Electric Cumat and

Magnetic Field

UNIT 12

GNETISM OF

Structure
12.1

Introduction
Objectives

12.3 Magnetic Field Due to a ~kgnetisedMaterial

12.4 The Auxiliary Field H (Magnetic Intensity)
12.5 Relationship between B and H for MagneticMaterial
12.6

Magnetic Circuits

12.8

Terminal Questions

12.9

Solutions and Answers

12.1 INTRODUCTION

In Block 2 of this course you have studied the behaviour of dielectric materials in
response to the external electric fields. This was done by investigating their properties
in terms of electric dipoles, both ~laturaland induced, present in these materials and
their lining up in the electric field. The macroscopic properties of these materials were
studied using the so-called polarization vector P, the electric dipole moment per unit
volume.
The magnetic properties of materials has a similar kind of explanation, albeit in a more
complicated form, due to the absence of free magnetic monopoles. The magnetic
dipoles in these materials are understood in t e r m of the so-called Amperian current
loops, first*introducedby Ampere.
All materials are, in some selae, magnetic and exhibit magnetic properties of different
kinds and of varying intensities. As you know, all materials, can be divided into three
main categories: (i) Diamagnetic; (ii) Paramagnetic and (3) Ferromagnetic materials. In
this unit, we shall study the macroscopic behaviour of these materials.
We understood the macroscopic properties of the dielectric materials using the fact that
the atoms and ~noleculesof these substances contain electrons, which are mobile and
are responsible for the electric dipoles, natural and induced, in these substances. The
polarisation of these substances is the gross effect of the alignment of these dipoles.
Similarly we describe the magnetic properties of various materials in terms of the
magnetic dipoles in these ~naterials.
In Unit 11, we have already explained diamagnetism and para~nagnetismin terms of
magnetic dipoles. In this unit, first, we will mention the origin of ferromagnetism.
Later, we will develop a description of the macroscopic properties of magnetic material.
With Unit 12, we end our study of magnetism, In the next Block we will deal with the
situation where both electric and magnetic fields will vary with time. This will lead,
ultimately, to the 'four differential equations known as Maxwell's equations.

Objectives
After studying this unit you should be able to :
0
understand and explain the terms: ferromnagnelism, amperian current,
magnetisation, magnetic intensity HI magneticsusceptibility, magnetic
permeability, relative permeability,
0

relate lnagnetisatio~lM (which is experimet~tallymeasureable) and the atomic

currents (which is 11ot measureable) within the material,
derive and utldersta~ldthe differential and integral equations for M and H and apply
these to calculate fields for simple situatioils,

F l e c t k C u m t and
Magnetic Field

Fig. 12.18 Domain

Cansider rwo electrons o n atoms

that are close to each other. If the
electron spins are parallel, they
stay away from each otl~erdue to
Pauli principle, therehy reducing
their caulomh energy of repulsion.
On theother hand, if these spins
are anti-parallel. the electrons can
come close lo eachother and their
wulonlb energy is higher. 'Thus, by
making their spins parallel, Ihe
electrons can reduce their energy.

w,p

interreIate B, H, M,

relate B & H for various magnetic al~dnon-magnetic materials,

derive all equatior~in allalogy with Ohtn's law for ;I ~nagneticcircuit.

a i ~ dX,

Ferrolnaglietic materials are those materials, which respoi~dvery slrongly to the

presence of magnetic fields. In such ~natcrials,the lnaglletic dipole monlenl of the
atoms arises due to the spins of ullpaired eleclrolls. These lelld lo liue up parallel lo each
other. Such a line-up does not occur over the whole material, but it occurs over a srnall
volume, known as 'domain'., as shown ill Fig. 19.1. However, these volulnes are large
compared to the atomic or lnolecular dimensions. Such line-ups take place even in Ihe
absence of an external magnetic field. You must be wollderil~gabout the nature of
forces that cause the spin lnagnetic nlonlellts of different atoms to lille up parallel to
each other. This can be explailled o111yby using quantum mechal~icalidea of "exchar~ge
forces". We will not go into the details of exchallge forces. About this, you will study ill
other courses of physics, but we are giving you sollle idea of exchange forces in the
margin remark.
In an unmagnctized fernmagnetic material, the lnagnetic momc~ltsof different domails
are rai~doinlyoriented, and the resulting inagnetic,moment of the ~naterial,as a whole is
zero, as shown in Fig. 12.9. However, in the presence of an extcrllal ~llagnelicfield, tile
magnetic moments of the domairls line-up in such a manner a s to give n net magnetic
mornent to the material in the direction of the field. The mechanism by which this
happens is thatJhe domains with the magnetic ~nornentsin the favoured directions
increase in size at the expense ofthe other domains, as show11 in Fig. 12.3a.

