Magnetic Circuits and Induction 33

SOLUTION
(a) In Example 2.1 the value of l was found as 1.152 Wb-T for Bc = 1.2 T. Therefore, for sinusoidai variation of Bc,
l = 1.152 sin 314t Wb-T
The emf is
dl
e= =361.7cos 314t V
dt
lc 0.4
(b) Rc = =
m 0 m r Ac 4 p ¥ 10- 7 ¥ 6000 ¥ 16 ¥ 10- 4
= 3.317 ¥ 104
lg 6 ¥ 10 - 4
Rg = = = 29.856 ¥ 104
m 0 Ag 4 p ¥ 10- 7 ¥ 16 ¥ 10- 4
(c) From Example 2.1
i = 1.06 A
l 1.152
\ L= = =1.09 H
i 1.06
It can also be found by using Eq. (2.21). Thus
N2 N2 (600) 2
L = N 2P = = =
R Rc + Rg (3.316 + 29.84 ) ¥ 104
= 1.08 H
(d) The energy stored in the magnetic field is from Eq. (2.32)
l l l 1 l2
Wf =
Ú0
id l =
Ú 0 L
dl =
2 L
2
1 (1.152)
= ¥ = 0.6144 J
2 1.08

When a magnetic material undergoes cyclic magnetization, two kinds of power losses occur in it—hysteresis
and eddy-current losses—which together are known as core-loss. The core-loss is important in determining
heating, temperature rise, rating and efficiency of B
transformers, machines and other ac run magnetic Bm
devices. g b

Hysteresis Loss c
dw1= HdB
Figure 2.18 shows a typical hysteresis loop of a d
ferromagnetic material. As the mmf is increased from 0 a Hm H
zero to its maximum value, the energy stored in the field
Bf
per unit volume of material is
Bb = Bm
Ú-Bf
HdB = area ofabgo e

Fig. 2.18 Hysteresis loss

34 Electric Machines

As H is now reduced to zero, dB being negative, the energy is given out by the magnetic field (from the
exciting coil back to the voltage source) and has a value
Bc
ÚBb = Bm
HdB = area cbg

The net energy unrecovered in the process is area ofabco which is lost irretrievably in the form of heat and
is called the hysteresis loss. The total hysteresis loss in one cycle is easily seen to be the area of the complete
loop (abcdefa) and let it be indicated as wh (hysteresis loss/unit volume). Then hysteresis loss in volume V of
material when operated at f Hz is
Ph = whVf W (2.35)
In order to avoid the need for computation of the loop area, Steinmetz gave an empirical formula for
computation of the hystersis loss based on experimental studies according to which
Ph = kh f B nm W/m3 (2.36)
where kh is a characteristic constant of the core material, Bm is the maximum flux density and n, called the
Steinmetz exponent, may vary from 1.5 to 2.5 depending upon the material and is often taken as 1.6.

Eddy-current Loss
When a magnetic core carries a time-varying flux, voltages are induced in all possible paths enclosing the flux.
The result is the production of circulating currents in the core (all magnetic materials are conductors). These
currents are known as eddy-currents and have power loss (i 2R) associated with them called eddy-current loss.
This loss, of course, depends upon the resistivity of the material and lengths of the paths of circulating currents
for a given cross-section. Higher resistivity and longer paths increase the effective resistance offered by the
material to induced voltages resulting in reduction of eddy-current loss. High resistivity is achieved by adding
silicon to steel and hence silicon steel is used for cores conducting alternating flux. Dividing up the material
into thin laminations along the flow of flux, with each lamination lightly insulated (varnish is generally used)
from the adjoining ones, increases the path length of the circulating currents with consequent reduction in
eddy-current loss. The loss in fact can be shown to depend upon the square of lamination thickness. The
lamination thickness usually varies from 0.3 to 5 mm for electromagnetic devices used in power systems and
from about 0.01 to 0.5 mm for devices used in electronic applications where low core-loss is desired.
The eddy-current loss can be expressed by the empirical formula
pe = ke f 2B2 W/m3 (2.37)
wherein
ke = K¢e d 2/r (2.38)
d being the thickness of lamination and r the resistivity of material.
It is only an academic exercise to split the core-loss into its two components. The core loss in fact
arises from two types of flux variations: (i) flux that has a fixed axis and varies sinusoidally with time as
in transformers (this is the type visualized in the above discussion), (ii) flux density is constant but the flux
axis rotates. Actually in ac machines as well as in armature of dc machines the flux variation comprises both
these types occurring simultaneously. The core-loss is measured experimentally on material specimen and
presented graphically. Typical values of the specific core-loss (W/kg of material) are displayed in Figs 2.19
(a) and (b) for cold-rolled grain-oriented (crgos) steel. It is easy to see from these figures that for reasons
mentioned above specific core loss is much higher in machines than in transformers.
 Magnetic Circuits and Induction 35

5 50

4 40
Specific loss (W/kg)

Specific loss (W/kg)

0.5 mm 0.4
mm
3 30

0.35 mm
2 20

0.35 mm
crgos
1 10

0 0
0.8 1.0 1.2 1.4 1.6 1.8 0.8 1.2 1.6 2.0 2.4
B Flux density (T ) Flux density (T )
(a) (b)

Fig. 2.19 Core-loss at 50 Hz: (a) transformers, (b) machines

EXAMPLE 2.9 The total core loss of a specimen of silicon steel is found to be 1500 W at 50 Hz. Keeping
the flux density constant the loss becomes 3000 W when the frequency is raised to 75 Hz. Calculate separately
the hysteresis and eddy current loss at each of those frequencies.

SOLUTION From Eqs. (2.36) and (2.37) for constant flux density, total core loss can be expressed as
P = Af + Bf 2 or P/f = A + Bf
1500/50 = A + 50 B or 30 = A + 50 B (i)
3000/75 = A + 75 B or 40 = A + 75 B (ii)
Solving Eqs. (i) and (ii), we get A = 10, B = 2/5
Therefore P = 10f + 2/5 f 2 = Ph + Pe (iii)
At 50 Hz Ph = 10 ¥ 50 = 500 W
Pe = 2/5 ¥ 2500 = 1000 W
At 75 Hz Ph = l0 ¥ 75 = 750 W
Pe = 2/5 ¥ (75)2 = 2250 W

2.7 PERMANENT MAGNETS

The permanent magnet is an important excitation source (life long) commonly employed for imparting energy
to magnetic circuits used in rotating machines and other types of electromechanical devices. There are three
classes of permanent magnet materials (or hard magnetic materials) used for permanent magnet dc (PMDC)
motors: Alnicos, ceramics (ferrites) and rare-earth materials. Alnico magnets are used in motors up to
200 kW, while ceramic magnets are most economical in fractional kW motors. The rare-earth magnetic
materials are very costly, but are the most economic choice in very small motors. Latest addition is
neodymium-iron boron (Nd FeB). At room temperature, it has the highest energy product (to be explained
later in this section) of all commonly available magnets. The high permeance and coercivity allow marked
reductions in motor frame size for the same output compared to motors using ferrite (ceramic) magnets. For
very high temperature applications Alnico or rare-earth cobalt magnets must be used.
12 Electric Machines

2.1 INTRODUCTION

The electromagnetic system is an essential element of all rotating electric

machinery and electromechanical devices as well as static devices like the
transformer. The role of the electromagnetic system is to establish and control
electromagnetic fields for carrying out conversion of energy, its processing and
transfer. Practically all electric motors and generators, ranging in size from fractional horsepower units
found in domestic appliances to the gigantic several thousand kW motors employed in heavy industry and
several hundred megawatt generators installed in modern generating stations, depend upon the magnetic
field as the coupling medium allowing interchange of energy in either direction between electrical and
mechanical systems. A transformer though not an electromechanical conversion device, provides a means
of transferring electrical energy between two electrical ports via the medium of a magnetic field. Further,
transformer analysis runs parallel to rotating machine analysis and greatly aids in understanding the latter.
It is, therefore, seen that all electric machines including transformers use the medium of magnetic field for
energy conversion and transfer. The study of these devices essentially involves electric and magnetic circuit
analysis and their interaction. Also, several other essential devices like relays, circuit breakers, etc. need the
presence of a confined magnetic field for their operation.
The purpose of this chapter is to review the physical laws governing magnetic fields, induction of emf
and production of mechanical force, and to develop methods of magnetic-circuit analysis. Simple magnetic
circuits and magnetic materials will be briefly discussed. In the chapters to follow, how the concepts of this
chapter are applied in the analysis of transformers and machines will be shown.

2.2 MAGNETIC CIRCUITS

The exact description of the magnetic field is given by the Maxwell’s equations* and the constitutive
relationship of the medium in which the field is established.

