3.

8 Design of Transformer

3.2 Specifications of a Transformer
The transformer specifications are their rating and performance expectations. A designer
needs to know the following important specifications with regards to the design of the
transformer.
1. Rating/Capacity/Output-in kVA
2. Voltage–Voltage rating of HV and LV windings with or without tap changers and
tappings
3. Frequency – f in Hz
4. Number of phases – one or three
5. Type of winding connection in case of three-phase transformers – star-star, star-delta,
delta-delta, delta-star with or without grounded neutral
6. Rating – continuous or short time
7. Cooling – natural or forced
8. Type – core or shell, power or distribution
9. Efficiency, per unit impedance, location (i.e., indoor, pole or platform mounting,
etc.), temperature rise, etc.

3.3 Design of Transformer

The design of transformer includes determination of its main dimensions namely the
overall width and height based on the geometry of windows, distance between limbs and
diameter of circumscribing circle for various types of core. Design also includes design of
yoke, core and winding. Estimation of no load current, iron loss and copper loss facilitates
the temperature rise calculations based on which cooling systems are designed. Calculation
of resistances and leakage reactances are used for the determination of the efficiency and
regulation.
Hence, the design of transformer can be split into following four steps:
1. Determining output equation relating output with machine dimensions
2. Determining overall dimensions
3. Design of cooling apparatus
4. Estimation of quantities such as no load current, iron losses, resistance and leakage
reactance and so on.

3.3.1 Output Equation of Single-phase Transformer

The kVA rating of a single-phase transformer is given as follows:

Q(kVA) = Vp I p ×10−3 (or) Vs I s × 10−3 (3.1)

 Ep I p ×10−3 [ ∵ Terminal voltage, Vp  Induced voltage, Ep ] (3.2a)

(or)  Es I s ×10−3 [ ∵ Vs  Es ] (3.2b)

where p → primary, V and E → voltage, s → secondary and I → current.
 Design of Transformer 3.9
Design of Transformer 3.9
Design
The voltage induced in a transformer winding is given as follows:
of Transformer 3.9
The voltage induced in a transformer winding is given as follows:
Ex = 4.44fφmTx (3.3)
The voltage induced in a transformer winding
Ex = 4.44fφ Tx is given as follows: (3.3)
where m

where Ex = 4.44fφmTx(3.3)
f – frequency (Hz)
– frequency
fwhere
φm – Flux
f – (Wb)(Hz)
frequency (Hz), φm – Flux (Wb), T – number of turns, x – can be p or s, denoting
φm––number
T
primaryFluxor(Wb)
of turns
secondary.
x –– can
T number
be p of
or turns
s, denoting primary or secondary
x – can be p or s, denoting primary or secondary
3.3.2 Output Equation of Single-phase Transformer (Core-type)
3.3.2 output equation of single-phase Transformer (Core-type)
The cross-section
3.3.2 output of a 1φ core-type
equation transformer Transformer
of single-phase is shown in Fig. 3.9. The arrangement of the
(Core-type)
The cross-section of a 1φ core-type transformer is shown in Fig. 3.9. The arrangement of the
winding
The is shownof
cross-section in Fig. 3.10.
winding is shown inaFig.
1φ core-type
3.10. transformer is shown in Fig. 3.9. The arrangement of the
winding is shown in Fig. 3.10.

Window area
Aw area
Window
Hw Aw Iron area
Hw Ai area
Iron
Ai
Ww
Ww

fig. 3.9 | Cross-section of 1φ core-type transformer

Fig.
fig. 3.9 | Cross-section of 1φ core-type transformer

H L L H
H L L H H L L H
L H H L
V V V V V V V V
V V V V V V V V

a1 a2
a1 a2
fig. 3.10 | Winding arrangement of 1φ core-type transformer
Fig.
fig. 3.10 | Winding arrangement of 1φ core-type transformer
The output of the transformer in kVA is related to its main dimensions. Hence, the output
canThe output of the
be determined intransformer
terms of areainofkVA is related
core/iron (Ai),toarea
its main dimensions.
of window Hence,
(Aw) and the output
window space
can be determined in terms of area of core/iron
factor (Kw). Hence, the following steps are followed. (Aii
), area of window (Aww) and window space
factor (Kww ). Hence, the following steps are followed.
3.10 Design of Transformer

In case of single-phase transformers, there is one primary and secondary winding. Half
of primary and half of secondary are placed on one limb and the other half of each of the
winding is placed on the other limb. Each limb has half of the primary and secondary
windings as shown in Fig. 3.10. Hence, the total copper area in window is given as follows:

Ac = Total copper area of [ primary winding + secondary windiing ]

= [ number of turns in primary × area of primary conductor ]
+ [ number of turns in secondary × area of secondary conductor ]
= Tp Ap + Ts As (3.4)

Assuming same current density, δ in both primary and secondary windings, we also get
Ip Is
δ= = (3.5)
Ap As

Ip Is
⇒ Ap = and As = (3.6)
δ δ

Substituting Eqs. (3.5) and (3.6) in Eq. (3.4), we get

Ip Is
Ac = Tp + Ts
δ δ
1
(
= Tp I p + Ts I s (3.7)
δ
)
Assuming, TpIp = Ts Is = AT(3.8)

where AT is the magnetic mmf.

Substituting Eq. (3.8) in Eq. (3.7), we get
AT + AT 2 AT
Ac = = (3.9)
δ δ

We know that the ratio of copper area (Ac) in window to the total window area (Aw) is
window space factor, which is given by

copper area in window A

Kw = = c (3.10)
total window area Aw

⇒ Ac = K w Aw (3.11)

Equating, Eqs. (3.9) and (3.11), we get

2AT
= K w Aw
δ
δ K w Aw
⇒ AT = (3.12)
2
 Output Equation of Single-phase Transformer (Shell-type) 3.11
Design of Transformer

From Eq. (3.2a),

Q = Ep I p × 10− 3
Q = Ep I p ×10−3
−33
=
=EEtt ×
×TTpp × ×10
× II pp × 10− (3.13)
(3.13)
Ep
[where Volt/turn, Et = Ep ⇒ Ep = Et × Tp ] (3.14)
[where Volt/turn, Et = Tp ⇒ Ep = Et ×Tp ] (3.14)
Tp
Substituting T I = AT from Eq. (3.8) in Eq. (3.13), we get
Substituting TppIpp= AT from Eq. (3.8) in Eq. (3.13), we get
Q = Et × AT × 10−3
Q = Et × AT × 10−3
Substituting Eq. (3.12) in the above equation, we get
Substituting Eq. (3.12) in the above equation, we get
δK A
Q = Et × δ K w Aw × 10−3
Q = Et × w2 w ×10−3
2
Substituting Ep / Tp from Eq. (3.3) in the above equation, we get
Substituting Ep / Tp from Eq. (3.3) in the above equation, we get
δK A
Q = 4.44× f ×φm × w w × 10−3
δ K 2A
Q = 4.44× f ×φm × w w ×10−3
2
= 2.22 f φmδ K w Aw × 10−3 (3.15)
= 2.22 f φm δ K w Aw ×10−3 (3.15)
Substituting φm = Bm (maximum flux density) × Ai (area of core or area of iron) in the above
equation, weφget
Substituting m = Bm (maximum flux density) × Ai (area of core or area of iron) in the above
equation, we get
Q = 2.22 f Bmδ K w Aw Ai × 10− 3
(3.16)
Q = 2.22 f Bm δ K w Aw Ai ×10−3 (3.16)
Equation (3.16) is the output equation of a single-phase core-type transformer and it relates
Equation
the output(3.16)
kVA is the output
rating equation ofand
to its dimensions a single-phase
[Link]-type transformer and it relates
the output kVA rating to its dimensions and parameters.

3.3.3 output
3.3.3 Output equation
Equation of single-phase
of Single-phase Transformer
Transformer (shell-type)
(Shell-type)
The
The cross-section
cross-section of
of the
the core
core of
of 1φ
1φ shell-type
shell-type transformer
transformer is
is shown
shown in
in Fig.
Fig. 3.11.
3.11. The
The
arrangement of winding is shown in Fig. 3.12.
arrangement of winding is shown in Fig. 3.12.