Fig. 122 : l l ~ domains

e
in an
unmagnctised bar of
iron. m e m w s show
the alignmen1
dirtclion or the
rnagnecc moment in
each domain.

Fi.123:
In a ferromagnetic ninlerial domniu changes, resulting in a net magnetic momenl, occur through
(e) domain growth and (b) domnin realignment,

In addition, the magnetic momel~tsof the entire domaills can rotate, as shown in
Fig. 12.3b. The material is thus magnetised. If, after this, lhe extenla1 lnagl~cticfield is
reduced'to zero, there still remains a considerable amount of lnaglleti7aIio11in the
material. The material gets permanently magnetized. The behaviour ol I'erromagnelic
materials, under the action of changing magnetic fields, is quite tolnplicaled and
exhibits the phenomenon, of hysteresis which literally means 'lagging behind'. You
will study more about this in Sec. 12.5.
Above a certain temperature, called 'Curie Temperature', because the forces of
thermal agitation domillate 'exchange' forces, the do~naillslosc their dipole ~noments.
The ferromagnetic material begills to behave like a paramagnetic ~naterial.When
cooled, it recovers its ferromagnetic properties.
Finally, we briefly mention two other types of nagn net ism wllich are closely related to
ferromagnetisrn. These are anti-ferromagnetism and ferrimagnetism ( also called
ferrites). In this course, we will not study the physics of antiferro- and ferrimagnetism.
The main reason for mentioning these materials is that they are of techi~ological
imporlance, being used in magnetic recording tapes, antenna and in colnputer memory.
In antifemmagnetic substances, the 'exchange' forces, as we menlioned earlier, play
the role of setting the adjacent atoms into antiparallel alignmen1 of their equal ~nagnetic

moments, that is, adjacent magnetic moments are set in opposite directions, as shown in
Fig. 12.4 a.
Such substances exhibit little or no evidence of magnetism present in the body.
However, if these substances are heated above the temperature known as Neel
temperature, the exchange force ceases to act and the substance behaves like any other
paramagnetic material.
In ferrimagnetic substances, known generally as femtes, the excha~igecoupling locks
the magnetic moments of the atoms i n the material into a pattern, as show11in
Fig. 12.4b. The external effects of such an alignment is intcnncdiate betwee11
ferromagnetisin and a~ltiferromagnetism.Again, here the exchange coupliiig disappers
above a certain temperature.
Thus, we find that the magnetization of tlie materials is due to permanent (and induced)
magnetic dipoles in these ~nalerials.The magnetic dipole moments in these materials
are due to the circulalillg electric currents, k~iowiias ampcrian currcnts at the atomic
and molecular levels. You arc expccted lo understand the correct relationship between
magnetization in a material and the amperian currents, togetherwith the basic
difference (and somelimes similarities) between tlie behaviour of the magnetic materials
in ~nagnelicfields, and dielectrics (and conductors) in electric fields.
Though physics of paramagn~ticand ferrolnagnetic materials have analogues in the
electric case, dian~agnetismis peculiar to ~naglietism.The student is advised to read the
matter in this unit and find the analogies and appreciate the differences, if any, by
referring back to the units on dielectrics. In the next section, we will find out the
relationship between the macroscropic quantity M, which is experilnentally measurable
and the atomic currents (a microscopic quantity) within the material which is not
measurable. With the help of this relationship, we can find out the magnetic field that
lnagnetised matter itself produces.

12.3 MAGNETIC FIELD DUE TO A MAGNETISED

MATERIAL
In Unit 5, we have.described the macroscopic propeflies of dielectric materials in tcrnls
of the polarization vector I', the origin of which is in the dipole lnolnents of its natural
or iliduced electric dipoles. We shall adopt a similar procedure in the study of magnetic
materials. You would be tempted to say that we should carry over all the equatio~lsin
the study of diclectrics to magnelic materials. One way of doing this would be to
replace the electric field vcctor E by D, then replace P by an allalogous quantily which
we,shall call lnagnetizatio~ivector M which is the lnagiietic dipole moment per unit
volume. Further, we replace the polarization charge density p, by magnetic 'charge'
density p, whatever that means, by writing V . M - p, just as we had V . I' = p,.
In fact, people did something Iike this, and they believed that magnetic charges or
monopoles exist. They have built a whole theory of electromagrielis~non this
assumption. However, we know that magnetic 'charges' or monopoles have not yet
been detected in any experiment so far, despite a long search for them. Now, we know
that the magrletization of matter is due lo circulating currenls within the atoms of the
materials, This was origirially suggested by Ampere, and we call these circulating
currents as 'amperian' current loops. These currents arise due to either the orbilal
motion of electrons in the atolns or their spila. These currents, obviously, do not
involve large scale charge trailsport in the magnetic materials as in the case of
conduction currents. These currents are also known as mag~ietizationcurrents, and we
shall relate these currents to the lnagnetizatioli vector M.
Let us consider a slab of uniformly magnetised material, as shown in Fig. 12.5a.
It contains a large number of atomic magnetic dipoles (evenly distributed throughout its
volume) all pointing in the same direction. If y is the magnetic moment of each dipole
then the magnetisation M will be the product of y and the number of oriented dipoles
per unit volume. You know that the dipoles can be indicated by tivy current loops.
Suppose the slab collsists of many tiny loops, as shown in Fig. 12.5b. Let us consider
any tiny loop of area a, as shown in Fig. 1 2 . 5 ~In
. terms of magnetisation M , the
magnitude of dipole moment p is written as follows:

Madz

where dz is the thickness of the slab.