—◊B = 0 and —◊D = r

∂D –12
—¥H = J + and D = e0 E ; e0 = 8.85 ¥ 10
∂t
wherein J = conduction current density and D = displacement current density, negligible for slowly-varying
fields D = e0 E ; e0 = 8.85 ¥ 10–12 F/m).
 Magnetic Circuits and Induction 13

Such description apart from being highly complex is otherwise not necessary for use in electric machines
wherein the fields (magnetic and electric) are slowly varying (fundamental frequency being 50 Hz) so that
the displacement current can be neglected. The magnetic field can then be described by Ampere’s law and is
solely governed by the conduction current. This law is in integral form and is easily derivable from the third
Maxwell’s equation (by ignoring displacement current) by means of well-known results in vector algebra.
The Ampere’s law is reproduced as follows:

Ú J ◊ ds = Ú H ◊ dl
s
(2.1)

wherein J = conduction current density

H = magnetic field intensity
s = the surface enclosed by the closed path of length l
ds = differential surface
dl = differential length
Consider the example of a simple electromagnetic system comprising an exciting coil and ferromagnetic
core as shown in Fig. 2.1. The coil has N turns and carries a constant (dc) current of i A. The magnetic field is
established in the space wherein most of the total magnetic flux set up is confined to the ferromagnetic core
for reasons which will soon become obvious. Consider the flux path through the core (shown dotted) which in
fact is the mean path of the core flux. The total current piercing the surface enclosed by this path is as follows:

Ú J ◊ ds = Ni
s
Hc

i
Core mean
+ length lc
(Ferromagnetic)
e N
Leakage fiux

–
ca b
Exciting coil

Fig. 2.1 A simple magnetic system

Hence Eq. (2.1) acquires the form

Ni =
Ú H ◊ dl (2.2)
l
Since N is the number of coil turns and i the exciting current in amperes, the product F = Ni has the units of
ampere-turns (AT) and is the cause of establishment of the magnetic field. It is known as the magnetomotive
force (mmf ) in analogy to the electromotive force (emf ) which establishes current in an electric circuit.
14 Electric Machines

The magnetic field intensity H causes a flux density B to be set up at every point along the flux path which
is given by
B = mH = m0 mrH (for flux path in core) (2.3a)
and B = m0H (for flux path in air) (2.3b)
The units of flux density are weber (Wb )/m2 called tesla (T). The term m0 is the absolute permeability of
free space and has a value of
m0 = 4p ¥ 10–7 henry (H)/m
The permeability m = m0 mr of a material medium is different from m0 because of a certain phenomenon
occurring in the material. The term mr is referred to as relative permeability of a material and is in the range of
2000-6000 for ferromagnetic materials (see Sec. 2.3). It is, therefore, seen that for a given H, the flux density
B and, therefore, the flux over a given area

f=
Ú B◊ds
s
will be far larger in the magnetic core in Fig. 2.1 than in the air paths. Hence, it is safe to assume that the
magnetic flux set up by mmf Ni is mainly confined to the ferromagnetic core and the flux set up in air paths
is of negligible value. The flux set up in air paths is known as the leakage flux as if it leaks through the core;
some of the leakage flux paths are shown chain-dotted in Fig. 2.1. There is no way to avoid magnetic leakage
as there are no magnetic insulators in contrast to electric insulators which confine the electric current to the
conductor for all practical purposes. The effect of the leakage flux is incorporated in machine models through
the concept of the leakage inductance.
The direction of field intensity is H and so the direction of flux f is determined from the Right Hand Rule
(RHR). It is stated as:
Imagine that you are holding a current carrying conductor in your right hand with the thumb pointing in
the direction of current. Then the direction in which the fingers curl gives the direction of flux. In case of a
coil you imagine that you are grasping the coil in right hand with the thumb in the direction of current; then
the fingers curl in the direction of flux.
The reader may apply RHR to the exciting coil in Fig. 2.1 to verify the direction of flux as shown in the
figure.
The magnetic field intensity H is tangential to a flux line all along its path, so that the closed vector
integration in Eq. (2.2) along a flux-line reduces to closed scalar integration, i.e.

Ni = Ú H · dl (2.4)
l
With the assumption of negligible leakage flux, the flux piercing the core cross-section at any point
remains constant. Further, from the consideration of symmetry it immediately follows that the flux density
over straight parts of the core is uniform at each cross-section and remains constant along the length; such
that H is constant along the straight parts of the core. Around the corners, flux lines have different path lengths
between magnetic equipotential planes (typical ones being ab and cd shown in Fig. 2.1) so that H varies from
a high value along inner paths to a low value along outer paths. It is reasonable to assume that H shown dotted
along the mean path will have the same value as in straight parts of the core (this mean path technique renders
simple the analysis of magnetic circuits of machines and transformers).
 Magnetic Circuits and Induction 15

It has been seen previously that the magnetic field intensity along the mean flux path in the core can be
regarded constant at Hc. It then follows from Eq. (2.4) that
F = Ni = Hc lc (2.5)
where F = mmf in AT and lc = mean core length (m)
From Eq. (2.5)
Ni
Hc = AT/m (2.6)
lc
If one now imagines that the exciting current i varies with time, Eq. (2.6) would indicate that Hc will vary
in unison with it. Such fields are known as quasi-static fields in which the field pattern in space is fixed but
the field intensity at every point varies as a replica of the time variation of current. This simplified field picture
is a consequence of negligible displacement current in slowly-varying fields as mentioned earlier. In a quasi-
static field, the field pattern and field strength at a particular value of time-varying exciting current will be the
same as with a direct current of that value. In other words, a field problem can be solved with dc excitation
and then any time variation can be imparted to it.
Now, the core flux density is given by,
Bc = mc Hc tesla (T)
and core flux (assumed to be total flux) is given by,
f= Ú B◊ds = Bc Ac Wb
s
where Ac = cross-sectional area of core and flux in the limbs is oriented normal to cross-sectional area. Then
from Eq. (2.6)
Ni F
f = mc HcAc = or f= = FP (2.7)
Ê lc ˆ R
ÁË m A ˜¯
c c
F l
where R= = c = reluctance* of the magnetic circuit (AT/Wb) (2.8)
f mc Ac
and P = 1/R = permeance of the magnetic circuit. It is, therefore, seen that by certain simplifying assumptions
and field symmetries, it has been possible to lump the distributed magnetic system into a lumped magnetic
circuit described by Eq. (2.7) which is analogous to Ohm’s law in dc circuits. The electrical circuit analog
of the magnetic system (now reduced to a magnetic circuit) is shown in
f (∼ i)
Fig. 2.2 wherein F (mmf ) is analogous to E (emf ), R (reluctance) is
analogous to R (resistance) and f (flux) is analogous to i (current).
F (∼ E) R (∼ R)
The analogy though useful is, however, not complete; there being two
points of difference: (i) magnetic reluctance is nondissipative of energy
unlike electric resistance, (ii) when F is time-varying, the magnetic Fig. 2.2 Electrical analog of the
circuit still remains resistive as in Fig. 2.2, while inductive effects are simple magnetic circuit of
bound to appear in an electric circuit. This is because there is no time- Fig. 2.1
lag between the exciting current and the establishment of magnetic flux
(quasi-static field).

* Unit of reluctance is AT/Wb and will not be specified every time in examples.
16 Electric Machines

The lumped magnetic circuit and its electrical analog are useful concepts provided the permeability (m) of
the core material and, therefore, the core reluctance is constant as is tacitly assumed above. This, however,
is not the case with ferromagnetic materials, but when air-gaps are involved, the assumption of constant
reluctance is generally valid and leads to considerable simplicity in magnetic circuit analysis.
In more complicated magnetic circuits—with multiple excitations and series-parallel core arrangement—
the general theorems of electric circuits apply, i.e. Kirchhoff’s voltage (mmf ) law and Kirchhoff’s current
(flux) law. This is illustrated in Example 2.3.

B-H Relationship (Magnetization Characteristic)

In free space (also nonmagnetic materials), the permeability m0 is constant so that B-H relationship is linear.
This, however, is not the case with ferromagnetic materials used in electric machines, wherein the B-H
relationship is strictly nonlinear in two respects—
(T )
saturation and hysteresis. Hysteresis non-linearity B
Saturation zone
is the double valued B-H relationship exhibited in
cyclic variation of H (i.e. exciting current). This
nonlinearity is usually ignored in magnetic circuit
calculations and is important only when current
wave shape and power loss are to be accounted
Linear zone
for. This is discussed in Sections 2.3 and 2.6. A (constant m)
typical normal B-H relationship (magnetization
characteristic) for ferromagnetic materials is shown
in Fig. 2.3. It has an initial nonlinear zone, a middle lnitial nonlinear zone
0
almost linear zone and a final saturation zone in H(AT/ m)

which B progressively increases less rapidly with H Fig. 2.3 Typical normal magnetization curve of
compared to the linear zone. In the deep saturation ferromagnetic material
zone, the material behaves like free space.
Due to considerations dictated by economy, electric machines and transformers are designed such that
the magnetic material is slightly saturated (i.e. somewhat above the linear zone). In exact magnetic circuit
calculations the nonlinear magnetization curve has to be used necessitating graphical/numerical solutions.

