be taken as the line integral of the electric field at the centre
of the broad face of the wave-guide.)

4. A circular wave-guide has a radius a. It carries a TE mode of

propagation in which the field vectors are independent of the angular
co-ordinate. Derive the fields for the lowest mode of this type and
calculate the attenuation of the guide for this mode.

158
5. For a wave-guide 40 mm by 60 mm, calculate the cut-off frequency in
in the TEal mode, together with the phase velocity and wavelength at a
frequency 50 per cent above that of cut-off.
[2.5 GHz; 4 x 10 8 m/s; 107.2 mm]

6. (a) Consider a parallel plane wave-guide with plate separation of

200 mm with the TEl mode excited at 1 GHz. Find the propagation
coefficient y, the cut-off frequency and Ag the wavelength in the
guide. Take Er = 1.
(b) If the space between the conducting plates is now filled with
a dielectric material of E 4, find new values for y, f and A .
r c g
[(a) y = j13.85; 750 MHz; 455 mm; (b) y j39.3; 375 MHz; 160 mm]

7. The fields of the dominant TEal mode in a rectangular wave-guide

are of the form

E = Eo cos (ky sin e)ej(wt - kz cos e)

x

H Ho cos e cos (ky sl'n e)ej(wt - kz cos e)

y

j
Hz = Hoj sin e sin (ky sin 6)e (wt - kz cos e)

where k 2 w2~E and Eo/Ho I(~/E) = n. If the wave-guide walls are

at x = ± a and y = ± b, what is the cut-off wavelength in terms of
these guide dimensions?

From the above field expressions, describe whether or not the waves
would be disturbed by narrow slits cut in the wave-guide walls,
y = ± b, with the slits parallel to (a) the x axis, (b) the z axis.

Calculate the total time-average power flow along the guide.

[A = A ; (a) slits parallel to the x axis no effect;
g [1 - (A/AC)2H
(b) slits parallel to the z axis considerable effect; abnlHol2 cos 6]

8. The cross-section of a rectangular wave-guide has dimensions 30 mm

by 20 mm. Calculate the cut-off frequency (fc) for the fundamental
mode and sketch the field pattern of such a wave within the guide at
a frequency of 2f. Explain what will happen should an attempt be
c
made to transmit a wave of frequency less than f c '

For frequencies above the cut-off frequency, sketch the relation-

ship of the wavelength in the guide to free-space wavelength as a
function of f /f, deriving any formulae used. Hence show that the
c
phase velocity of the wave in the guide is greater than the velocity
of light, and comment.
[C.E.I. Part 2, Communication Engineering, 1979] [5000 MHz]

159
9. Derive an expression for the intrinsic wave impedance of a two-
phase wave-guide for both the TE and TM modes of operation, in terms
of wavelength A and guide wavelength Ag .
).l 1
377~ 2:. 2
Ag e: r

10. Prove by Maxwell's equations that it is impossible for a TEM

wave to be propagated inside a conducting tube, whether the wave-
guide is cylindrical or rectangular. (Hint: prove V is constant.
therefore E = 0 and hence H = 0.)

11. A rectangular wave-guide is filled with a dielectric material

of permittivity e: except for a section of length ~ which is empty
as shown in figur~ 6.8. The dimensions of the guide have been chosen
such that the dominant TEIO mode propagates in the dielectric section
but is evanescent in the empty section. A wave is incident from the
left and a matched load terminates the guide at the right.

Calculate the power dissipated in the load (in relation to the

power in the incident wave) as a function of the length ~.

[1 (a\:8 8Z f
+ sinh a~]

Figure 6.8

12. A waveguide comprises two extensive and perfectly conducting

flat sheets set parallel and distance a apart in air. The propa-
gation mode in the z-direction is such that the component field am-
plitudes in the y-direction are

H = 0 E
y
= A sin(~x/a) exp (-~z)
Y
Derive the other components of E and H, and sketch the field patterns
in the guide. Derive an expression for the average power transmitted
between the sheets in the z- direction per unit y-axis width.

[C.E.I. Part 2, E.F.N., 1980]

160
APPENDIX STOKES' THEOREM

Yl ,...
n
,.
',",

"
/';s
'"
-
E
x
Figure Al
Consider a square of area /';S in the x-y plane as shown in figure AI.
Let the electric field E have components E and E as shown. Now the
x y
work per coulomb required to move a charge around the perimeter of
the square is given by the line integral of E around the perimeter.

f
This work equals the total e.m.f. around the perimeter. that is

e.m.f. = E dt
Dividing by the area /';S and taking the limit of this ratio as /';S
approaches zero yields the curl of E normal to /';S at the point around
which /';S shrinks to zero. Thus
. IE dt
llm/,;s+O ~ = curInE (AI)

where curInE is the component of the curl of E normal to the area /';s.

Consider now a surface of area XIYl as shown in figure AI. Let

the area be divided into infinitesimal areas as suggested. From
equation Al the work per coulomb to carry a charge around an
infinitesimal loop divided by its area is equal to the curl of E at
the point. If the curl of E is integrated over the entire area xIYl,
all contributions to the total work cancel except the work along the
periphery of the area XIYl.

Hence it can be stated that the integral of the normal component

of the curl over the area xIYl must equal the line integral of E
around the periphery of the area.

IE dt = [ (curl E) ds (A2)

161
or f E dt = Is (V x E) ds (A3)

The relationship expressed in equation A2 or A3 is referred to as

Stokes' theorem as applied to electric fields. In general, the
theorem states that the line integral of a vector function around a
closed contour c is equal to the curl of that vector function over
any surface having the contour c as its bounding edge.

162
BIBLIOGRAPHY

It is beyond the scope of a book of this type to give an extensive

coverage of the many fine published books on field theory. Below is
a list of those books which the author believes will be useful to
the student for a study of the subject.

Carter, G.W., The Electromagnetic Field in its Engineering Aspects,

2nd ed. (Longman, London, 1967)
Hayt, W.H. Jr., Engineering Electromagnetics, 3rd ed. (McGraw-Hill,
New York, 1974).
Jordan, E.C., and Balmain, K.E., Electromagnetic Waves and Radiating
Systems, 2nd ed. (Prentice-Hall, Englewood Cliffs, N.J., 1968).
Kraus, J.D., and Carver, D.R., Electromagnetics, 2nd ed. (McGraw-Hill,
New York, 1973).
Morton, A.H., Advanced Electrical Engineering, (Pitman, London, 1973).
Parton, J.E., and Owen, S.J.T., Applied Electromagnetics, (Macmillan,
London and Basingstoke, 1975).
Seely, S., Introduction to Electromagnetic Fields, (McGraw-Hill,
New York, 1958).
Staniforth, J.A., Microwave Transmissions, (English Universities
Press, London, 1972).

163