Window area Window area

Aw Aw

Hw Iron area Hw
Ai

Ww Ww

fig. 3.11 | Cross-section of 1φ shell-type transformer

Fig.
3.12
3.12 Design
Design of
of Transformer
Transformer

HV LV LV HV

fig. 3.12
Fig. 3.12 || Winding
Winding arrangement
arrangement of
of 1φ
1φ shell-type
shell-type transformer
transformer
Here, both the windings are wound over the central limb and so each window has all the
Here, both the windings are wound over the central limb and so each window has all the
primary and secondary windings.
primary and secondary windings.
Hence, the total copper area in the window is given by
Hence, the total copper area in the window is given by
A = T A +T A
Acc= TppApp+ TssAss
2IT
=2IT
= δ
δ
2 AT
=2 AT = K w Aw
= δ = K w Aw
δ
The aforementioned equation is same as Eqs. (3.9) and (3.11) and so the output equation
for The
a 1φaforementioned equationis issame
shell-type transformer sameasasthe
Eqs. (3.9) and
output (3.11)of
equation and
1φso the output
core-type equation
transformer
for a 1φ shell-type
and is given by transformer is same as the output equation of 1φ core-type transformer
and is given by
Q = 2.22 f Bmδ K w Aw Ai × 10−3
Q = 2.22 f Bm δ K w Aw Ai ×10−3
The above equation is the output equation of a single-phase shell-type transformer.
The above equation is the output equation of a single-phase shell-type transformer.
3.3.4 output equation of Three-phase Transformer
3.3.4
The kVAOutput
rating ofEquation of Three-phase
a three-phase transformer isTransformer
given by
The kVA rating of a three-phase transformer is given by
Q (kVA) = 3 Vp I p × 10−3 (or ) 3Vs I s × 10−3 (3.17)
−3 −3
Q (kVA) = 3 Vp I p ×10 (or) 3Vs Is ×10 (3.17)
 3Ep I p × 10−3 [∵ Vp  Ep ] (3.18a)
 3Ep I p ×10−3 [∵ Vp  Ep ] (3.18a)
(or)  3Es I s × 10−3 [∵ Vs  Es ] (3.18b)
(or)  3Es I s ×10−3 [∵ Vs  Es ] (3.18b)
where p – primary, V, E – voltage, s – secondary, I – current.
where p – primary, V, E – voltage, s – secondary, I – current.
3.3.5 output equation of Three-phase Transformer (Core-type)
3.3.5 Output
The cross-section Equation of Three-phase
of a 3φ core-type Transformer
transformer is shown in (Core-type)
Fig. 3.13. The arrangement of
winding
The is shown of
cross-section in aFig.
3φ 3.14.
core-type transformer is shown in Fig. 3.13. The arrangement of
winding is shown in Fig. 3.14.
 Output Equation
Output Equation of
of Single-phase
Single-phase Transformer
Transformer (Shell-type) 3.13
Design of Transformer
(Shell-type) 3.13

Window
Window area
area Window
Window area
area
A
Aw A
Aw
w w

H
Hww Iron H
Hww
Iron area
area
A
Aii

W
Www W
Www

fig.
Fig.
fig. 3.13
3.13 || Cross-section
Cross-section of
of 3φ
3φ core-type
core-type transformer
transformer

H
H L
L L
L H
H
L
L H
H H
H L
L L
L H
H H
H L
L

V
V V
V V
V V
V V
V V
V V
V V
V V
V V
V V
V V
V

a1
a a2
a
1 2

fig.
fig. 3.14
Fig. 3.14 || Winding
Winding arrangement
arrangement of
of 3φ
3φ core-type
core-type transformer
transformer

Each
Each limb
limb carries
carries both
both the
the primary
primary and
and secondary
secondary windings
windings ofof one
one phase.
phase. So,
So, in
in each
each
window,
window, there are two primary and two secondary windings. There are two windows, and
there are two primary and two secondary windings. There are two windows, and
in
in each
each window,
window, the
the total
total copper
copper area
area in
in the
the window
window is is given
given by
by
AAc c=
A ==2T
2T AApp+
2TppAc ++2T
2Ts sA
2T
p p AAs s(3.19)
s s
(3.19)
(3.19)
Following
Following the
the same
same steps
steps as
as used
used inin the
the derivation
derivation of of output
output equation
equation of
of single-phase
single-phase
transformer, substituting Eq. (3.6) in Eq.(3.19),
transformer, substituting Eq. (3.6) in Eq. we
(3.19), we
Eq.(3.19), we get get
get
II pI I
= 2Tpp pp+
AAcc= + 2TssIIsss
A c =22TT p δ +22TT s δ
δδ δδ
2
(( ))
= =22 TpI I p++TTsII s (3.20) (3.20)
= δδ T Tpp I pp + Tss I ss (3.20)
δ
3.14 Design of Transformer

Substituting Eq. (3.8) in Eq. (3.20), we get

2 4 AT
Ac = ( AT + AT ) = (3.21)
δ δ

Similarly, equating Eqs. (3.21) and (3.11), we get

4AT
= K w Aw
δ

δ K w Aw
⇒ AT = (3.22)
4
From Eq. (3.18a),
Q = 3Ep I p ×10−3

Substituting Eq. (3.14) in the above equation, we get

Q = 3Et ×Tp × I p ×10−3 (3.23)

Substituting Eq. (3.8) in Eq. (3.23), we get

Q = 3Et × AT ×10−3

Substituting Eq. (3.22) in the above equation, we get

δ K w Aw
Q = 3Et × ×10−3
4
Substituting Ep/Tp from Eq. (3.3) in the above equation, we get
δ K w Aw
Q = 3 × 4.44 × f ×φm × ×10−3
4

= 3.33 f φm δ K w Aw ×10−3 (3.24)

Substituting φm = Bm (maximum flux density) × Ai (area of core) in Eq. (3.24), we get

Q = 3.33 f Bm δ K w Aw Ai ×10−3 (3.25)

Equation (3.25) is the output equation of a three-phase core-type transformer.

3.3.6 Output Equation of Three-phase Transformer (Shell-type)

The cross-section of a 3φ shell-type transformer is shown in Fig. 3.15. The arrangement of
winding is shown in Fig. 3.16.
This transformer has six windows and each window carries LV and HV windings of one
phase only.
Copper area in each window is given by

Ac = Tp Ap + Ts As (3.26)
 OutputEquation
Output EquationofofThree-phase Design of(Shell-type)
Three-phaseTransformer
Transformer Transformer 3.15
3.15
(Shell-type) 3.15

Windowarea
Window area Windowarea
Window area
AAw w AAw w

HHw w HHw w

WWw w WWw w

HV LV
HV LV LV HV
LV HV

Windowarea
Window area Windowarea
Window area
AAw w AAw w
Iron
Iron
HHw w HHw w
area
area
AAi i
HV LV
HV LV LV HV
LV HV
WWw w WWw w

Windowarea
Window area Windowarea
Window area
AAw w AAw w
HV LV
HV LV LV HV
LV HV
HHw w HHw w

WWw w WWw w
fig.
fig. 3.16
3.16|||Winding
Fig.3.16 Winding
Winding
arrangement
arrangement of 3φ
arrangementofof3φ shell-type
3φshell-type
shell-type
fig.
fig. 3.15
3.15|||Cross-section
Fig.3.15 Cross-section
Cross-sectionofof 3φ
of3φ shell-type
3φshell-type transformer
shell-typetransformer
transformer transformer
transformer
transformer

Following the same steps as used in the derivation of output equation of single-phase
Followingthe
Following thesame
samesteps
stepsasasused
usedininthe
thederivation
derivationofofoutputoutputequation
equationofofsingle-phase
single-phase
transformer, substituting Eq. (3.6) in Eq. (3.19), we get
transformer, substituting Eq. (3.6) in Eq. (3.19), we
transformer, substituting Eq. (3.6) in Eq. (3.19), we get get
Ip I
Ac = TpI pI p + TIssI ss
A = T
Acc= Tpp δ+ Ts s δ+ T
δδ δδ
Tp I p + Ts I s
=TTppI pI p++TTs sI sI s (3.27)
== δ (3.27)
(3.27)
δδ
Substituting Eq.
SubstitutingEq.
Substituting (3.8)
Eq.(3.8) in
(3.8)in Eq.
inEq. (3.20),
Eq.(3.20), we
(3.20),we get
weget
get
AT +
AT + AT
AT
A =AT + AT
AAccc== δ
δδ
222AT
AT
AT
=
== (3.28)
(3.28)
(3.28)
δδδ
3.16 Design of Transformer

Similarly, equating Eqs. (3.21) and (3.11), we get

2AT
= K w Aw
δ
δ K w Aw
⇒ AT = (3.29)
2
From Eqn. (3.18a),
Q = 3Ep I p ×10−3

Substituting Eqn. (3.14) in the above equation, we get

Q = 3Et ×Tp × I p ×10−3 (3.30)

Substituting Eqn. (3.8) in Eq. (3.30), we get

Q = 3Et × AT ×10−3

Substituting Eq. (3.22) in the above equation, we get

δ K w Aw
Q = 3Et × ×10−3
2
Substituting Ep/Tp from Eq. (3.3) in the above equation, we get
δ K w Aw
Q = 3 × 4.44 × f ×φm × ×10−3
2
= 6.66 f φm δ K w Aw ×10−3 (3.31)

Substitution φm = Bm (maximum flux density) × Ai (area of core) in Eq. (3.31), we get

Q = 6.66 f Bm δ K w Aw Ai ×10−3 (3.32)

The above equation is the output equation of a three-phase shell-type transformer.