Core with Air-gap

Transformers are wound on closed cores as in Fig. 2.1. Rotating machines have a moving element and must
therefore have air-gaps in the cores out of necessity. A typical magnetic circuit with an air-gap is shown in
Fig. 2.4. It is assumed that the air-gap is narrow and the flux coming out of the core passes straight down the
air-gap such that the flux density in the air-gap is the same as in the core. Actually as will soon be seen, that
the flux in the gap fringes out so that the gap flux density is somewhat less than that of the core. Further, let
the core permeability mc be regarded as constant (linear magnetization characteristic).
The mmf Ni is now consumed in the core plus the air-gap. From the circuit model of Fig. 2.4(b) or directly
from Fig. 2.4(a)
Ni = Hc lc + Hglg (2.9a)
Bc Bg
or Ni = lc + lg (2.9b)
mc m0
 Magnetic Circuits and Induction 17

Hc

Mean core length, lc

i
Cross-sectional
+ area A

e Air-gap; Hg, lg
N
Rc f
_

Ni Rg

(a) (b)

Fig. 2.4 A typical magnetic circuit with air-gap and its equivalent electric circuit

Assuming that all the core flux passes straight down the air-gap (it means no fringing (see Fig. 2.5))
Bg = Bc
\ f = Bc A = Bg A (2.10)
Substituting Eq. (2.10) in Eq. (2.9b)
Ê l ˆ Ê lg ˆ
Ni = f Á c ˜ + f Á (2.11)
Ë mc A ¯ Ë m0 A ˜¯
Recognizing various quantities in Eq. (2.11)
F = f (Rc + Rg) = fReq (2.12)
lc
where Rc = = core reluctance
mc A
lg
Rg = = air-gap reluctance
m0 A
From Eq. (2.12)

F F /R
g
f= = (2.13)
Rc + Rg 1 + Rc / Rg
R
But c
= 0 lc 1
Rg clg
because mc is 2000 to 6000 times m0 in ferromagnetic materials. The permeability effect predominates the
usual core and air-gap dimensions even though lc lg. It then follows from Eq. (2.13), that
f ª F/Rg (2.14)
which means that in a magnetic circuit with air-gap(s), core reluctance may be neglected with no significant
loss of accuracy. This assumption will be generally made in modelling rotating machines. The effect of core
saturation (reduction of core permeability) will be introduced as a correction wherever greater accuracy is
desired.
18 Electric Machines

Magnetic Circuit Calculations

Normally magnetic circuit calculations involve two types of problems. In the first type of problem it is
required to determine the excitation (mmf ) needed to establish a desired flux or flux density at a given point
in a magnetic circuit. This is the normal case in designing electromechanical devices and is a straight forward
problem. In the second category, the flux (or flux density) is unknown and is required to be determined for a
given geometry of the magnetic circuit and specified mmf. This kind of problem arises in magnetic amplifiers
wherein this resultant flux is required to be determined owing to the given excitation on one or more control
windings. A little thought will reveal that there is no direct analytical solution to this problem because of the
non-linear B-H characteristic of the magnetic material. Graphical/numerical techniques have to be used in
obtaining the solution of this problem.

Leakage Flux
In all practical magnetic circuits, most of the flux is confined to the intended path by use of magnetic cores but
a small amount of flux always leaks through the surrounding air. This stray flux as already stated is called the
leakage flux, Leakage is characteristic of all magnetic circuits and can never be fully eliminated. Calculations
concerning the main magnetic circuit are usually carried out with the effect of leakage flux either ignored or
empirically accounted for. Special studies of leakage must be made for ac machines and transformers since
their performance is affected by it.

Fringing
At an air-gap in a magnetic core, the flux fringes out into neighbouring air paths as shown in Fig. 2.5;
these being of reluctance comparable to that of the gap. The result is nonuniform flux density in the air-
gap (decreasing outward), enlargement of the effective air-
Fringing flux
gap area and a decrease in the average gap flux density. The
fringing effect also disturbs the core flux pattern to some
depth near the gap. The effect of fringing increases with the Air
air-gap length. Corrections for fringing in short gaps (as used Core gap Core
in machines) are empirically made by adding one gap length
to each of the two dimensions making up its area. For the
example of the core with the air-gap previously presented, the Fig. 2.5 Flux fringing at air-gap
gap reluctance would now be given by
lg
Rg =
m0 Ag
which will be less than the previous value as Ag > A.
It can be shown theoretically that the magnetic flux leaves and enters the surface of an infinitely permeable
material normally. This will be nearly so in ferromagnetic materials which have high permeability. In electric
machines a small amount of the tangential flux component present at iron surfaces will be neglected.

Stacking Factor
Magnetic cores are made up of thin, lightly insulated (coating of varnish) laminations to reduce power loss
in cores due to the eddy-current phenomenon (explained in Sec. 2.6). As a result, the net cross-sectional area
 Magnetic Circuits and Induction 19

of the core occupied by the magnetic material is less than its gross cross-section; their ratio (less than unity)
being known as the stacking factor. Depending upon the thickness of laminations, stacking factor may vary
from 0.5–0.95, approaching unity as the lamination thickness increases.

EXAMPLE 2.1 The magnetic circuit of Fig. 2.4(a) has dimensions: Ac = 4 ¥ 4 cm2, lg = 0.06 cm, lc =
40 cm; N = 600 turns. Assume the value of mr = 6000 for iron. Find the exciting current for Bc = 1.2 T and
the corresponding flux and flux linkages.

SOLUTION From Eq. (2.9), the ampere-turns for the circuit are given by
Bc Bg
Ni = lc + lg (i)
m0 mr m0
Neglecting fringing
Ac = Ag therefore Bc = Bg
Bc Ê lc ˆ
Then i= + lg ˜
m0 N ÁË mr ¯
1.2 Ê 40 ˆ
= Á + 0.06˜ ¥ 10–2 (ii)
4 p ¥ 10- 7 ¥ 600 Ë 6000 ¯
= 1.06 A
Ê 2/3ˆ
The reader should note that the reluctance of the iron path of 40 cm is only Á = 0.11 of the reluctance of the
Ë 6 ˜¯
0.06 cm air-gap.
f = Bc Ac = 1.2 ¥ 16 ¥ 10–4 = 19.2 ¥ 10–4 Wb
Flux linkages, l = Nf = 600 ¥ 19.2 ¥ 10–4 = 1.152 Wb-turns
If fringing is to be taken into account, one gap length is added to each dimension of the air-gap constituting the area.
Then
Ag = (4 + 0.06) (4 + 0.06) = 16.484 cm2
Effective Ag > Ac reduces the air-gap reluctance. Now

19.2 ¥ 10 - 4
Bg = = 1.165 T
16.484 ¥ 10 - 4
From Eq. (i)
1 Ê Bc lc ˆ
i= + Bg l g ˜ (iii)
m0 N ËÁ mr ¯

1 Ê 1.2 ¥ 40 ¥ 10 - 2 ˆ
= -7 Á + 1.165 ¥ 0.06 ¥ 10 - 2 ˜
4 p ¥ 10 ¥ 600 Ë 6000 ¯
= 1.0332 A

EXAMPLE 2.2 A wrought iron bar 30 cm long and 2 cm in diameter is bent into a circular shape as
shown in Fig. 2.6. It is then wound with 600 turns of wire. Calculate the current required to produce a flux
of 0.5 mWb in the magnetic circuit in the following cases:
(i) no air-gap;
(ii) with an air-gap of 1 mm; mr (iron) = 4000 (assumed constant); and
20 Electric Machines

(iii) with an air-gap of 1 mm; assume the following data for the magnetization of iron:
H in AT/m 2500 3000 3500 4000
B in T 1.55 1.59 1.6 1.615

SOLUTION
(i) No air-gap
30 ¥ 10 - 2
Rc = = 1.9 ¥ 105 30 cm, core length
4000 ¥ 4 p ¥ 10 - 7 ¥ p ¥ 10- 4 i

Ni = fRc
N = 600 1 mm
0.5 ¥ 10 - 3 ¥ 1.9 ¥ 105
or i = fRc /N = = 0.158 A
600 –4 2
Ac = p ¥ 10 m
(ii) Air-gap = 1 mm, mr (iron) = 4000
Rc = 1.9 ¥ 105 (as in part (i)
1 ¥ 10- 3
Rg = = 25.33 ¥ 105 Fig. 2.6
4 p ¥ 10- 7 ¥ p ¥ 10- 4
R(total) = Rc + Rg = 27.1 ¥ 105