(12.1)

Magnetism of Makrials-I1

Fig. 124 :Relative onentalion ul'

electron spins in (n)
antiferromagetic niatcrinl
and (b)ferrite.

Electric C u m t lad
Magnetic Field

(dl
Fig.125: (a) A thin slab dunilomly magnetized material, with Lhe dipoles indicated by (h) and (c) tiny
current loops is cquivdent to (d) n ribbon ofcurrent I flowing around the boundnry.

If the tiny loop has a circulating current I, the11dipole lnoinent of the tiny loop is given by
y

la

(12.2)

Equating (12.1) and (12.2) we get

I
M = - or I
dz

Mdz.

Here we have assumed that the current loops correspondi~igto magnetic dipoles are

1
I
I

I
I

I
I

>

84

large enoughso that magnetisation does not vary appreciably rrom one loop to the next,
s o Eq. 12.3 shows that the current is the same in all current loops o l Fig. 12.5b. Notice
that within the slab, currents flowing in the various loops cancel, because everytime if
there is one going in one particular direction, then a c o ~ i t i ~ ~ uone
o u s is going io the
exactly opposite direction. At the boundary of the slab, there is no adjacent loop to do
the cancelling. Hence the whole thing is equivalent to the single loop of c u r r e d I
flowing around the boundary, as shown in Fig. 12.5d. Therefore, the thin slab of
lnagnetised inaterial is equivalent to a single loop carrying the current Mdz . Hence, the
magnetic field at any point extenla1 to the slab, is the same as !hat of the currelit Mdz .
In case there is non-uniform magnetization in the material, the atr_l!nic currents iu the
(arnperian) circulating current loops do not have the same inagtlitude at all points inside
the material and, obviously, they do not callcel each other out illside such a material.
Still we will find that magnetised matter is equivalent to a curre~ltdistributio~i
J = curl M. Let us see how we have anived at this relation.
In the non-uniformly magnetised material consider two little blocks of the volurne
'Ax A y A z, cubical in shape adjacent t~ each other along y-axis (see Fig. 12.6a). Let us
call these blocks '1' and '2' respectively. Let the z-compo~ienld l M i n these blocks be
M z ( y ) and Mz ( y + Ay ) respectively.

,
L

Magnetism oCMateri&-I1

~ ethe
t ampenan currents circulating round the block '1' be I ( 1 ) and round the block
'2' be1( 2 ).Using Eq. (12.3) and refemng to Fig. 12.6a we write,
Ix(1) = Mz(y)Az
and

1,(2)

M,(y+Ay)Az

Fi.116: Two adjacent chuncksolnmgnetised malerial, with a larger arrow on the w e to the right in
(a) and above in (b), suggesting greater magnetisation rt that p h i On the surface where they join
them i s r net c u m n t in the x-direction

At the interface of the two blocks, there will be two contributions to the total current:

I( 1 ) flowing in the negativex-direction, produced due to block 1, and1( 2 ) flowing in

the positivex-direction produced due to Block 2. The total current in the positive
x-direction is the sum :

I f the magnetisation in the first block

is M ( x. y. :). the magnetisation in
the second block is

The tcomponent of magnetisation

ofthe first block in termsoil. ( 1 )is
wrillen as

Eq.(12.4) gives the net rnagnetii.ation current in the material at a point in the
x-direction in t e r n of the z-component of M. The current per unit area, i.e., current
density Jm flowing in thex-direction is give11as follows :

where Ay & is the area of cross-section of one such block for the current A Ix.Hence

In these equations, we have put suffixes x to the currents to indicate that, at the interface
of the blocks, the current is along the x-axis.
There is another way of obtaining the current flowing inx-direction by considering
these two tiny blocks, one above the other, along the z-axis, as shown in Fig. 12.6b. We
obtain the relation as

By superimposition of these two situations, we get

Eq. (12.7) is obviously the x-component of a vector equation relating Jm and the curl of
M.Combining this withy and r components, we obtain

Eq. (12.8) is a more general expression representing the relationship between the
.
magnetisation and the equivalent current. We see from Eq. (12.8) that inside a
uniformly magnetized material in which case M = constant; we have Jm = 0. This is
true. See Eq. (12.8), the current is only at the surface of the material where the

M,Az

I,(l)

Similarly, the:-component o l
magnet isation oithe second block
neglecting high-order ternis which
vanish in the limit where cach'block
becomes very sn~all.isgiven by

Electric C u m t nnd
Magnetic Field

magnetization has a discontinuity (dmpi'ng from a finite M to zero). Inside a

non-uniformly magnetized material, J, is nonzero.
We shall see in the next section that J,, which is introduced to explain the origin of
magnetisation in a material, is made to make its exit from the equation, and only the
conductioncurrent density indicating the actual charge transport and which is
experimentally measurable remains.