3.4 Volt Per Turn of Winding

The rating of a transformer per phase is given as

Q (kVA) = VI ×10−3 = 4.44 f φm TI ×10−3

= 4.44 f φm AT × 10-3 [ ∵ TI = AT](3.33)

Here, the term, φm is called the magnetic loading and AT is called electric loading.
The magnetic loading, φm = BmAi gives a measure of cross-section of iron core. As the
magnetic loading increases, the flux density Bm and hence the core loss increases which
reduces the efficiency of operation.
Similarly, the electric loading AT gives a measure of the cross-section of the winding.
Increase in electric loading increases the number of turns, resistance and hence results in
increased copper losses reducing the efficiency.
 Choice of Flux Density 3.17

So, higher values of φm and AT will only reduce the efficiency of the transformer. For
economical design of transformer, the ratio of φm and AT is maintained constant, which
depends on the type of the transformer.
φ φ
The ratio, x = m ⇒ AT = m (3.34)
AT x

Substituting Eq. (3.34) in Eq. (3.33), we get

2
φm
Q = 4.44 f ×10−3
x

x ×10 3
⇒ φm = × Q
4.44 f

Now, Volt per turn of winding is given by

Et = 4.44 f φm

x ×10 3
= 4.44 f × Q
4.44 f

⇒ Et = 4.44 f x ×10 3 × Q

⇒ Et = K Q

where K = 4.44 f x×10 3

Here, x and hence K are constants. Their value depends on type, construction and service
conditions. Table 3.4 gives the various values of K for different types of transformers.

Table 3.4 | Values of K for various types of transformers

Type of transformers K
Single-phase core type 0.75–0.85
Three phase core type( power) 0.6–0.7
Three phase core type( distribution) 0.45
Single phase shell type 1.0–1.2
Three phase shell type 1.3

3.5 Choice of Flux Density

The output equation and the emf equation indicate that the choice of flux density Bm affects
the core area Ai. Higher flux density reduces the core area. This will reduce the diameter of the
circumscribing circle, thereby reducing the length of mean turn of winding. Thus, there will be
3.18 Design of Transformer

a huge saving in the cost of copper and iron. But, a higher flux density, while saving copper and
iron, will also result in increased magnetizing current leading to saturation, increased iron losses
requiring elaborate cooling arrangements. The choice of Bm also depends on the type and service
conditions of the transformer. It has already been pointed out that a distribution transformer
should be designed for lower iron losses giving good all day efficiency. Hence, for distribution
transformers, a comparatively lower flux density is assumed. The permissible values of Bm for
various types of transformers are given in Table 3.5. The variation depends on the rating of the
transformer as well, higher values for higher rating.

Table 3.5 | Values of flux density for various types of transformers

Type of transformers Value of flux
density
Distribution transformer-hot rolled silicon steel 1.1 to 1.4 Wb/m2
Power transformer-hot rolled silicon steel 1.2 to 1.5 Wb/m2
Transformers using cold rolled grain oriented steel
Up to 132 kV 1.55 Wb/m2
For 275 kV 1.6 Wb/m2
For 400 kV 1.7 Wb/m2

3.6 Choice of Current Density

The choice of specific electric loading, i.e., the current density (δ), will affect the transformer
dimensions and efficiency. A higher value of current density will result in smaller dimensions
of transformer. But with increased electric loading heating, losses and hence efficiency
decreases. Higher value of current density will result in excessive temperature rise, resulting
in damage in insulation due to local heating. Even with improved methods of cooling and
with well-ventilated winding, the choice of current density is limited by the local heating.
The type of transformer also affects the choice of current density as the requirement of copper
and iron losses is different for power and distribution transformers. Table 3.6 gives the range
of permissible values of current density for different types of transformers.

Table 3.6 | Values of current density for various types of transformers

Type of transformers Value of current
density (A/mm2)
Standard distribution transformer and 1.5 to 2.6
small power transformer
Medium and large power transformer 2.4 to 3.4
Large power transformers 5.4 to 6.3

3.7 Design of Core

The core forms the magnetic circuit of the transformer along with the yoke. The core is built
up of silicon steel laminations of 0.3–0.5 mm thickness with varnish insulation in between.
 Design of Core 3.19

The core section of a small size transformer is rectangular with circular or rectangular coils.
For higher sizes, circular cores with circular coils are preferred over rectangular coils as the
former occupies less space and uses lesser amount of copper. Also, the mechanical forces
developed under short circuit conditions, change the shape of square or rectangular sections
and damage the coil and insulation. This is not so in circular section. Hence, a circular core
section and circular coils are generally used. The different types of core laminations used for
core-type and shell-type transformers are represented in Fig. 3.17.

Shell-type Laminations Core-type Laminations

“E-I” Laminations “E-E” Laminations “L” Laminations “U-I” Laminations

fig. 3.17 | Types of core laminations used for core-type and shell-type transformers
Fig.

As a laminated core needs to be used in a transformer, a larger number of laminations of

different sizes are required to build a circular core. It also becomes difficult to arrange them
together to get a circular section. So, it is a general practice to approximate a circular core as
a stepped core with infinite number of steps. However, the maximum number of steps can
be restricted for practical reasons with minimum being one. A stepped core also enables the
use of circular coil. It is also called as cruciform type of core if the number of steps are two.
Figure 3.18 gives the types of cores and Fig. 3.19 shows the core sections of multi-stepped,
square and four-stepped cruciform cores. The shape of the core has to be selected based
on minimum wastage of useful space and for minimum length of copper. The imaginary
circle passing through the corners of the core section is known as circumscribing circle or

Cruciform (2 step)
Cruciform(2step)

Stepped 3 step
3step

Types of core Square 4 step

4step

Rectangle

fig. 3.18 | Types of cores

Fig.
3.20
3.20 Design
Design of
of Transformer
Transformer

fig.
fig. 3.19
Fig. 3.19 || Core
Core sections
sections of
of multi-stepped,
multi-stepped, square
square and
and four-stepped
four-stepped cruciform
cruciform cores
cores

circum
circum circle.
circle. The
The dimensions
dimensions forfor obtaining
obtaining the
the maximum
maximum areaarea in
in the
the given
given circum
circum circle,
circle, for
for
various
various number
number of of steps,
steps, are
are given
given in
in Fig
Fig 3.19.
3.19.
Since
Since the
the core
core is
is prepared
prepared byby stacking
stacking the
the laminations
laminations together
together which
which are
are insulated
insulated from
from
each
each other
other by
by aa thin
thin coating
coating of
of varnish,
varnish, assembled
assembled core
core includes
includes the
the area
area of
of insulation
insulation also.
also.
The
The stacking
stacking factor
factor that
that relates
relates the
the gross
gross core
core section
section with
with the
the net
net core
core section
section is
is defined
defined asas

Net Net core section, A

Netcore
coresection,
section, A
Aii =i K = Ks (Stacking factor)  (3.35)
Gross section,
Gross AgiKss(Stacking
core section, = (Stacking factor)
factor) (3.35)
(3.35)
Grosscore
core section, A
Agi
gi