0.5 ¥ 10 - 3 ¥ 27.1 ¥ 105

\ i= = 2.258 A
600
(iii) Air-gap = 1 mm; B-H data as given
0.5 ¥ 10 -3
Bc = Bg = = 1.59 T (fringing neglected)
p ¥ 10- 4
Bg 1.59
Hg = =
m0 4 p ¥ 10- 7
1.59 ¥ 1 ¥ 10- 3
ATg = Hg lg = = 1265
4 p ¥ 10- 7
From the given magnetization data (at Bc = 1.59 T),
Hc = 3000 AT/m
ATc = Hc lc = 3000 ¥ 30 ¥ 10–2 = 900
AT (total) = ATc + ATg
= 900 + 1265 = 2165
2165
i= = 3.61 A
600

EXAMPLE 2.3 The magnetic circuit of Fig. 2.7 has cast steel core with dimensions as shown:
Mean length from A to B through either outer limb = 0.5 m
Mean length from A to B through the central limb = 0.2 m
In the magnetic circuit shown it is required to establish a flux of 0.75 mWb in the air-gap of the central
limb. Determine the mmf of the exciting coil if for the core material (a) mr = (b) mr = 5000. Neglect
fringing.
 Magnetic Circuits and Induction 21

Mean flux path

f2 f1

1 ¥ 1 cm
i f

N
0.02 cm
2 cm 2
0.02 cm 0.025 cm
3 1
B
1 ¥ 1 cm

Fig. 2.7

SOLUTION
(a) mr = , i.e. there are no mmf drops in the magnetic core. It is easy to see from Fig. 2.7 that the two outer limbs
present a parallel magnetic circuit. The electrical analog of the magnetic circuit is drawn in Fig. 2.8(a). Various
gap reluctances are:
0.025 ¥ 10 - 2
Rg1 = = 1.99 ¥ 106
4 p ¥ 10- 7 ¥ 1 ¥ 10- 4
0.02 ¥ 10 - 2
Rg2 = 1.592 ¥ 106
4 p ¥ 10- 7 ¥ 1 ¥ 10- 4
0.02 ¥ 10 - 2
Rg3 = = 0.796 ¥ 106
4 p ¥ 10- 7 ¥ 2 ¥ 10- 4
From Fig. 2.8(b),
Ni = 0.75 ¥ 10–3 (Rg3 + Rg1 || Rg2)
= 0.75 ¥ 10–3 (0.796 + 0.844) ¥ 106
= 1230 AT

f = 0.75 mWb

0.75 mWb
N Ni
Rg Rg R g || R g
2 1 1 2
Rg Rg
3 3

(a) (b)

Fig. 2.8 Electrical analog of Fig. 2.7

(b) mr = 5000. This means that the reluctance of magnetic core must be taken into consideration. The analogous
electric circuit now becomes that of Fig. 2.9.
22 Electric Machines

Since gap lengths are negligible compared to core lengths, various core reluctances can be calculated as follows:

0.5
Rc1 = = 0.796 ¥ 106
4 p ¥ 10- 7 ¥ 5000 ¥ 1 ¥ 10- 4
0.75 mWb
Rc2 = Rc1 = 0.796 ¥ 106 R c2
Ni
R c1
0.2 6
R c3
Rc3 = = 0.159 ¥ 10
4 p ¥ 10- 7 ¥ 5000 ¥ 2 ¥ 10- 4 R g2 R g1
R g3
The equivalent reluctance is
Req = (Rc1 + Rg1) || (Rc2 + Rg2) + Rc3 + Rg3)
Fig. 2.9
27.86 ¥ 23.86
= ¥ 106 + 0.955 ¥ 106 = 1.955 ¥ 106
51.72
Now Ni = fReq
= 0.75 ¥ 10–3 ¥ 1.955 ¥ 106
= 1466 AT

EXAMPLE 2.4 The magnetic circuit of Fig. 2.10 has cast steel core. The cross-sectional area of the
central limb is 800 mm2 and that of each outer limb is 600 mm2. Calculate the exciting current needed
to set up a flux of 0.8 mWb in the air gap. Neglect magnetic leakage and fringing. The magnetization
characteristic of cast steel is given in Fig. 2.16.

1 mm

400 400
mm 500 turns mm
160 mm

Fig. 2.10

SOLUTION
0.8 10- 3 1
Air gap Bg = ¥ = 1 T and Hg = AT/m
800 10- 6 4 p ¥ 10- 7
1
Fg = ¥ 1 ¥ 10–3 = 796 AT
4 p ¥ 10- 7
Central limb Bc = Bg = 1 T
From Fig. 2.16 Hc = 1000 AT/m
Fc = 1000 ¥ 160 ¥ 10–3 = 160 AT
 Magnetic Circuits and Induction 23

Because of symmetry, flux divides equally between the two outer limbs. So
f (outer limb) = 0.8/2 = 0.4 mWb
0.4 ¥ 10 - 3
B (outer limb) = = 0.667 AT
600 ¥ 10- 6
F (outer limb) = 375 ¥ 400 ¥ 10–3 = 150 AT
F (total) = 796 + 160 + 150 = 1106 AT
Exciting current = 1106/500 = 2.21 A

EXAMPLE 2.5 The magnetic circuit of Fig. 2.11 has a cast steel core whose dimensions are given below:
Length (ab + cd) = 50 cm Cross-sectional area = 25 cm 2
Length ad = 20 cm Cross-sectional area = 12.5 cm2
Length dea = 50 cm Cross-sectional area = 25 cm2
Determine the exciting coil mmf required to establish an air-gap flux of 0.75 m Wb. Use the B-H curve
of Fig. 2.16.

0.25 cm
e
F c

Fig. 2.11

SOLUTION Assuming no fringing the flux density in the path abcd will be same, i.e.

0.75 ¥ 10 - 3
B= = 0.3 T
25 ¥ 10 - 4
B 0.3 ¥ 0 .25 ¥ 10 - 3
Fbc = lbc = = 60 AT
m0 4 p ¥ 10- 7
Hab = Hcd (from Fig. 2.16 for cast steel for B = 0.3 T) = 200 AT/m
Fab+cd = 200 ¥ 50 ¥ 10–2 = 100 AT
\ Fad = 60 + 100 = 160 AT
160
Had = = 800 AT/m
20 ¥ 10 - 2
24 Electric Machines

Bad (from Fig. 2.16) = 1.04 T

fad = 1.04 ¥ 12.5 ¥ 10–4 = 1.3 mWb
fdea = 0.75 + 1.3 = 2.05 mWb
2.05 ¥ 10 - 3
Bdea = = 0.82 T
25 ¥ 10 - 4
Hdea (from Fig. 2.16) = 500 A T/m
Fdea = 500 ¥ 50 ¥ 10–2 = 250 AT
F = Fdea + Fad = 250 + 160 = 410 AT

EXAMPLE 2.6 A cast steel ring has a circular cross-section of 3 cm in diameter and a mean circumference
of 80 cm. A 1 mm air-gap is cut out in the ring which is wound with a coil of 600 turns.
(a) Estimate the current required to establish a flux of 0.75 mWb in the air-gap. Neglect fringing and
leakage.
(b) What is the flux produced in the air-gap if the exciting current is 2 A? Neglect fringing and leakage.
Magnetization data:
H (AT/m) 200 400 600 800 1000 1200 1400 1600 1800 2020
B (T) 0.10 0.32 0.60 0.90 1.08 1.18 1.27 1.32 1.36 1.40

SOLUTION
f = 0.75 ¥ 10–3 Wb 1.6
1.5
0.75 ¥ 10 - 3 1.4
Bg = f/A = 2 =1.06 T
Ê 0.03 ˆ
p ¥Á 1.2
Ë 2 ˜¯
1.0
Bc = Bg (no fringing)
B(T)

Bc
Reading from the B-H curve drawn in Fig. 2.12, 0.8 Ni = HcIc + I
Bc m0 g
Hc = 900 AT/m 0.6

lc = 0.8 m (air-gap length can be neglected) 0.4

ATc = Hc lc = 900 ¥ 0.8 = 720 0.2
1.06
ATg = ¥ 10–3 = 843
4 p ¥ 10- 7
0
0 400 Hc 800 1200 1600 2000
Ni = ATc + ATg = 720 + 843 = 1563 H in AT/m
1563 Fig. 2.12
Therefore i= = 2.6 A
600