12.4 THE AUXILIARY FIELD H

INTENSITY)

GNETIC

'

So far we have been considering that magnetisation is due to current associated with
atomic magnetic moments and spin of the electron. Such currents are known as bound
currents or magnetisation amperian current. The current density J, in Eq. (12.8) is the
bound current set up within the material. Suppose you have a piece of magnetised
material. What field does this object produce? The answer is that the field produced by
this object is just the field produced by the bound currents established in it. Suppose we
wind a coil around this magnetic material and send through this coil a certain current I.
Then the field produced will be the sumof the field due to bound currents and the field
due to current I. The current I i s known as the free current because It is flowing through
the coil and we can measure it by connecting an ammeter in series with the coil. (In
case the magnetic material happens to be conductor, the free current will be the current
are those caused by
flowing through the material itself.) Remember that free curre~~ts
external voltage sources, while the internal currents arise due to the motion of the
electrons in the atoms. The current is free, because someone has plugged a wire into a
battery and it can be started and stopped with a switch. Therefore, the total current
density J can be written as

where Jfrepresents the free current density.

Let us use Ampere's law to find the field. In differential form, it is written as (See Unit 9)

VxB

WJ

(9.46)

Using Eq.(12.9), Ampere's law would then take the form as follows:

As mentioned earlier, we have no way to measure J, experimentally, but we have a

way to express it in terms of a measurable quantity, the magnetization vector M through
the Eq.(12.8). We then have

Eq.(12.10) is the differential equation for the field

in terms of its source Jfi

the free current density. This vector is given a new symbol H, i.e.,

The vector H is called the magnetic 'intensity' vector, a name that rightly belongs to B,
but, for historical reasons, has been given to H. Using Eq.(12.11). Eq.(12.10) becomes

V x H = 31

(12.12)

In other words, H is related to the free cumnt in the way B is related to the total
current, bound plus free. This surely has made you think over the purpose of
introducing the new vector field A. For practical reasons the vector H is very useful as
it can be calculated from the knowledge of external current only; whereas B is related to
the total current which is not known. Eq. (1312) can also be written in the integral form
as

!
86 ' '

Magnetism oCMnttrinls-11

(11.13)
where Ijis the conduction current through the surface bounded by the path of the line
integral on the left. Here the line integral of H is around the closed path which may or
may not pass through the material.This equation can be used to calculate W, even in the
presence of the magnetic material.
SAQ I

Fig. 12.7 shows a piece o f imn wound by a coil canying a current of 5A. Find the value
of J H . dl around the path (I), (2) and (3). Also state for which path(s)
B = B and B ;. N.
Fmm Eq. (12.3), we see that the units in which M is measured is amperes per meter.
Eq. (12.11) shows that the vector W has the units asM, hence W is also measured in
amperes per metre. The electrical engineers working with electromagnets, transformers,
etc., call the unit o f H as ampere turns per metre. Since 'turns', which is supposed to
imply the number of turns in the coil canying a current, is dimensionless, it need not
confuse you.
Fig.127: SAQl

Magnetic properties of substance

In paramagnetic and diamagnetic materials, the magnetisation is maintained by the
field. When the field is removed, M disappears. In fact, it is found that M is .
proportional to B, provided that the field is not too strong. Thus
It is conventional to express Eq. (12.14) in terms of H instead of B. Thus we have

xmH

(12.15)

The constant of proportionality -A,is called the magnetic susceptibility of the material.
It is a dimensionless quantity, which varies fmrn one substance to another. We can
cl~aracterisethe magnetic properties of a substance by %.It is negative for diamagnetic
substances and positive for paramagnetic materials. Its magnitude is very small
compared io unity, that is ( X, I < < 1. For vacuum X, is zero, sirice M can only exist in.
magnetised matter. We give below a short table giving the values of X , for diamagnetic
and paramagnetic substances at room temperature.