TheThestacking factor usually varies from 0.850.85

to [Link].
The stacking
stacking factor
factor usually
usually varies
varies from
from 0.85 toto 0.9.
In
In order to determine the utilization of the available space
space of transformer
transformer core, the
the ratios,
In order
order to
to determine
determine the
the utilization
utilization of
of the
the available
available space ofof transformer core,
core, the ratios,
ratios,
net
net core area to circumscribing circle’s area and gross core area to circumscribing circle’s area
net core
core area
area to
to circumscribing
circumscribing circle’s
circle’s area
area and
and gross
gross core
core area
area to
to circumscribing
circumscribing circle’s
circle’s area
area
are
are determined.
are determined.
determined.
3.7.1 Square
3.7.1
3.7.1 square Core
square Core
Core
Square
Square cross
cross section
section

dd aa

fig.
fig. 3.20
Fig. 3.20 ||| Square
Square core
core
From
From Fig.
Fig. 3.20,
3.20, the
the gross
gross core
core area
area for
for square
square core
core is
is given
given by
by
From Fig. 3.20, the gross core area for square core is given by
A = aa×
gi =
Agi ×aa,, (3.36
(3.36 a)
a)
Agi = a× a, (3.36a)
where
where aa –– side
side of
of square.
square.
where
And ais–observed
side of square.
And it is observed from
it from Fig.
Fig. 3.20
3.20 that
that
And it is observed from Fig. 3.20 that
diagonal of square =
diagonal of square = diameter
diameter ofof circumscribing
circumscribing circle
circle
diagonal of square=
=dd = diameter of circumscribing circle

=d
 Design of Core 3.21

Also, for any square, diagonal (d) = 2a

d
Hence, a= d (3.36 b)
Hence, a = 2 (3.36b)
2
Substituting Eq. (3.36 b) in Eq. (3.36 a), we get
Substituting Eq. (3.36b) in Eq. (3.36a), we get
d2
Agi = d 2 = 0.5d22 (3.36 c)
Agi = 2 = 0.5d (3.36c)
2
We know that
We know that
Net core area, A = Stacking factor × gross core area
Net core area, Aii = Stacking factor × gross core area
A = K × Agi
Aii = Kss × Agi
Assuming stacking factor = 0.9 and substituting Eq. (3.36 c) in the aforementioned equation,
Assuming stacking factor = 0.9 and substituting Eq. (3.36c) in the aforementioned equation,
we get
we get
A = 0.9 × 0.5d 2
Aii = 0.9× 0.5d 2

= 0.45d 2
= 0.45d 2
To determine
To determine the
the utilization
utilization ofof available
available space
space in in transformer,
transformer, area
area of
of circumscribing
circumscribing
circle is
circle is required
required to
to be
be found.
found. It
It is
is given
given by
by
π d22
Acc =
A = πd
44
cc

where dd –– diameter
where diameter ofof circle.
circle. Hence,
Hence, the
the ratio
ratio of
of net
net core
core area
area toto circumscribing
circumscribing circle’s
circle’s area
area
is given by
is given by
AAi 0.45d22
i == 0.45d2 = =00..58
58
Acc
A cc ππdd2
44
And, the
And, the ratio
ratio of
of gross
gross core
core area
area to
to circumscribing
circumscribing circle’s
circle’s area
area is
is given
given by
by
Agi 0.5d22
Agi = 0.5d = 0.64
Acc = π d22 = 0.64
Acc πd
44
3.7.2 Stepped
steppedCore
Core
The stepped core has two regions, region 1 and region 2, as shown in Fig. 3.21.

d a
1 b
θ

fig. 3.21 || Stepped core

Fig.
3.22 Design of Transformer

From Fig. 3.21, the gross core area is given by

Agi = Area of region 1 + Area of region 2

where
Area of region 1 = ( a×b)

Area of region 2 = ( a − b) × b
Hence,
Agi = ab + ab − b 2

= 2 ab − b 2 (3.37)

It is required to express the terms ‘a’ and ‘b’ in terms of ‘d’, which is the diameter of
A Agi
circumscribing circle, in order to find the ratios, i and .
Acc Acc
From Fig. 3.21,
a b
cos θ = and sin θ =
d d

⇒ a = d cos θ and b = d sin θ (3.38)

Substituting Eq. (3.37) in Eq. (3.38), we get

2
Agi = 2× d sin θ × d cos θ − (d sin θ)

= 2d 2 sin θ cos θ − d 2 sin 2 θ

Agi = d 2  sin 2θ − sin 2 θ  [ ∵ 2 sin θ cos θ = sin 2θ ] (3.39)

 
The maximum area for a given d is obtained when θ is maximum. So, to get maximum θ,
Agi is differentiated with respect to θ and equated to zero.
dAgi
=0
dθ

Hence, on differentiating Eq. (3.39) with respect to θ and equating it to zero, we get
dAgi
= d 2 [ 2 cos 2θ − 2 sin θ cos θ ] = 0
dθ

⇒ 2 cos 2θ = 2 sin θ cos θ

⇒ 2 cos 2θ = sin 2θ
sin 2θ
⇒ =2
cos 2θ
 Design of Core
Design of Core
3.23
3.23
⇒ tan 2θ = 2
⇒ tan 2θ = 2
⇒ 2θ = tan−1 (2) and θ = 31.72° (3.40)
⇒ 2θ = tan−1 (2) and θ = 31.72° (3.40)
2 (stepped) (Cruciform)
Hence, with
Hence, withθ =θ=3131 °, °we
.72.72 , we
getget maximum
maximum area
area
of of for
core corea for a given
given d. Hence,
d. Hence,
aa==ddcos
cosθθ==0.85d
0.85d
and
and b = d sin θ = 0.53d
b = d sin θ = 0.53d 50
100 80
Using Eq. (3.40) in Eq. (3.39), we get
Using Eq. (3.40) in Eq. (3.39), we get
A = 0.618 d 2
Agi = 0.618gid 2 50
80
Net core area, Ai = 0.9× Agi
Net core area, Ai = 0.9 × Agi
=0.56 d 2
= 0.56 d 2 100
 ∵ stacking factor is 3 (stepped)
   ∵ stacking factor is  85
 assumed to be 0.9  

 assumed to be 0. 9 
 
Ai 0.56 d 22 40
Ai = 0.56 d = 0.71 66
Acc = π d 22 = 0.71 85
Acc πd
4
4
A 2
gi = 0
Agi 0..618
618dd 2 = 0.79
= 2 = 0.79
A
Acc π
πdd 2 87 87 100 4 (stepped)
cc
44

Hence, it
Hence, it is
is proved
proved thatthat the
the area
area ofof 34
circumscribing circle can be more effectively
circumscribing circle can be more effectively 57
used by
used by increasing
increasing the
the number
number of ofsteps
stepsofofthe
the 74
core. Table 3.7 shows the ratio of net, gross
core. Table 3.7 shows the ratio of net, gross core
core area
area to area
to the the area of circumscribing
of circumscribing circle
circle for Fig.
fig. 3.22
3.22 || Types
Types of
of stepped
stepped cores
cores
for various cores represented in Fig.
various cores represented in Fig. 3.22. 3.22.

Table 3.7 | 3.7

Table Ratio|ofRatio
net, gross
of net,core area
gross to area
core the area of area
to the circumscribing circle forcircle
of circumscribing various cores cores
for various
ratio
Ratio square core
Square core 2-stepped 3-stepped 4-stepped
A /area of the circumscribing circle, Acc 0.64 0.79 0.84 0.87
gi/area of the circumscribing circle, Acc
Agi 0.64 0.79 0.84 0.87
Ai/area of the circumscribing circle, Acc 0.58 0.71 0.75 0.78
Ai/area of the circumscribing circle, Acc 0.58 0.71 0.75 0.78

Note that the values in the second row of Table 3.7 can be obtained by multiplying the
Note that the values in the second row of Table 3.7 can be obtained by multiplying the
respective values in the first row. This is because, Ai = Agi × Stacking factor. Stacking factor
respective values in the first row. This is because, Ai = Agi × Stacking factor. Stacking factor
is assumed as 0.9.
is assumed as 0.9.
3.24 Design of Transformer

3.8 Design of Yoke

The yoke of a transformer connects the legs and provides the least reluctance path. The yoke
portion does not have windings and so the iron loss in yoke can be minimized by increasing
the yoke area, thereby reducing the operating flux density. Generally, the yoke area is made
20% more than the leg area. The yoke sections can be rectangular or cruciform. Figure 3.23
shows the yoke for rectangular and cruciform types of leg. A cruciform leg can also have a
rectangular yoke as shown in Fig. 3.23.
3.23

Dy
a
Hy a

a
a
a
a
Rectangular leg and yoke Cruciform leg and yoke Rectangular yoke and
cruciform leg

fig. 3.23 | Yoke section for various types of legs

Fig.