(b) The excitation is now given and the flux is to be determined from the B-H curve given. The problem must,
therefore, be solved numerically/graphically. It is solved here graphically. Now
B2
Ni = l + Hc lc; (Bg = Bc) (i)
m0 g
This is a linear equation in Bc and Hc; the second equation is the nonlinear B-H curve. The intersection of the
two for a given Ni will yield the solution. For this problem
Ni = 600 ¥ 2 = 1200 AT
 Magnetic Circuits and Induction 25

Substituting various values in Eq. (i)

Bc
1200 = ¥ 10–3 + 0.8 Hc (ii)
4 p ¥ 10- 7
This equation is plotted in Fig. 2.12, by locating the points
Hc = 0, Bc = 1.5
Bc = 0, Hc = 1500
The intersection gives the result
Bc = 0.78 T
p
f = BcA = 0.78 ¥ (0.03)2 = 0.55 mWb
4

2.3 MAGNETIC MATERIALS AND THEIR PROPERTIES

From the magnetic point of view* a material is classified according to the nature of its relative permeability
(mr). All nonmagnetic materials are classified as paramagnetic, mr slightly greater than 1, and diamagnetic,
mr slightly less than 1. For all practical purposes, mr of these materials can be regarded as unity, i.e. their
magnetic properties are very much similar to that of free space. Such materials are not of interest to us in this
treatise.
Materials which are of interest to us are those whose relative permeability is much higher than that of free
space. These can be classified as ferromagnetic and ferrimagnetic. Ferromagnetic materials can be further
subdivided as hard and soft. Hard ferromagnetic materials include permanent magnet materials, such as
alnicos, chromium steels, certain copper-nickel alloys and several other metal alloys. Soft ferromagnetic
materials are iron and its alloys with nickel, cobalt, tungsten and aluminium. Silicon steels and cast steels
are the most important ferromagnetic materials for use in transformers and electric machines. Ferrimagnetic
materials are the ferrites and are composed of iron oxides—MeO. Fe2O3, where Me represents a metallic
ion. Ferrites are also subgrouped as hard (permanent magnetic) and soft (nickel-zinc and manganese-zinc)
ferrites. Soft ferrites are quite useful in high frequency transformers, microwave devices, and other similar
high-frequency operations. There is a third category of magnetic materials, known as superparamagnetic,
made from powdered iron or other magnetic particles. These materials are used in transformers for electronics
and cores for inductors. Permalloy (molybdenum-nickel-iron powder) is the best known example of this
important category of magnetic materials.

Properties of Magnetic Materials

Magnetic materials are characterized by high permeability and the nonlinear B-H relationship which
exhibits both saturation and hysteresis. The physics of these properties is explained by the domain theory of
magnetization usually taught in junior level courses.
The B-H relationship for cyclic H is the hysteresis loop shown in Fig. 2.13 for two values of maximum
flux density. It is easily observed from this figure that B is a symmetrical two-valued function of H; at any
given H, B is higher if H is reducing compared to when H is increasing. This is the basic hysteresis property
in which B lags behind H. It can also be recognized as a memory-type non-linearity in which the material

* For the theory of magnetization based on atomic structure of materials a suitable book on material science may be
consulted.
 Analysis of
AC Circuits

�3.1�
We have discussed the network theorems with reference to resistive load and dc sources. Now, all the
theorems will be discussed when a network consists of ac sources, resistors, inductors and capacitors. All the
theorems are also valid for ac sources.

�3.2�
Mesh analysis is useful if a network has a large number of voltage sources. In this method, currents are
assigned in each mesh. We can write mesh equations by Kirchhoff’s voltage law in terms of unknown mesh
currents,

� � Find mesh currents I1 and I2 in the network of Fig. 3.1.

3 j4

100�45 V j 10 j 10
I1 I2

�
Solution Applying KVL to Mesh 1,
100�45° − (3 j4)I1 − j10(I1 − I2) 0
(3 j14)I1 − j10I2 100 �45° …(i)
Applying KVL to Mesh 2,
−j10 (I2 − I1) j10 (I2) 0
j10I1 0 …(ii)
I1 0
Substituting I1 in Eq. (i),
j10 I 2 100 5
100 45
I2 10 135 A
j10
3.2�Circuit Theory and Networks—Analysis and Synthesis

� � Find mesh current I1, I2 and I3 in the network of Fig. 3.2.

5 j2 j5

2
10�30 V 3 5
I1 I2 I3 j2

Fig. 3.2

Solution Applying KVL to Mesh 1,

10 �30° − (5 − j2) I1 − 3 (I1 − I2) 0
(8 − j2) I1 − 3I2 10�30° … (i)
Applying KVL to Mesh 2,
−3 (I2 − I1) − j5I2 − 5 (I2 − I3) 0
−3I1 (8 j5) I2 − 5I3 0 … (ii)
Applying KVL to Mesh 3,
–5 (I3 – I2) – (2 – j2) I3 0
–5I2 + (7 – j2) I3 0 … (iii)
Writing Eqs (i), (ii) and (iii) in matrix form,
8 j2 3 0 I1 10 30
3 8 j5 5 I2 0
0 5 7 j2 I3 0
By Cramer’s rule,
10 30 3 0
0 8 j5 5
0 5 7 j2
I1 1.43
433 38 7 A
8 j2 3 0
3 8 j5 5
0 5 7 j2
8 j 2 10 30 0
3 0 5
0 0 7 j2
I2 0.693 2.2 A
8 j2 3 0
3 8 j5 5
0 5 7 j2
8 j2 3 10 30
3 8 j5 0
0 5 0
I3 0.476 13.8 A
8 j2 3 0
3 8 j5 5
0 5 7 j2
 3.2�Mesh Analysis

� In the network of Fig. 3.3, find the value of V2 so that the current through (2 + j3)
ohm impedance is zero.
5 2 j3 4

30�0 V j5 6 V2
I1 I2 I3

Applying KVL to Mesh 1,

30�0° − 5I1 − j5 (I1 − I2) 0
(5 j5) I1 − j5I2 30 �0° …(i)
Applying KVL to Mesh 2,
−j5 (I2 − I1) − (2 j3) I2 − 6 (I2 − I3) 0
−j5I1 (8 j8) I2 − 6I3 0 …(ii)
Applying KVL to Mesh 3,
−6(I3 − I2) − 4I3 − V2 0
−6I2 10I3 −V2 …(iii)
Writing Eqs (i), (ii) and (iii) in matrix form,
5 j5 j5 0 I1 30 0
j 5 8 j8 6 I 2 0
0 6 10 I3 V2
By Cramer’s rule,
5 j 5 30 0 0
j5 0 6
0 2 1 0
I2 0
5 j5 j5 0
j 5 8 j8 6
0 6 10
( )( ) ( ))(( ) 0
j1500
V2 35.36 45 V
30 j 30

� Find the value of the current I3 in the network shown in Fig. 3.4.
j4 10�30 V
4

20�0 V j 10 20
I1 I2

10 4 j4

I3

20

�
 Circuit Theory and Networks—Analysis and Synthesis

Applying KVL to Mesh 1,

20�0° − (4 − j4) I1 − j10 (I1 − I2) − 10 (I1 − I3) 0
(14 j6) I1 − j10I2 − 10I3 20 �0° … (i)
Applying KVL to Mesh 2,
−j10 (I2 − I1) − 10�30° − 20I2 − (4 − j4) (I2 − I3) 0
−j10I1 (24 j6) I2 − (4 − j4) I3 −10�30° … (ii)
Applying KVL to Mesh 3,
−10(I3 − I1) − (4 − j4) (I3 − I2) − 20I3 0
−10I1 − (4 − j4) I2 (34 − j4) I3 0 … (iii)
Writing Eqs (i), (ii) and (iii) in matrix form,
14 j6 j10 10 I1 20 0
j10 24 j6 (4 j 4) I 2 10 30
10 (4 j 4)
j4 34 j 4 I3 0
By Cramer’s rule,
14 j6 j10 20 0
j10 24 j6 10 30
10 (4 j 4) 0
I3 0 44 14 A
14 j 6 10 10
j10 24 j 6 ( 4 j 4)
10 ( 4 j4
j 4) 34 j 4

� Find the voltage VAB in the network of Fig. 3.5.

1 j 50 100

B I2

100 4 j200
96
I1

10�0 V

Applying KVL to Mesh 1,

− 96 I1 − (100 4 j200) (I1 − I2) 10 �0° 0
(200 j200) I1 − (104 j200) I2 10 �0° …(i)
Applying KVL to Mesh 2,
− (1 − j50 − 100) I2 − (100 4 j200) (I2 − I1) 0
− (104 j200) I1 (205 j150) I2 0 …(ii)
Writing Eqs (i) and (ii) in matrix form,

200 j 200 (104 j 200) I1 10 0

(104 j200
j 200) 205 j150 I2 0
 3.2�Mesh Analysis

By Cramer’s rule,

10 0 200
0 205 j150
I1 0 051 10 A
200 (104 200
200 j150
200 200 10 0
j 200 0
I2 045 26 34 A
200 104 j 200
j 200 j150
2

26 34 (4 j 200) ( 0.051 0 0.045

V

� For the network shown in Fig. 3.6, find the voltage across the capacitor.