Paramagnetic

Aluminium

Paramagnetic

Sodium

Paramagnetic

Tungsten

Paramagnetic

Oxygen

Diamagnetic

Bismuth

Diamagnetic

Copper

Diamagnetic

Silver

Diamagnetic

Gold

We have not given a table for the susceptibilities of fernmagnetic subs&nces as f i ,

depends not only on H but also on the previous mangetic history of the material.
Using Eq. (12.11) in the fonn

B
we have

w(W+M)

Electric Cumnt and

Mmgnttic Field

..
where

(12.17)
c~ = w K ,

clo( I + % )

p is called the permeability of the medium and Km is called the 'relative'

permeability. We see that p has the same dimensions as po and Km is dimensionless.
In vaccum xm = 0 and p = po. Relative permeability Km differs from unity by a very
small amount as Km = ( 1 + X , ). K,,,for para- and ferromagnetic materials are greater

than unity and for diamagnetic material it is less than unity.

The magnetic properties of a material are completely specified if any one of the three
quantities, magnetic susceptibility, )(F, relative permeability K, or permeability p is
known.
Example 1

A toroid of aluminium of, length lm, is closely wound by 100 turns of wire canying a
steady current of 1 A. The magnetic field B in the toroid is found to be
1.2567 x
wbm-'. Find (i) H, (ii) %, and Km(iii) M in the toroid and (iv) equivalent
surface magnetization arrent I,.

Solution

i)

According to Eq.(12.1 3)

To evaluate H produced by the current, we consider a circular integration path

along the toroid. H is constant everywhere along this path of length lm. The
number of cumnt turn threading this integration path is 100 x 1A . Since H is
everywhere parallel to the circular integration path, we get
H x l m = 100xlA

(ii) From Eq. (12.16)

(iii) From Eq. (12.15)

M=xmH

(iv) I, = ML

In this solution, we have assumed B, H and M to be uniform over the cross-section of

the toroid and along the axis of the toroid.

Try to do the following SAQ.

Magnetism of Mr~nls-11

SAQ 2
~n air-core solenoid wound with 20 turns per centimetre canies a current of 0.18 A.
Find N and B at the center of the solenoid. If an iron core of absolute permeability
6 x lom3H m-I is inserted in the solenoid, find the value of H and B?
( w = 4xx 1 o d 7 ~ m - l )

12.5 RELATIONSHIP BETWEEN B AND H FOR

GNETIC
The specific dependence of M on B will be taken up in this section. The relationship
between M and B or equivalently a relationship between B and H depend on the nature
of the magnetic material, and are usually obtained from experiment.
A convenient experimental arrangement is a toroid with any magnetic material in its
interior. Around the tomid, two coils (primary and secondary) are wound, as shown in
Fig.12.8.

Fh.128 i Anangcmcnt for invatiptlng Lbe rebtion between B and M, or I3 and H, in r rnagoetic rnalrria~'

If we consider the radius of the cross-section of the toroidal windings to be small in

comparison with the radius of tbe toroid itself, the magnetic field within the toroid can
be considered to be approximately uniform. A current passing through the primary coil
establishes H. The establishment of the current in the primary coil induces an
electromotive force (emf). By measuring the induced voltage, we can determine
changes in flux and hence, in B inside the magnetic material. If we take H as the
independent variable, and if we keep the track of the changes in B starting from B 0,
we can always know what B is for a particular value of H. In this way, we can obtain a
B-H curve for different types of magnetic material.
The experiment described above can be carried out for diamagnetic and paramagnetic
materials by commencing with I = 0 and slowly increasing the value of I to obtain a
series of values of B and H . A plot of B against H for these substances is shown in the
Fig. 12.9(a). We see that the graph is a straight line as expected from the relation
B = w(l+Xrn)H
(12.16)

hl291 latema1 mmp& hld (B) versus applied magnetic &Id (H) for diBemml lypa of rnqnetlc
rnatrrtk (a) In d l r n n p d i c and prunrgnclic mrlrrhb, Lbc trhhwhlp Is bear. @) In
fermmgaetic mrlnlab, Lbt dolrtimdp depends m Ihc s l r c ~ q t hd t b c r p p W lkkl and OD h e
p u t history of (be materid, In (b), lbt Lkkl rtrmgthr .low tbt v d a l axis u e much gruLcr lbrn
along Lbe hmimn(.l .xis, Arrows Indlutc Lhc d W o n in wbieb Lbc Oclbr ut changed

Electric Currenlmd
Magnetic Field

where pa and -hare conslants. The slope of the graph is given by ( 1 + -h) from which
call be determined using the following relation:

For diamagnetic substances, slope <

slope > po so that y, > 0.

making X, < 0. For paranlag~leticmaterials

If in the experiment givenabove we use ferromagnetic materials like iron, we obtain a

lypical B - Hcurve as showri in Fig. 13.9(b).
i)

At I = 0 , i.e., when H = 0, B is zero. When I is increased, B and H are

detennined for increasing values of 1 .At first, B increases with H along the curve
'a'. At some high value ofH, the curve (shown by the dashed line in the figure)
becomes linear, indicatillg that M ceases to increase, as the material has reached
saturation with all the domain dipole molilents in the saliie direction.