For a rectangular yoke, the depth Dyy of the yoke will be same as the depth of the core.
This will be equal to the width of the largest core stamping, ‘a’. The height of the yoke Hyy can
in general be taken as 1–1.5 times the width of the leg.
Area of yoke,
Area =yD=y D
Ay A
of yoke, ×yH×y H =y1.15 to 1.20
= 1.15 times
to 1.20 AgiA
times (for Hot Rolled Silicon Steel)
gi (for Hot Rolled Silicon Steel)
=yA=giA(for
Ay A gi
Cold
(for Rolled
Cold Silicon
Rolled Steel)
Silicon Steel)

3.9 Window and Core Proportions

The area of the window can be determined from the output equation of the transformer.

Area windowAAw== QQ inin

kVAkVA
Area of of
window w 33
2.222
.22
oror
3.333
.33
oror
6.666 δ BmAAK
.66f δf B K ××
1010
m i i ww
If WIfwW
is the width of the window and Hw is the height of the window, then Aw = Ww × Hw
w is the width of the window and Hw is the height of the window, then Aw = Ww ×
in order to limit the the
Hw in order to limit leakage reactance
leakage of the
reactance transformer,
of the Hw H
transformer, is made more than the Ww. In
w is made more than the Ww.
practice, the ratio H /W
In practice, the ratiowH /W w lies between 2.5 and 3.5.
lies between 2.5 and 3.5.
w w

3.10 Overall
3.10 overall Dimensions
Dimensions
After
After designing
designing the
the core,
core, yoke
yoke sections
sections and
and the
the window
window sections,
sections, the
the overall
overall dimensions
dimensions
could be computed as follows.
could be computed as follows.
3.10.1 Single-phase
3.10.1 single-phase Core-type
Core-type Transformer
Transformer
Figure
Figure 3.24
3.24 shows
shows the
the overall
overall dimensions
dimensions ofof aa single-phase
single-phase core-type
core-type transformer,
transformer, where
where H
H
–– overall
overall height, W – overall width, D – distance between the adjacent limbs, d – diameter of
height, W – overall width, D – distance between the adjacent limbs, d – diameter of
 Overall Dimensions 3.25
Overall Dimensions 3.25

circumscribing
circumscribing circle, Www –– width
circle, W width of
of the
the window,
window, H Hww –– height
height of
of the
the window, Hyy –– Height
window, H Height of
of
yoke, D – depth of yoke.
yoke, Dyy – depth of yoke.
HH= =HHw + 2H
w + 2Hyy
DD= =WWw +a
w+a
W=
W =DD++ aa
D = a
Dyy = a
Horizontal cross section

one limb
Width of
Ww

Hy

Hw H

Hy

W
a a
d d
Width of two limbs

Vertical cross section

Fig.
fig. 3.24
3.24 || Overall
Overall dimensions
dimensions of
of single-phase
single-phase core-type
core-type transformer
transformer

3.10.2
3.10.2 Three-phase
Three-phase Core-type
Core-type Transformer
Transformer
Figure
Figure 3.25
3.25 shows
shows the
the overall
overall dimensions
dimensions of
of aa three-phase
three-phase core-type
core-type transformer,
transformer, where
where
D=
D =W Www +
+ dd
H=
H =H Hww +
+H Hyy
WW= = 2D2D ++ aa

3.10.3 single-phase Shell-type

3.10.3 Single-phase shell-type Transformer
Transformer
In shell-type
In shell-typetransformer
transformershown
shown in in Fig.
Fig. 3.26,
3.26, thethe
corecore section
section will will be rectangular
be rectangular in shape.
in shape. Only
the central limb carries windings and so the flux in the central limb is twice that of the of
Only the central limb carries windings and so the flux in the central limb is twice that the
outer
outer limbs. Hence, the area of cross-section of the central limb is twice that of
limbs. Hence, the area of cross-section of the central limb is twice that of the other limbs. the other limbs.
3.26
3.26 Design
Design of
of Transformer
Transformer

Horizontal cross section

D

Hww

Www
d
Hyy

Hww H

Hyy
W
a a a
d d d
Width of three limbs
Vertical cross section

fig. 3.25 | Overall dimensions of a three-phase core-type transformer

Fig.

Horizontal
cross
section a
2a b
b

Www Www

a
Vertical
cross
section Hww H

fig. 3.26
Fig. 3.26 || Overall
Overall dimensions
dimensions of
of aa single-phase
single-phase shell-type
shell-type transformer
transformer
 Overall Dimensions
Overall 3.27
Dimensions 3.27

The dimensions
The dimensions are
are given
given by
by
Gross
Grosscore
coresection
sectionofofthe
thecentral limb==2a2a××b b
centrallimb
Over all
Over all width,
width, WW==2W
2Ww++4a4a
w
Overall height, HH==HH
ww++2H
2H y=
y= HH
ww++2a2a
Note: The
Note: Theratio b/2a=
ratiob/2a =2.5
2.5
3.10.4 Three-phase
3.10.4 Three-phase shell-type
Shell-type Transformer
Transformer
Figure 3.27
Figure 3.27 shows
shows thethe overall
overall dimensions
dimensions of of aa three-phase
three-phase core-type
core-type transformer.
transformer. The
The
dimensions are given
dimensions are given by by
Gross
Grosscore
coresection
sectionofofthe
thecentral limb==2a2a××b b
centrallimb
Overall width,
Overall width, WW==2W
2Ww++4a4a
w
Overall height, H = 3Hw w++4H
H = 3H 4Hy = 3Hw +
y = 3Hw + 4a4a

Three phase core type

w

Depth of winding
Horizontal
section b

ww ww
a

2a Hw

Vertical
section a

fig. 3.27 | Overall dimensions of a three-phase shell-type transformer

Fig.
3.28 Design of Transformer

3.11 Design of Windings

The applied voltage V1 is approximately equal to the voltage induced:
V1 = E1 = 4.44 φm f T1 = Et T1
Number of primary turns (or turns/phase),
V1
T1 = in case of single-phase transformers
Et
V1ph
= in case of three-phase transformers
Et

Number of secondary turns (or turns/phase),

V2
T2 = in case of single-phase transformers
Et
V2ph
= in case of three-phase transformers
Et
Primary current (or current/phase),
kVA ×10 3
I1 = in case of single phase transformers
V1

kVA ×10 3
= in case of three-phase transformers
3V1ph

Cross-sectional area of primary winding conductor

I1
a1 = mm 2
δ
Secondary current (or current/phase)
kVA ×10 3
I2 = in case of single-phase transformers
V2
kVA ×10 3
= in case of three-phase transformers
3V2ph

Cross-sectional area of secondary winding conductor

I2
a2 = mm 2
δ
Knowing the number of turns and cross-sectional area of the primary and secondary
winding conductors, number of turns/layer in a window height of Hw and number of layers
in a window width of Ww can be found out.
 Design
Design of Windings 3.29
of Windings 3.29
The
The various
various types
types of
of windings
windings used
used in
in transformers
transformers are
are shown
shown in
in Fig.
Fig. 3.28.
3.28.

Winding
types

Continuous
Cylindrical Helical Layer
disc type

fig. 3.28 | Different types of windings used in transformers

Fig.