1 3
I2

j2 1
j2
1

j3 3
j1
5�0 V

Solution Applying KVL to Mesh 1,

0 2) 0
5 0 …(i)
Applying KVL to Mesh 2,
2

2I 3 0 …(ii)
Applying KVL to Mesh 3,
) j2 0
) 0 …(iii)
Writing Eqs (i), (ii) and (iii) in matrix form,
2 j3 I1 5
2 5 2 I2 0
3 2 2 j2 I 0
 Circuit Theory and Networks—Analysis and Synthesis

By Cramer’s rule,

5 5 1
2 0 j2
3 0 2 2
I2 130 51 A
2 j3
2 5 2
( j2
2 5 0
2 5j2 0
2 0
I3 0 91 A
2 j3
2 5 2
2
V 2 0 91 . 130 51 3 03 . V

� Find the voltage across the 2 � resistor in the network of Fig. 3.7.

3 j

2
2 30 A 8 45 V
I1 I2 �
j

Solution For Mesh 1,

I1 � 2 0 …(i)
Applying KVL to Mesh 2,

)( ) j 45 � 0
� 45 …(ii)
Substituting I1 in Eq. (i),

8 45

I2 � � 9 65 A
2 1
V 2 2 0 3 19 65 78 84 37 V
 3.2�Mesh Analysis

� Find the current through 3 � resistor in the network of Fig. 3.8.

3 5

j2

� 1 1 0 A j1
I1 I2 I3
10 0 V

Solution Applying KVL to Mesh 1,

10 �0 j 2I1 3 1 1 (I1 I 2 ) 0
( ) I1 I 2 10 �0 …(i)
Meshes 2 and 3 will form a supermesh.
Writing current equation for the supermesh,
I3 I2 1�0 …(ii)
Applying KVL to the outer path of the supermesh,
1( 2 1 ) 5I 3 j1 3 0
I1 I 2 (5 1) I 3 0 …(iii)
Writing Eqs (i), (ii) and (iii) in matrix form,
4 j2 1 0 I1 10�0
0 1 1 I2 1�0
1 1 (5 j1) I3 0
By Cramer’s rule,
10�0 1 0
1�0 1 1
0 1 (5 1)
I1 2.11 28.01 A
4 j2 1 0
0 1 1
1 1 (5 1)
I3 I1 2.11 28.01 A

� Find the currents I1 and I2 in the network of Fig. 3.9.

2Vx
6
�

9 0 V Vx �3
�j 3
� I1 � I2

�
 Circuit Theory and Networks—Analysis and Synthesis

Solution From Fig. 3.9,

Vx � j 3( 1 2) …(i)
Applying KVL to Mesh 1,
9 0 6 1 j 3 ( I1 I 2 ) � 0
( ) I1 j 3 2 9 0 …(ii)
Applying KVL to Mesh 2,
j3 ( 2 1) 2Vx 3 2 0
j3 jj3
3I1 2 [ jj3
3 ( I1 I 2 )] 3 2 0
j9 (3 jj999)) 2 0 …(iii)
Writing Eqs (ii) and (iii) in matrix form,

6 j3 j3 I1 9 0
�
j9 3 j9 I2 0

By Cramer’s rule,
9 0 j3
0 3 j9
I1 � � 1 3 2.49 A
6 j3 j3
j9 3 j9
6
j3 9 0
j9 0
I2 � � 1.244 15 95 A
6 j3 j3
j9 3 j9

� Find the voltage across the 4 � resistor in the network of Fig. 3.10.

6�30 V j2

Ix j1

2 4
2Ix
I1 I2

Solution From Fig. 3.10,

Ix I1 …(i)
Applying KVL to Mesh 1,
�2 1 6 30 j1( 1 � 2 ) � 2I x 0
�2 1 6 30 j1 1 � j1I 2 � 2 1 0
( ) I1 j1 2 6 30 …(ii)
 3.3�Node Analysis

Applying KVL to Mesh 2,

2 0
�4 0
) 0 …(iii)
Writing Eqs (ii) and (iii) in matrix form,
1 1 I1 30
1 4 I2 0

By Cramer’s rule,
1 6 0
2 1 0
I2 4 . A
1 1
1 �4
V (0 74 � 2. 2 � . 1 V

3.3�
Node analysis uses Kirchhoff’s current law for finding currents and voltages in a network. For ac networks,
Kirchhoff’s current law states that the phasor sum of currents meeting at a point is equal to zero.

� In the network shown in Fig. 3.11, determine Va and Vb.

j6 Va 3 Vb j5

10 0 V j6 j4 j4

Applying KCL at Node a,

0
3
1 1 1 1 10
j j6 3 j6
. V 1 67 90 …(i)
Applying KCL at Node b,
V V
0
1
1 1 1 1
� V 0
3 3 j j1
� . b 0 …(ii)
 Circuit Theory and Networks—Analysis and Synthesis

Adding Eqs (i) and (ii),

� j1.25
25 b 1.67 � 90
1 67 � 90
Vb 1 34 0 V
� j1.25
Substituting Vb in Eq. (i),
0 33 a 0.33(1
(1.34 0 ) 1.67 90
1 73 755 17
Va 5 244 75.17
1 V
0 33

� For the network shown in Fig. 3.12, find the voltages V1 and V2.
5 V1 4 V2 2

50 0 V j2 j2 50 90 V

�
Applying KCL at Node 1,
V1 0 0 V1 V1 V2
0
5 j2 4
1 1 1 1
V1 V2 10 0
5 j2 4 4
(0.45 � j 0.5) V1 0 25 2 10 0 …(i)
Applying KCL at Node 2,
V2 V1 V2 V2 � 50 90
0
4 j2 2
1 1 1 1
� V1 V2 25 90
4 4 � j2 2
�0.25V1 ( .7755 j 0. ) 2 25 90
25 …(ii)
Writing Eqs (i) and (ii) in matrix form,
0 45 j 0.5 0.25 V1 10 0
�0 25
5 0.75 j 0 5 V2 25 90
By Cramer’s rule,
10 0 0.25
j 25 0 75 j 0.5
V1 24.7 7 . 5 V
0 45 j 0.5 0.25
�0 25 0.75 j 0 5
0 45 j 0.5 10 0
�0
0 25 25 90
V2 34 3344 52
5 882 V
0 45 j 0.5 0.25
�0 25 0.75 j 0 5
 3.3�Node Analysis

� Find the voltage VAB in the network of Fig. 3.13.

j5
2 2
A j 10

I
I � 10 0 A 1

3 B j4

Applying KCL at Node 1,

V1 V2 V1
10 �0
2 3 j4
1 1 1
V1 V2 10�0
2 3 j4 2
(0.62 j 0.16 )V1 0 V2 10�0 …(i)
Applying KCL at Node 2,
V2 V1 V2 V2
0
2 j5 j10
1 1 1 1
V1 V2 0
2 2 j5 j10
0.5V1 (0 5 j 0.3)V2 0 …(ii)
Writing Eqs (i) and (ii) in matrix form,
0 62 j 0.16 05 V1 10 �0
05 0 5 j 0 3 V2 0
By Cramer’s rule,
10 �0 0.5
0 0.5 j 0.3
V1 21.88 56 42 V
0 62 j 0.16 05
05 0 5 j0 3
0 62j 0.16 10 �0
05 0
V2 18.7 87.42
4 V
0 62 j 0.16 05
05 0 5 j0 3
VA V2 18.77�87 42 V
V1 �56 4
21.88�
VB ( j4) ( j4) 17 455�93 32 V
3 j4 3 j4
VAB VA VB ( .7�87.. ) ( .45�93.. ) 2.23�34.1 V
 Circuit Theory and Networks—Analysis and Synthesis

� Find the node voltages V1 and V2 in the network of Fig. 3.14.