ii)

If, afterreaching saturation, we decrease the current in the coil to bring H back to
zero, the B-Hcurve falls along the curve 'b'. When H reaches zero, there is still
some B left implying that even when/ = 0, there is still some magnetizatio~~
or M
left in the specimen. The li~aterialis pennaiieiitly magnetized. The value of B for
H = 0 is called remanence.

iii) If the current is reversed in the primary coil and made to increase its value, the
B-H curve rulls along the curve 'b ' until B becomes zero at a certain value of H.
This value o f H is called the coercive force. If we co~~tinue
to increase the value of
the current i ~the
i negative direction, the curve col~tinuesalong '6 ' until the
saturatioli is reached again.
iv) The current is now decreased until it becomes zero once again. This corresponds to
H = 0 , bul B is not zero and has magnetization in the opposite direction. Here we
reverse the current again, so that the current in the coil is once Illore along the
positive direction. With the increasing current in this direction, the curve continues
along the curve 'c ' to meet the curve 'b ' at saturation.
If we alternate the current between large positive and negative values, the B-Hcurve
goes back and forth along ' b ' and ' c ' in a cycle. This cycle curve is called hysteresis
curve. It shows that B is 1101a single valued function of H , but depends on the previous
treatment of the material.
The shape of the hysteresis loop varies very widely from one substance to another.
Those substances, like steel, alnico, etc., from which pennanent magnets are made,
have a very wide hysteresis loop with a large value of the coercive force (see Fig.
12.10). However, those substances, like soft iron, permalloy, etc., from which
electromagnets (temporary magnet) are made, should have large remaliance but very
small coercive force. Those ferromagnetic materials, which are used in the cores o f
transformers, like iron-silicon (0.8-4.8%) alloys, have very liarnjw hysteresis loop.

Fig.1210: h r hysteresiscurves for a few materials. ~ w e s ' ( aaod

) @) are respectively for specimen of a d '
iron and steel mnteriels,

12.6 MAGNETIC CIRCUITS

A magnetic circuit is the closed path take11by the lnaglietic flux set up i11 an electric
machine or apparatus by a magnetising force. (The maglietising force may be due to a
currelit coil or a pennanent magnet.)

'90
_-ei

In order to study the resernblarice between a ~iiagneticcircuit and an electric circuit, we

will develop a relation corresponding to Ohii's law for a magnetic circuit. Let us
consider the case of an iron rillg (Fig. 11.11) ~ilagl~etised
by a current flowing through a
coil wound closely over it. Suppose,

I
N

= current flowing in the coil

=

number of tunls in the coil

I = length of the magnetic circuit

(mean circulnference of the ring)
A
p

area of cross-section of the ring

pernleability of iron.

In this case, all the magnetic flux produced is confined to the iron ring with very little
leakage (we shall see the reason for this later). We have seen earlier that H illside the

'I4

Fig. 12.1 1: Magnetic circuit

J H .= ~NI
I

(from Ampere's law)

where the path of integration is along the axis of the ring. As the line integral of electric
field E over a circuital path is the electromotive force (e.m.f), by analogy, the line
integral of H is termed.as magnetomotive force (M.M.F.)

... M.M.F. = J H .

dl

NI.

At every point along fhis path in the ring, we write

Further if Q is the magnetic flux given by @

BA, then H

- a/@,

hence

where we have taken @ outside the integral as it is constant at all cross-sections of the
ring. Eq. (12.19) rerni~~ds
us of a similar equation for an electric circuit colltaining a
source of E.M.F., namely,
.
emf.

- current

x resistance =

IJ

The Eqs. (12.19) and (12.20) suggest that:

i)

The magnetomotive force ( J H . ~) is analogous wit11 e.1n.f. (JE

ii)

The magnetic flux

is analogous with current I in Ohm's law.

iii) The magnetic ~ s i s t a l l c eknown as reluctance

resistance (J

. ti^ ).

y)

:. M.M.F.= flux x relucta~ice

- .- M.M.F.
NI

is analogous with electric

. .

Electric Current and

Magnetic Field

If we take p to be constant throughout the ring then

where L is the letigth o f the ring. However, we must recogrlise the significant
difference betweeti all electric circuit and a inag~leticcircuit:
Energy is colltinuously being dissipated in the resistallce of the electric circuit,
i)
whereas no energy is lost in the relucta~lceof the ~nagneticcircuit.
ii)

The electric current is a true flow of the eleclro~lsbul there is no flow of such
paflicle in a magnetic flux.

iii) At a given teinperature, the resistivity p is iildepeildcilt of current, while the

1

correspondillg quantity - i n reluctailce varies with inag~ieticflux a.

CI

Reluctances in Series : Let us assume that the toroid is made of no re thaii one
ferromag~ietic~naterial,each of which is of the same cross-sectional area A , but with
different pern~eabilitiespi, p ~. ,. . . .