3.11.1 Type
3.11.1 Type and
and Choice
Choice of
of Windings
Windings
Differenttypes
Different typesof of windings
windings namely,
namely, cylindrical,
cylindrical, helical,helical,
layer andlayer and continuous
continuous disc
disc windings
windings
are generallyareused
generally usedupon
depending depending uponand
the voltage thecurrent
voltage and current
ratings. ratings.
The various typesThe
of
various types
windings of windings
frequently used infrequently
transformer used in transformer
construction construction
are explained are explained as
as follows.
follows.
Cylindrical winding
Cylindrical winding
The cylindrical winding may use circular, rectangular or strip conductors. It is generally
The cylindrical winding may use circular, rectangular or strip conductors. It is generally
wound with many layers. The layered cylindrical winding with sheet conductors is shown
wound with many layers. The layered cylindrical winding with sheet conductors is shown
in Fig. 3.29(a).
in Fig. 3.29(a).
Helical
Helical winding
winding
This
This type
type of
of winding
winding may
may be
be of
of single
single or
or double
double helical
helical type.
type. They
They are
are usually
usually wound
wound around
around
aa bakelite
bakelite cylinder. In a single helical type, the turns are wound in the axial direction. There
cylinder. In a single helical type, the turns are wound in the axial direction. There
is
is only
only one
one turn
turn in
in each
each winding
winding layer.
layer. If
If the
the winding
winding consists
consists ofof many
many strands
strands inin parallel,
parallel,
they
they are
are called
called double
double helical
helical winding.
winding. Helical
Helical windings
windings are are usually
usually used
used for
for low-voltage
low-voltage
winding
winding where the number of turns is less and current is high. Since the current is
where the number of turns is less and current is high. Since the current is high,
high, aa
large number of strips are used in parallel and are placed side by side in a radial
large number of strips are used in parallel and are placed side by side in a radial direction. direction.
These
These parallel
parallel connected
connected conductors
conductors can can bebe transposed
transposed at at regular
regular intervals
intervals toto equalize
equalize the
the
resistance
resistance and leakage reactance of each conductor. A single helical winding is
and leakage reactance of each conductor. A single helical winding is shown
shown in in
Fig.
Fig. 3.29(b).
3.29(b).
Continuous
Continuous disc
disc winding
winding
The
The disc
disc winding
winding consists
consists of
of number
number of
of flat
flat coils
coils or
or discs
discs connected
connected in
in series
series or
or parallel.
parallel. The
The
discs
discs are
are formed
formed by
by single
single or
or number
number of
of strips
strips in
in parallel
parallel wound
wound spirally.
spirally. These
These windings
windings are
are
used
used for
for high
high rated
rated transformers.
transformers. A
A simple
simple disc
disc winding
winding isis shown
shown in
in Fig.
Fig. 3.29(c).
3.29(c).
Layer winding
This type of winding consists of several cylindrical layers wound concentrically and are
connected in series as shown in Fig. 3.29(d). They are used for windings rated above 100 kV,
with the number of layers dependent on the voltage rating. Generally, the outer layers are
made shorter to distribute the winding capacitance uniformly.
Once the choice of winding is properly made, then it is designed in such a way that it
fits properly with the calculated window dimensions accommodating provision for ducts,
insulation and clearances. Table 3.8 gives a guideline about the choice of windings based on
the voltage ratings.
3.30 Design
3.30 Design of
of Transformer
Transformer

12 11 10 9 8 7 6 5 4 3 2 1

Core Low
leg voltage
winding

(a)
(b)

Layer Winding

Spacer

Transposition

(b) (c) (d)

fig.
Fig. 3.29
3.29 || Different
Different types
types of
of windings
windings used
used inin transformers:
transformers: (a)
(a) Continuous
Continuous disc
disc type
type winding,
winding,
(b) Helical winding, (c) Layer winding and (d) Cylindrical winding
(b) Helical winding, (c) Layer winding and (d) Cylindrical winding

Table 3.8 | Choice of windings based on the voltage ratings of transformers

Table 3.8 | Choice of windings based on the voltage ratings of transformers
Type of transformers kVa Voltage in kV Winding
Type of transformers kVA Voltage in kV Winding
Transformers for 1.6–100 HV – 3.3 to 11 Foil, crossover
Transformers for rural 1.6–100 HV – 3.3 to 11 Foil, crossover
rural supply LV – 0.44 Helical
supply
LV – 0.44 Helical
Distribution 100–1000 HV – 11 to 33 Foil, crossover, disc, multilayer
Distribution small
transformer, 100–1000 HV – 11 to 33 Foil, crossover, disc,
helical
transformer, small
power transformer multilayer helical
power transformer LV – 0.44 Helical, multilayered helical
LV – 0.44 Helical, multilayered helical
Transformers 10,000–20,000 HV – 33 to 66 Disc
Transformers
for secondary for 10,000–20,000 HV – 33 to 66 Disc
secondary transmission
transmission LVLV
– 11
– 11 DiscDisc
or helix
or helix
Transformers
Transformers for
for 45,000–15,00,00
45,000–15,00,00HVHV– 132 to to
– 132 440
440 DiscDisc
or multi-layered helix
or multi-layered helix
primary
primary transmission
transmission LVLV
– 11,33,66
– 11,33,66 DiscDisc
or helix
or helix
Transformers
Transformers at
at 72,000–35,00,00
72,000–35,00,00HVHV– 132 to to275
– 132 275 DiscDisc
or multi-layered helix
or multi-layered helix
generating
generating stations
stations LVLV
– 1122
– 1122 DiscDisc
or helix
or helix
 Reactance Calculation 3.47

Resistance of the secondary winding/phase,

ρLmt Ts
Rs =
as

Mean length of turn of the primary winding,

Lmt = π × mean diameter of the winding
Mean length of turn of the secondary winding,
Lmt = π × mean diameter of the winding
Where resistivity of copper at 60°C, r = 2.1 × 10−6 ohm-cm or 2.1 × 10−8 ohm-m or 0.021
ohm/m/mm2
Lmtp and Lmts are the length of mean turns of primary and secondary windings, respectively.
Tp, Ts are the corresponding number of turns
ap, as are the area of cross-sections of the conductors
Total resistance per phase referred to primary side,

 Tp 2
R1 = Rp +   Rs
 Ts 

3.13 Reactance Calculation

The reactance of the windings primarily involves estimation of the leakage flux. Reactance
calculations are required to make an estimate the regulation of the transformer to be
designed.
(a) Useful flux
The flux that links with both primary and secondary windings and is responsible in
transferring the energy electromagnetically from primary to secondary side is called
the useful flux. The path of the useful flux is in the magnetic core.
(b) Leakage flux
The flux that links only with the primary or secondary winding and is responsible
in imparting inductance to the windings is called the leakage flux. The path of
the leakage flux depends on the geometrical configuration of the coils and the
neighbouring iron masses.
(c) Reactance
1. Leakage reactance = 2πfl × inductance = 2πf × (Flux linkage/current)
2. Flux linkage = flux × number of turns
3. Flux = (mmf or AT)/Reluctance = AT × permeance ( ∧ )
aµ0 µr
4. Permeance ( ∧ ) = 1/Reluctance =
l
where a = area over which the flux is established, l = length of the flux path.
If xp and xs are the leakage reactances of the primary and secondary windings, then the
total leakage reactance of the transformer referred to primary winding
3.48 Design of Transformer

 Tp 2
X p = xp + xs′ = xp + xs  
 Ts 

Similarly, the leakage reactance of the transformer referred to secondary winding

 T 2

Xs = xp′ + xs = xp  s  + xs
 Tp 

Estimation of the leakage flux or reactance is always difficult, on account of the complex
geometry of the leakage flux path and great accuracy is unobtainable. A number of
assumptions are to be made to get a usable approximate expression. Validity or the accuracy
of the expression is checked against test data.

3.13.1 Leakage Reactance of a Core-type Transformer with ­Concentric

LV and HV Coils of Equal Height or Length
The following are the assumptions considered for the leakage reactance of the core-type
transformer.

•• Effect of magnetizing current is neglected.

•• Reluctance and effect of saturation of iron are neglected.
•• All the mmf is assumed to be used to overcome the reluctance of coil height.
•• Leakage flux distribution in coil and in the space between the LV and the HV coils is
assumed to be parallel to the leg axis.
•• Equal axial lengths for both primary and secondary windings.
•• The flux paths are parallel to the winding along the axial height.
•• Ip Tp = Is Ts = Total mmf
•• Length of mean turn of both windings is equal.