V1 j1 V2

2 2V2 j2 2 30 A

�
Solution Applying KCL at Node 1,
V1 V1 V2
2V2
2 j1
1 1 1
V1 2 V2 0
2 j1 j1
( .5 j1)
j1)
1) (2 j1)
1) 2 0 …(i)
Applying KCL at Node 2,
V2 V1 V2
2�30
j1 j2
1 1 1
V1 V2 2 30
j1 j1 j2
j1 j05 2 2�30 …(ii)
Writing Eqs (i) and (ii) in matrix form,
0.5 1 ( 2 j1) V1 0
j1 j 0 .5 V2 2�30
By Cramer’s rule,
0 ( 2 1)
2�30 j 0 .5
V1 2.466�130.62 V
0.5 1 ( 2 j1)
j1 j 0.5
0.5 j1 0
j1 2 �30
V2 1.2233�167.49 V
0.. ((22 j1)
j1 j 0.5

� In the network of Fig. 3.15, find the voltage V2 which results in zero current through
4 � resistor.
5 V1 4 V3 2

50 0� V j2 j2 V2

�
 3.3�Node Analysis

Solution Applying KCL at Node 1,

0
5 2 4
1 1 1 1
10 �0
5 j2 4
.45 0 ) . 10 0 …(i)
Applying KCL at Node 3,
V
0
4 2 2
1 1 1 1
0 5V
4 4 j 2
0 5V …(ii)
Writing Eqs (i) and (ii) in matrix form,
5 0 25 V1 10�0
. j V
By Cramer’s rule,
10 0 .25
V 5 05 10 .7 0. 0 125 V
V1
04 5 0 25 5 1 5
j0 5
. 5 10
. 5V 05 0 25
V
5 0.25 0 55 1 5
j0 5

I4 0
4

10 0 7 0. 0.125 25
5 1 5 0 55 15 95
0 25V
V 01 0 )
5 5
V 26.26 2 V
5

� Find the voltage across the capacitor in the network of Fig. 3.16.
12 30 V
V1 V2

2 60 A j1 2 j2

�
 Circuit Theory and Networks—Analysis and Synthesis

Solution Nodes 1 and 2 will form a supernode.

Writing the voltage equation for the supernode,
V1 V2 12�30 …(i)
Applying KCL to the supernode,
V1 V2 V2
2 60
j1 2 j2
( ) ( .5 j 0. ) 2 2�60 …(ii)
Writing Eqs (i) and (ii) in matrix form,

1 1 V1 12�30
j1 0 5 j 0.5 V2 2�60
By Cramer’s rule,
1 12�30
j1 2�60
V2 18.55�157.42 V
1 1
j1 0 5 j 0.5
Vc V2 18.55�157
57.42 V

�
The superposition theorem can be used to analyse an ac network containing more than one source. The
superposition theorem states that in a network containing more than one voltage source or current source,
the total current or voltage in any branch of the network is the phasor sum of currents or voltages produced
in that branch by each source acting separately. As each source is considered, all of the other sources are
replaced by their internal impedances. This theorem is valid only for linear systems.

� Find the current through the 3 j4 ohm impedance.

5 j5

3
50 90� V 50 0� V
j4

�
Solution 5� j5 �

Step I When the 50 �90° V source is acting alone (Fig. 3.18) I

3�
(3j 4)( jj5)
ZT 5 6 35�23 2 50 90 V
3 j9
j4 �
50 �90
IT 7 87�66 8 A
6 35
35��23 2
�
2.86 Basic Electrical Engineering

2.7 STA DELTA T ANS MAT N

When a circuit cannot be simpli�ed by normal series–parallel reduction technique, the
star-delta transformation can be used.
Figure 2.175(a) shows three resistors RA, RB and RC connected in delta.
Figure 2.175(b) shows three resistors R1, R2 and R3 connected in star.

Fig. 2.175 Delta and star networks

These two networks will be electrically equivalent if the resistance as measured

between any pair of terminals is the same in both the arrangements.

2.7.1 Delta to Star Transformation

Referring to delta network shown in Fig. 2.175(a),
the resistance between terminals 1 and 2 = RC || (RA + RB)
R ( R + RB )
= C A (2.4)
RA + RB + RC
Referring to the star network shown in Fig. 2.175(b),
the resistance between terminals 1 and 2 = R1 + R2 (2.5)
Since the two networks are electrically equivalent,
R ( R + RB )
R1 + R2 = C A (2.6)
RA + RB + RC
RA ( RB + RC )
Similarly, R2 + R3 = (2.7)
RA + RB + RC
RB ( RA + RC )
and R3 + R1 = (2.8)
RA + RB + RC
Subtracting Eq. (2.7) from Eq. (2.6),
RB RC - RA RB
R1 – R3 = (2.9)
RA + RB + RC
 2.7 Star-Delta Transformation 2.87

Adding Eq. (2.9) and Eq. (2.8),

RB RC
R1 = (2.10)
RA + RB + RC
RA RC
Similarly, R2 = (2.11)
RA + RB + RC
RA RB
R3 = (2.12)
RA + RB + RC
Thus, star resistor connected to a terminal is equal to the product of the two delta
resistors connected to the same terminal divided by the sum of the delta resistors.

2.7.2 Star to Delta Transformation

Multiplying the above equations,
RA RB RC2
R1 R2 = (2.13)
( RA + RB + RC )2
RA2 RB RC
R2 R3 = (2.14)
( RA + RB + RC )2
RA RB2 RC
R3 R1 = (2.15)
( RA + RB + RC )2
Adding Eqs (2.13), (2.14) and (2.15),
RA RB RC ( RA + RB + RC )
R1 R2 + R2 R3 + R3 R1 =
( R A + RB + RC )2
RA RB RC
=
RA + RB + RC
= RAR1
= RB R2
= RC R3
R1 R2 + R2 R3 + R3 R1
Hence, RA =
R1
R2 R3
= R2 + R3 +
R1
R1 R2 + R2 R3 + R3 R1
RB =
R2
2.88 Basic Electrical Engineering

R3 R1
= R1 + R3 +
R2
R1 R2 + R2 R3 + R3 R1
RC =
R3
R1 R2
= R1 + R2 +
R3
Thus, delta resistor between the two terminals is the sum of two star resistors connected
to the same terminals plus the product of the two resistors divided by the remaining third
star resistor.
Note: When three equal resistors are connected in delta, the equivalent star resistance is
given by
RD RD R
RY = = D
RD + RD + RD 3
or RD = 3RY

Fig. 2.176

Example 1
Convert the star circuit into its equivalent delta circuit.
A

2W

N
W W
B C

ig. 2.177 [May 2015]

Solution Converting the given star network to delta,

2¥6
RA = 2 + 6 + = 11 W
4
6¥4
RB = 6 + 4 + = 22 W
2
Fig. 2.178
 2.7 Star-Delta Transformation 2.89

4¥2
RC = 4 + 2 + = 7.33 W
6

Example 2
Find an equivalent resistance between terminals A and B.

Fig. 2.179

Solution Converting the two delta networks formed by resistors of 4.5 W, 3 W and
7.5 W into equivalent star networks,

Fig. 2.180

4.5 ¥ 7.5
R1 = R6 = = 2.25 W
4.5 + 7.5 + 3
7.5 ¥ 3
R2 = R5 = = 1.5 W
4.5 + 7.5 + 3
4.5 ¥ 3
R3 = R4 = = 0.9 W
4.5 + 7.5 + 3

Fig. 2.181
2.90 Basic Electrical Engineering

Simplifying the network,

Fig. 2.182

RAB = 7.45 W
Example 3
Find an equivalent resistance between terminals A and B.

Fig. 2.183

Solution Redrawing the network,

Fig. 2.184
 2.7 Star-Delta Transformation 2.91

Converting the delta network formed by three resistors of 10 W into an equivalent star
network,

Fig. 2.185

10 ¥ 10 10
R1 = R2 = R3 = = W
10 + 10 + 10 3

Fig. 2.186

Simplifying the network,

Fig. 2.187
RAB = 10 W
2.92 Basic Electrical Engineering

Example 4
Calculate Rxy for the circuit shown in Fig. 2.188.

ig. 2.188 [De 2012]

Solution Redrawing the network.

Fig. 2.189

Converting the delta network formed by resistors of 10 W, 10 W and 45 W into an equivalent star
network,

10 ¥ 10
R1 = = 1.54 W
10 + 10 + 45
10 ¥ 45
R2 = R3 = = 6.92 W
10 + 10 + 45

Fig. 2.190
 2.7 Star-Delta Transformation 2.93

Simplifying the network,

(a)

(b)

(c)
Fig. 2.191
Rxy = 15 W

Example 5
Find an equivalent resistance between terminals A and B.

Fig. 2.192

Solution Converting the star network formed by resistors of 3 W, 4 W and 6 W into an

equivalent delta network,

Fig. 2.193

6¥4
R1 = 6 + 4 + = 18 W
3
6¥3
R2 = 6 + 3 + = 13.5 W
4
2.94 Basic Electrical Engineering

4¥3
R3 = 4 + 3 + =9W
6

Fig. 2.194

Simplifying the network,

Fig. 2.195
RAB = 6 || (1.35 + 0.9)
= 6 || 2.25
= 1.64 W

Example 6
Find an equivalent resistance between terminals A and N by solving outer delta ABC.