Fig.12.12 : (a) A magnetic circuit conlpmed of several ~naterials:Reluctances in series. (I)) 1 G e t i c circuii
consisting of two loops: Reluctances in parallel.

Then, (see Fig. 12.12) as before, we have

NI = J H A

where the itltegnls on the right are taken over axial paths in the materials 1,2, ....
There fore,

s o that the total relucta~lceof the given magnetic circuit is given by

% = 81 + & + . ....

I
I

(11.23)

Reluctances in Parallel : We shall next illustrate the case of a magnetic circuit i11 which the
reluctances are in parallel.Fig. 12.12b shows such a magnetic circuit. The current canying
coils have N tun6 each, canying a current[amperes. The magnetic flux threading the
coil splits into two paths with fluxes @ I and 4): as shorv~iin the figure. dhviously,
=
+ 02.We assume that the area of cross-sectionA is coilsta~iteverywkre in the
circuit.
Let the lengths of the paths ABCD, DA and DEFA shown in the figure be L, L1, L2
respectively. For the pathABCDA , we have

~irnilarlyfor the closed path ADEFA , we haye

Notice that we have used CLI and y: for the paths AD and DEFA, As a ' s being
different for these paths, H s would be different. This inakes ps different in these paths.
Using @ = a 1 + 0.and Eq. (12.25), we write

Substituting the value of @I from the above equation in the Eq. (12.24), we have

This shows that the relu'ctaiices of th6 paths DA alld DEFA are in parallel as the
magnetic flux 4) splits into $1 and $2 along these paths respectively. The combined
reluctance 'Si of these paths is given, in tenns of the reluctances ?Ill and 8:of these
paths, as follows

Notice that the Eq. (12.24), (12.25) and = 0

1+ are the statements of Kirchoff's
laws for the inagnetic circuits.
Now we see why the magnetic flux does not leak through the air. Air forms a parallel
path for the flux, for air, p = po and for a ferromagiletic material p L. lo4 pc, hence
the aiipath is a very high reluctaiice path compared to that through the ferromagnetic
material. The ~nagiieticflux will follow the path of least reluctance, a situation similar
to that in the electric circuit.
The magnhtic circuit fonnulae are used by the electrical engineers in calculations
relating electromagnets, motors and dynamos. The problem is usually to find the
number of tunsand the current in the winding of a coil, which is required to produce a
certain flux density in the air gap of a11 electromagnet. Knowing the reluctance of the
circuit, M.M,F. is calculated froin the relation:

M.M.F. = flux x reluctance

Since M.M.F. is also NI (see Eq. (13.19)), the magnitude of aippere turns can be
calculated. Let us illustrate it by studying the magnetic circuit of an electromagnet.

Magnetic Circuit of an Electromagnet

The magnetic circuit o f an electromagnet coi~sistsof the yoke which fo& the base of
the magnet, the limbs on which the coil is wound, the pole pieces and the air gap. See
Fig. 1113. Let 11 be the effective length and a the area of cross-section of the yoke. If
11
1 1 is the permeability o f its iron, then -is
the teluctance of the yoke. Similarly the
I11 a1
1: .
13
reluctance of each limb is -and
the reluctance of each pole piece is -while
w a?
p3 03'
14
the reluctance of the air gap is -(because
pairI \ ~ ) .Hence the total reluctance ~f
~b a4

Ydro
Fig. 12.13 : Mqpclic circuit d
an dcclromagnd

the magnetic circuit is

-

11L

2 2

a
3

14

If the magnetic circuit carries one and the same flux cP across all its parts, then
according to Eq.(12.19),the number of ampere turns is:

Let us take another example of calculating the magnetic field B in the air gap of a toroid
of Fig. 12.14. Here the toroid is of a ferromagnetic materiaL (soft iron) with a small air
gap OFwidth 'd ' which is small compared to the length L of the toroid. For this case,
we have

NI = @

[vI$+

O being the flux through this magnetic circuit.

-w
B

-[u(L-d)+lul]

or
I

Air @

I& U L I~~q p a e t k w ~
&.lrw~

B =

(2.29)
I

NI ,PI
cld,+(v-w)d

This is the value of the magnetic field in the air gap. Read the following example which
shows bow the air gap effectively increases the length of the toroid.

Example 2
Compare the examples of a complete toroid of length L wound with a coil of N turn
each carrying a current I amperes and of a toroid of lenth( L - d ) with an air gap of
length d ( d <<L ) . Show that the air gap effectively increases the length of the toroid
by ( Km- 1 ) d, where Kmis relative permeability.