Let,
bp and bs = Radial depth of primary and secondary windings
Tp and Ts = Number of primary and secondary turns per phase for three phase
Ip and Is = Primary and secondary currents per phase for three phase
Lmtp/Lmts = Mean length of turn of primary or secondary windings, respectively
Lmt = Mean length of primary and secondary windings considered together
L0 = Circumference of the insulation portion or duct or both between LV and HV coils
Lc = Axial height or length of both LV and HV coils
The total flux linkage of the primary or secondary winding is due to
•• Leakage flux inside the primary or secondary winding and
•• Leakage flux in between the LV and HV coils

To determine the flux linkage due to the flux inside the coil, consider an elemental strip
dx at a distance ‘x’ from the edge of the LV winding.
Then the flux linkage of the LV winding due to the flux φx in the strip,
Ψx = φx × number of turns linked by φx
Ψx = ampere turns producing φx × permeance of the strip × number of turns linked by φx
 Reactance Calculation 3.49

 I pTp x   Lmtp dxµ0   Tp x 

 ×   
ψx =  ×

 bp   Lc   bp 


Considering that the mean length of the strip is approximately equal to Lmtp.
Therefore, the total flux linkage due to the flux inside the coil is given by
bp
 I pTp x   Lmtp dxµ0   Tp x 
 ×   
ψx = ∫  ×
 bp   Lc  
 bp 
0

I pTp2 µ0 Lmtp bp
ψx =
3 Lc

If one half of the flux φ0, in between the LV and HV windings, is assumed to be linking
with each winding, then the flux linkage of the primary winding due to half of the flux φ0 in
between LV and HV windings is
1
ψ0 = φ0 ×Number of turns linked by φ0
2
= ampere turns producing φ0 × permeance of the
duct × number of turns linked by flux φ0
1 L aµ
ψ0 = ×Tp I p × 0 0 ×Tp
2 Lc

Therefore, the total flux linkage of the primary winding is

µ0  Lmtp bp L0 a 
λ = ψ + ψ0 = I pTp2 ×  + 
Lc  3 2 

Assuming that Lmtp and L0 to be equal, the total flux linkage of the primary winding
becomes, with the assumption that Lmtp = L0

µ0 Lmtp  bp a 
λ = ψ + ψ0 = I pTp2 ×  + 
Lc  3 2 

Therefore, leakage reactance of the primary/phase,

 flux linkage 
xp = 2π f × 
 current 

µ0  Lmtp bp L0 a 
I pTp2 ×  + 
Lc  3 2 
x p = 2π f
Ip
 No Load Current of a Transformer 3.61

Im produces flux φm in the magnetic circuit and No

Ic supplies the of
Load Current noa Transformer
load losses of the
3.61
transformer.

m producesno
IThus, flux φm in
load the magnetic
input circuit and=IcV
to the transformer supplies the no load losses of the transformer.
1(I0 cos φ0) = V1 Ic
Thus,
no load input to the transformer = =VNo cos φlosses
1(I0load 0) = V(as
1 Ic the output is zero)
= No load losses (as the output is zero)
and input = output + losses
and input = output + losses
Since the copper loss under no load condition is almost negligible, the no load losses can
Since be
entirely thetaken
copperas loss
due under
to coreno load
loss condition is almost negligible, the no load losses can
only.
entirely be taken as due to core loss only. core loss
Thus, the core loss component of the no load current, I c =core loss , for single-phase
Thus, the core loss component of the no load current, I c = V , for single-phase
core loss/phase V11
transformers and I c =core loss/phase , for three-phase transformers.
transformers and I c = V1ph , for three-phase transformers.
V1ph
RMS value of magnetizing current,
RMS value of magnetizing current,
Peak magnetizing ampere turns
Im =
Peak magnetizing ampere turns 2T
Im = 1
2T1
where T1 is the number of turns of primary.
where 1 is the number
TheTmagnetic circuitofofturns of primary.
a transformer consists of both iron and air path. The iron path is
dueThe magnetic
to legs circuitand
and yokes of athe
transformer
air path isconsists
due to theof both iron and joints
unavoidable air path. The iron
created path
by the is
core
due to legs of
composed and yokes and
stampings of the air path
different is dueIf to
shapes. allthe
theunavoidable joints created
joints are assumed by the core
to be equivalent to
composed
an air gap ofof stampings
length, lg, of
thendifferent
the totalshapes.
ampere If all the for
turns joints
theare assumed to
transformer be equivalent
magnetic circuittois
an air to
equal gapAToffor
length, , then the
iron +lg800,000 lgBtotal
m . ampere
Therefore, turns for the transformer magnetic circuit is
equal to AT for iron + 800,000 lgBm. Therefore,
Total AT = AT for iron + 800,000 lg Bm
Total AT = AT for iron + 800,000 lg Bm
AT for iron + 800,000 lg Bm
I m = AT for iron + 800,000 lg Bm
Im = 2T1
2T1

Vp

Ic
I0

φº

90º flux φ

Im

fig.
Fig. 3.35
3.35 || No
No load
load current
current and
and its
its components
components

The
The no
no load
load current
current and
and its
its components
components of
of aa transformer
transformer in
in terms
terms of
of mmf,
mmf, number
number ofof
turns,
turns, power and voltage are derived for single- and three-phase transformers as follows.
power and voltage are derived for single- and three-phase transformers as follows.
3.62 Design of Transformer

3.15.1 Single-phase Transformer
Magnetizing current
The peak value of magnetizing current in terms of mmf and number of turns is given by
AT0
I m(pk) = (3.41)
Tp

where AT0 – Total magnetizing mmf under no load condition, Tp – Number of turns in
primary.
In Eq. (3.41),
AT0 = mmf of core + mmf of yoke + mmf of joints  (3.42)
= ATc + ATy + ATj
In the above equation,
(ATc) mmf of core = mmf per metre for maximum flux density in core ×
length of path traced by flux in core
= atc× 2lc = 2atc hw [ ∵ lc = hw (height of window)]
(ATy) mmf of yoke = mmf per metre for maximum flux density in yoke ×
length of path traced by flux in yoke
= aty× 2ly = 2aty ww [ ∵ ly = ww (width of window)]
atc, aty – mmf per metre for maximum flux density in core and yoke, respectively
2lc, 2ly – Total length of core and yoke, respectively
ATj = 800000 lgBm
(As all joints are assumed to be equivalent to an airgap (lg))
From Eq. (I), the rms value of magnetizing current for sinusoidal current is given by
AT0
I m (rms) = (3.43)
2Tp

If the magnetizing current is non-sinusoidal, its rms values are given by

AT0
I m (rms) = (3.44)
kpeak Tp
where kpeak – peak factor
Core loss or iron loss current
The core loss or iron loss component of no load current is given by
Pi or Pc
I c (I i or I l ) =
Vp
where Pi (Pc) – iron loss or core loss, Vp – Primary voltage.

3.15.2 Three-phase Transformer

The magnetizing current derived for single-phase transformer holds good for three-phase
transformer too, but the length of the core is 3lc.
 No Load Current of a Transformer 3.63

3.15.3 Magnetizing Volt – Ampere and Alternate Expression

for ­Magnetizing Current
The magnetizing Volt – Ampere (VA)M is expressed as the product of voltage induced
primary of transformer winding and magnetizing current.
Voltage induced in primary of transformer winding
= 4.44 f φm TP

= 4.44 f Bm Ai TP [ ∵ φm = Bm Ai ]

AT0
Magnetizing current =
2TP

Hence, magnetizing volt ampere,

AT0
( VA)M = 4.44 f Bm Ai Tp ×  (3.45)
2TP

Expressing mmf in terms of mmf per metre and length of iron path, we get

AT0 = ati × li (3.46)

Substituting Eq. (3.46) in Eq. (3.45), we get

4.44 f Bm Ai ati li
( VA)m =
2

The Magnetizing Volt Ampere per kg can be to found out by determining the ratio of
Magnetizing Volt Ampere to the weight of iron, which is given by

Wi = density × volume

We know that the density of iron is 7.8 × 103 kg/m3 and volume is given by product of
area of iron (Ai) and length of iron (li). Hence, using these values in the above equation, we
get
Wi = 7.8 ×10 3 × Ai × li (3.48)

Hence, magnetizing volt ampere per kg, obtained by dividing Eq. (3.47) by Eq. (3.48) is
4.44 f Bm ati
(VA)m / kg =
2 ×7.8 ×10 3
= 0.4025 f Bm ati ×10−3
 0.4 fBm ati ×10−3

Magnetizing VA/kg × weight of core

Magnetizing current, Im =
number of phases × voltage/phase
3.68 Design of Transformer

Eddy loss Pe = Ke V Bm2 f2 t2,

where Ke is a constant whose value depends on the resistivity of the core material used.
Copper loss: These are the losses occurring in the two windings of the transformer on load
conditions. Since these losses vary with the load, they are called load losses.
Calculation of total core loss

Total core loss = core loss in legs + core loss in yokes

The core loss can be estimated at design stage by referring to graph of core loss/kg versus
flux density.

Core loss in leg = loss/kg in leg × weight of leg in kg

= loss/kg in leg × volume of the leg (Ai Hw) ×

density of steel or iron used

Total core loss = Number of legs × loss/kg in leg × volume of the

leg (Ai Hw) × density of steel or iron used

Core loss in yoke = loss/kg in yoke × volume of yoke (Ay × mean

length of the yoke) × density of iron used

The density of iron or steel used for the transformer core lies between 7.55 and 7.8 g/cm3.
Calculation of copper loss
Since this is the I2R loss in the windings, it can be calculated from the resistance and current
rating of the winding.