Fig. 2.196

Solution Converting outer delta ABC into a star network,

12 ¥ 12
RY = =4W
12 + 12 + 12
 2.7 Star-Delta Transformation 2.95

M N

Fig. 2.197
Simplifying the network,

2 7

1.56

Fig. 2.198
RAN = 1.56 W

Example 7
Find an equivalent resistance terminals between A and B.

Fig. 2.199

Solution The resistors of 2 W and 4 W and the resistors of 4 W and 11 W are connected
in series.
2.96 Basic Electrical Engineering

Fig. 2.200

Converting the two outer delta networks into equivalent star networks,
6¥6
RY1 = =2W
6+6+6
15 ¥ 15
RY2 = =5W
15 + 15 + 15

Fig. 2.201

Simplifying the network,

Fig. 2.202

RAB = 23 W
 2.7 Star-Delta Transformation 2.97

Example 8
Find an equivalent resistance between terminals A and B.

ig. 2.203 [May 2014]

Solution Drawing the resistor of 30 W from outside,

Fig. 2.204

Converting the delta network formed by resistors of 20 W, 25 W and 35 W into an equivalent

star network,

Fig. 2.205

20 ¥ 35
R1 = = 8.75 W
20 + 35 + 25
20 ¥ 25
R2 = = 6.25 W
20 + 35 + 25
2.98 Basic Electrical Engineering

35 ¥ 25
R3 = = 10.94 W
20 + 35 + 25
Redrawing the network,

Fig. 2.206

Simplifying the network,

Fig. 2.207

RAB = 32.36 W
 2.7 Star-Delta Transformation 2.99

Example 9
Find an equivalent resistance between terminals A and B.

Fig. 2.208

Solution The resistors of 5 W and 25 W and the resistors of 10 W and 5 W are connected
in series.

Fig. 2.209

Converting the delta network formed by the resistors of 20 W, 5 W and 15 W into an

equivalent star network,

Fig. 2.210

20 ¥ 5
R1 = = 2.5 W
20 + 5 + 15
20 ¥ 15
R2 = = 7.5 W
20 + 5 + 15
5 ¥ 15
R3 = = 1.875 W
20 + 5 + 15
2.100 Basic Electrical Engineering

Redrawing the network,

(a)

(b)
Fig. 2.211

Converting the delta network formed by the resistors of 3.875 W, 37.5 W and 30 W into
an equivalent star network,

Fig. 2.212

3.875 ¥ 37.5
R4 = = 2.04 W
3.875 + 37.5 + 30
3.875 ¥ 30
R5 = = 1.63 W
3.875 + 37.5 + 30
37.5 ¥ 30
R6 = = 15.76 W
3.875 + 37.5 + 30
 2.7 Star-Delta Transformation 2.101

Simplifying the network,

(a)

(b)

(c)

(d) (e)
Fig. 2.213
RAB = 23.52 W

Example 10
Find an equivalent resistance between terminals A and B.

Fig. 2.214
Solution Converting the star network formed by the
resistors of 3 W, 5 W and 8 W into an equivalent
delta network,
3¥5
R1 = 3 + 5 + = 9.875 W
8
3¥8
R2 = 3 + 8 + = 15.8 W
5
5¥8
R3 = 5 + 8 + = 26.33 W
Fig. 2.215 3
2.102 Basic Electrical Engineering

Fig. 2.216

The resistors of 15.8 W and 5 W and the resistors of 26.33 W and 4 W are connected in
parallel.

Fig. 2.217

Converting the delta network into a star network,

3.8 ¥ 9.875
R4 = = 2.19 W
3.8 + 9.875 + 3.47
3.8 ¥ 3.47
R5 = = 0.77 W
3.8 + 9.875 + 3.47
3.47 ¥ 9.875
R6 = =2W
3.8 + 9.875 + 3.47
 2.7 Star-Delta Transformation 2.103

(a) (b)
Fig. 2.218

Simplifying the network,

(a)

(b)

(c) (d)
Fig. 2.219

RAB = 4.23 W

Example 11
Find an equivalent resistance between terminals A and B.

Fig. 2.220
2.104 Basic Electrical Engineering

Solution Converting the star network formed by the resistors of 3 W, 4 W and 5 W into
an equivalent delta network,

Fig. 2.221

5¥4
R1 = 5 + 4 + = 15.67 W
3
3¥ 4
R2 = 3 + 4 + = 9.4 W
5
5¥3
R3 = 5 + 3 + = 11.75 W
4
Similarly, converting the star network formed by the resistors of 4 W, 6 W and 8 W into
an equivalent delta network,

Fig. 2.222

6¥8
R4 = 6 + 8 + = 26 W
4
4¥8
R5 = 4 + 8 + = 17.33 W
6
6¥4
R6 = 6 + 4 + = 13 W
8
These two delta networks are connected in parallel between points A and B.

Fig. 2.223
 2.7 Star-Delta Transformation 2.105

The resistors of 9.4 W and 17.33 W are in parallel with a short. Hence, the equivalent
resistance of this combination becomes zero.
Simplifying the parallel networks,

Fig. 2.224
RAB = 6.17 || 9.78 = 3.78 W

Example 12
Find the value of current �owing through 6 W resistor.

ig. 2.225 [De 2015]

Solution Converting the parallel combination of the current source of 5 A and the resistor
of 5 W into an equivalent series combination of voltage source and series resistor,
5W 5W 6W

5W 5W 5W

25 V
5W

1W 5V
Fig. 2.226

Converting the delta network formed by three 5 W resistors into an equivalent star
network,

Fig. 2.227
2.106 Basic Electrical Engineering

By series-parallel reduction technique and adding two voltage sources,

Fig. 2.228

Again by series-parallel reduction technique,

13.67 W

3.33 W I 20 V

Fig. 2.229

20
I= = 1.18 A
13.67 + 3.33
I6W = I13.67W = I = 1.18 A
Example 13
Determine the current supplied by the battery.

Fig. 2.230

Solution Converting the delta network formed by resistors of 6 W, 6 W and 6 W into an

equivalent star network,

Fig. 2.231
 2.7 Star-Delta Transformation 2.107

Simplifying the network,

(a)

(b)

(c)
Fig. 2.232

50
I= = 7.14 A
5+2
Example 14
Calculate the value of current �owing through the 10 W resistor.

Fig. 2.233

Solution Between terminals A and B resistors of 8 W and 4 W are connected in series.

Similarly, between terminals F and E, resistors of 17 W and 13 W are connected in series.

Fig. 2.234
2.108 Basic Electrical Engineering

Converting delta ABC and DEF into an equivalent star network,

Fig. 2.235

Simplifying the network,

Fig. 2.236

180
I= =6A
4 + 16 + 10
By current-division rule,
48
I ¢ = I24 W = I10 W = 6 ¥ =4A
24 + 48
Example 15
Determine current �ow through the 20 W resistor in the following circuit in Fig. 2.237.

ig. 2.237 [De 2013]

 2.7 Star-Delta Transformation 2.109

Solution Converting the two outer delta networks into equivalent star networks,
15 ¥ 15
RY1 = =5W
15 + 15 + 15
36 ¥ 36
RY2 = = 12 W
36 + 36 + 36

Fig. 2.238

Simplifying the network,

(a)

(b)

Fig. 2.239
2.110 Basic Electrical Engineering

200
I= = 5.07 A
5 + 22.44 + 12
By current-division rule,
57
I20 W = I37 W = 5.07 ¥ = 3.07 A
57 + 37

Example 16
Find the current supplied by the battery.

Fig. 2.240

Solution Converting the star network formed by resistors of 40 W, 20 W and 50 W into

an equivalent delta network,

Fig. 2.241

40 ¥ 20
R1 = 40 + 20 + = 76 W
50
40 ¥ 50
R2 = 40 + 50 + = 190 W
20
20 ¥ 50
R3 = 20 + 50 + = 95 W
40
 2.7 Star-Delta Transformation 2.111

Fig. 2.242

The resistors of 190 W and 10 W and the resistors of 95 W and 30 W are connected in
parallel.

Fig. 2.243

Simplifying the network,

(a)

(b)

Fig. 2.244

15
I= = 0.542 A
22.67 + 5
2.112 Basic Electrical Engineering

Delta to star transformation Star to delta transformation

RB RC RR
R1 = RA = R2 + R3 + 2 3
RA + RB + RC R1
RA RC R1 R3
R2 = RB = R1 + R3 +
RA + RB + RC R2
RA RB R1 R2
R3 = RC = R1 + R2 +
RA + RB + RC R3

Exercise 2.5
2.1 Find the equivalent resistance between terminals A and B.

Fig. 2.245
[5 W]
2.2 Find the equivalent resistance between terminals A and B.

Fig. 2.246
[25 W]