I
I

Solution
In the case of a complete toroid without the air gap, we have B

NI/(:). In the

event of an air gap of length d, we have fmm Eq. (12.29):

NI ,Po

Pd+(P--lJ4l)d
Dividing both the numerator and the denominator by ppo, we get

B =

NI

1-CL[ ( L - d ) + E dw]

NI

so that

NI
1
-[L+(Km-l)d]
P'

I
I

If we compaE this formula with that for the complete toroid, we see that L is effectively
increased by ( K,,,
- 1 ) d.

Before ending this unit solve the following SAQ.

>

I
,
'

SAQ 3
A soft iron ring with a 1.0 cm air gap is wound with a coil of 500 turns and carrying a
cumnt of 2 A. The mean length of iron ring is 50 cm,its cross-section is 6 cq2, its
permeability is 2500 p a Calculate the magnetic induction in the air gap. Eind also B and
H in the iron ring.

'

94

i 1
C

Let us now sun1up what we have learnt in this unit.

The behaviour of the ferromagnetic materials is complicated on account of the

permanent magnetization and the phenomenon of hysteresis. This behaviour is
explained by the presence of the domains in these materials. In each domain the
dipole moments are locked to remain parallel due to 'exchange' force. However, in
the unmagnetised state, the magnetisation directions of differentdomains are
random, resulting in a zero net magnetisation. There also exist two other kinds of
magnetic materials : antiferromagneticand femmagnetic.
For non-uniform magnetisation, magnetised matter is equivalent to a current
distribution J = curl M,where M is magnetisation or magnetic moment per unit
volume.
The magnetic field produced by the magnetised material is obtained by Ampere's
law as follows:
VxB

Jf+J,

where J f is the h e current density which flows through the material and J , is the
bound current density which is associated with magnetisation. This gives

when B
e
,

(- B - )

M is a new field vector.

For paramagnetisin and diamagnetism B, M aiid H are linearly related to each

other except for ferromagnetic materials which exhibit hysteresis, a non-linear
behaviour.

The study of the electromagnets, motors and dynamos involves the problem of
current carrying coils containing ferromagnetic materials, i.e., it involves the study
of magnetic circuits. We speak of the magnetic circuits when all the magnetic flux
present is confined to a rather well-defined path or paths.

Magnetic circuit formula is:

magnetomotive force ( M.M.F. ). = flux x reluctance

M.M.F. is also equal to NI where N is the number of turns of the coil wound over .
the magnetic material and I the current flowing through each coil.
1
Reluctance 3 = -

Cur

where I, a and p are the length, area of cross-sectionand permeability of the material.
Additions of reluctances obey the same rules as additions of resistances.
I

12.8, TEWINAL QUESTIONS

1) Find the magnetizing field H and the magnetic flux density B at (a) a point of
105 mm from a long straight wire carrying a current of 15 A and (b) the center
of a 2000-turn solenoid which is 0.24m long and bears a currefit of
1.6A.(
4n x lom7H m" )

2)

A toroid of mean circumference 0.5 m has 500 turns: each carrying a current'of
'
0.15 A. (a) Find H and B if the tordid has an air core. (b) Find B and the
magnetization M if the core is filled with imn of relative permeability 5000.

3)

A toroid with 1500 turns is wound on an iron ring 360 mm2in cross-sectional area,
of 0.75-m mean circumference and of 1500 relative permeability. If the windings
carry 0.24A, find (a) the magnetizing field H (b) the m.m.f,, (c) the magnetic
induction B, (d) the magnetic flux, and (e) the refuctance of the circuit.

12.9 SOLUTIONS AND ANSWERS

SAQs
1) Path (1) encloses

= 5A

.: IH.d I
For Path (2), $H .dl
ForPath(3),

= 5A

71 = 35A

J H .=~2f
B

1OA

H forPath(1)

H for Paths (2) a~ld(3) because these paths pass through iron.

H E n l - (2000m-')(0.18~)

2)

-- -

360~rn-'

If an iron core of absolute permeability 6 x

then H remains unchanged i.e.,

H m-' is inserted in the solenoid

H = 360 Am-' ( unchanged )

andB = p H = ( 6 x lom3~ m - l ) . ( 3 6 0 ~ m - ' ) 2.16T

The expression for the magnetic induction in the air gap is

3)

NIP
[L+(Km-l)d]
Substituting the values, given in question, we get

B in the iron ring has tbe same value as in air, but H in iron is given by

Tenninal Questions
1)

2)

(a)

--

HI B

clo

For a toroid H

-w
x - -I 1
2m 1.l0

= nl, a

-I

2nr

d we &B

15A
(b)(O.l05rn)

22.7 A me'

(&/lo7) ( K J I )

pH. Thus

(b)

B = 5000 ( 0.188 mT) = 0.94T

Using B/w = H + M

d)

O = BA = ( 0.905 Wb mb2)

(e)

Reluctance

m.rn.f.
=

= 3.26 x
Wb.
106
360
= 1:l x 106~''
3 . ~ ~ 1 0 - ~

Magnetism of Materials-11