3.17 Effects of Change in Frequency in Parameters

of the Transformer
The input frequency of the transformer is one of the determining factors of the performance of
the transformer. When the input frequency is changed, the transformer iron loss and hence the
efficiency change. Similarly, the winding resistance, leakage reactance and voltage are affected
by the change in input frequency. The frequency change effects are discussed as follows.
1. Effect on core loss: Let us examine the effect of frequency variation on core loss if the
voltage remains constant.
We have the voltage, E = 4.44fBmAiT and eddy current loss, Pe = Ke f 2 Bm 2

If the voltage E remains constant, then the product (f Bm) remains constant. Thus,
eddy current loss will remain constant as long as E is kept constant, no matter the
(
x
frequency is changed. But the hysteresis loss is proportional to the product f Bm )
and is given as
x
Hysteresis loss, Ph = K h f Bm
The exponent x can take the value from 1.6 to 2.2 depending on the magnetic
characteristics of the core material.
 Optimum Design 3.69

If x is assumed to be 2, then
Ph = K h ( f Bm ) Bm

= K h K Bm
 (∵ E is constant, f Bm is constant and equal to K )
K K2
= Kh K = Kh
f f
Hence, hysteresis loss will decrease with increase in frequency.
2. Effect on winding resistance: The winding resistance increases due to skin effect
when the frequency is increased. This increase is negligible, when the change in
frequency is very small.
3. Effect on leakage reactance: Since the leakage reactance Xl = 2πfl, the leakage
reactance changes linearly with the increase in frequency.
4. Effect on voltage: Along with the change in frequency, if the flux density is also
changed, then a new voltage rating is possible. For the new voltage rating, the iron
loss and no load current can be calculated.

Table 3.9 gives the summary of effects of frequency variation on various parameters
discussed above.

Table 3.9 | Summary of effects of frequency variation on various parameters

Parameter Effect of change in frequency
Core loss Decreases with increase in frequency
with voltage maintained constant
Voltage Increases with increase in frequency

Leakage reactance Increases with increase in frequency

Resistance Increases with increase in frequency

3.18 Optimum Design
A transformer design may be optimized to minimize overall volume, weight, cost and losses,
etc. These requirements are generally conflicting and so it is possible to satisfy any one of the
above at one time.
We know that, from Eq. (3.16), the output equation of single-phase transformer is given by
Q (kVA) = 2.22 f Bm δ K w Aw Ai ×10−3
= 2.22 f Bm δ Ac Ai [ ∵ K w Aw = Ac ]
In the above equation, on assuming f, Bm and δ to be constant, the left out product of
Ac Ai is constant and is assumed to be given by

Ac Ai = y 2 (3.49)
3.70 Design of Transformer

We know that, from Eq. (3.34),

φm
x= (3.50)
AT
where
φm = Bm Ai (3.51)
And from Eq. (3.12),
δ K w Aw
AT =
2
δAc
= (3.52)
2
Substituting Eqs. (3.52) and (3.51) in Eq. (3.50), we get
Bm Ai 2Bm Ai
x= =
δ Ac δ Ac
2
On simplifying,
Ai δx
⇒ = = z (Assumption ) (3.53)
Ac 2Bm

The assumption is made such that z is a function of x, as Bm and δ are assumed to be constants.
Using Eq. (3.53) in Eq. (3.49) to determine Ai and Ac, we get
Substituting Ai = z Ac in Eq. (3.49), we get

zAc × Ac = y 2

⇒ zAc 2 = y 2

y
⇒ Ac = (3.54)
z
Ai
Substituting Ac = in Eq. (3.49), we get
z
Ai
× Ai = y 2
z

⇒ Ai 2 = zy 2

⇒ Ai = y z (3.55)

The total cost of transformer is given by

CT = ci + Cc (3.56)

where Ci – total cost of iron, Cc – total cost of copper conductor

 Optimum Design 3.71

And,
Ci = ci gi li Ai (3.57)

Cc = cc gc lmt Ac (3.58)
where Ci – specific cost of iron, Cc – specific cost
Substituting Eqs. (3.57) and (3.58) in Eq. (3.56), we get

CT = ci gi li Ai + cc gc lmt Ac (3.59)

Substituting Eqs. (3.54) and (3.55) in Eq. (3.59), we get

y
CT = ci gi li y z + cc gc lmt (3.60)
z
In order to determine the minimum cost, differentiating the above equation with respect to
z, and equating to zero we get

dCT 1 y 1 y
= ci gi li − cc gc lmt −3/2 = 0
dz 2 z 2 z
y y
⇒ ci gi li = cc gc lmt
z 3/2
z
On simplifying the above equation, we get
cc gc lmt
ci gi li =
z
Substituting Eq. (3.53) in the above equation, we get
Ac
ci gi li = cc gc lmt
Ai
⇒ ci gi li Ai = cc gc lmt Ac

⇒ ci Gi = cc Gc (3.61)
[ ∵ Gi = gi li Ai and Gc = gc lmt Ac ]
⇒ Ci = Cc (3.62)

Similarly, from Eqs. (3.61) and (3.62),

For minimum volume of transformer,
Volume of iron = Volume of conductor

Gi c
= c
Gc ci
For minimum weight of transformer,

Weight of iron = Weight of conductor

3.72 Design of Transformer

GGi =
i= GGcc
For minimum
For minimum losses
losses of
of transformer
transformer (maximum
(maximum efficiency),
efficiency),
Iron
Iron loss ==
loss Copper
Copper loss
loss

x 2xP2cPc
Pi P=i =

3.19 Cooling
3.19 Cooling of
of Transformer
Transformer
The losses developed in the transformer core and the winding cause heating of the
The losses developed in the transformer core and the winding cause heating of the
transformer parts. Based on the thermal gradient, the heat is transferred to the cooling
transformer parts. Based on the thermal gradient, the heat is transferred to the cooling
medium depending on the method of cooling. In very small transformer of a few kVA rating,
medium depending on the method of cooling. In very small transformer of a few kVA rating,
the cooling surface of the core and winding is sufficient to be cooled by normal circulation
the cooling surface of the core and winding is sufficient to be cooled by normal circulation
of atmospheric air. As the capacity of the transformer increases, the losses also increase, so
of atmospheric air. As the capacity of the transformer increases, the losses also increase, so
it is required to provide better cooling arrangement. The various cooling methods can be
it is required to provide better cooling arrangement. The various cooling methods can be
grouped as follows (Figs. 3.36 and 3.37).
grouped as follows (Figs. 3.36 and 3.37).

Types of
cooling

Oil cooling (Oil

Air cooling (Dry
immersed
type transformer)
transformer)

Natural Forced
Forced cooling Natural cooling
cooling cooling

Air Oil Oil Oil Oil Oil

Oil
Air forced Air natural natural natural forced Forced forced
natural Air blast Air Air Water Air Air Water
natural natural forced forced Natural forced forced

fig.
Fig. 3.36 | Types of cooling

Air
Air Natural
Natural (AN)
(AN)
•
•• This method uses
This method uses the
the ambient
ambient air
air as
as the
the cooling
cooling medium.
medium.
•
•• The natural circulation of surrounding air is
The natural circulation of surrounding air is utilized
utilized to
to carry
carry away
away the
the heat
heat generated
generated
by natural convection.
by natural convection.
•
•• A
A sheet
sheet metal
metal enclosure
enclosure is is used
used toto protect
protect the the windings
windings against
against mechanical
mechanical
damage.
damage.
•
•• This
This method
method is is used
used for
for small
small low-voltage
low-voltage transformers
transformers up up to
to 1.5
1.5 MVA.
MVA.
Air Forced Air Natural (AFAN)
Air Forced Air Natural (AFAN)
•
•• In
In this
this method,
method, air air is
is circulated
circulated through
through the the transformer
transformer withwith the
the help
help of
of aa fan
fan and
and
cooled in a heat exchanger by natural circulation
cooled in a heat exchanger by natural circulation of air. of air.
•
•• This
This method
method is is not
not commonly
commonly used.
used.
 CoolingofofTransformer
Cooling 3.73
Transformer 3.73

Main tank Main tank

Pump Pump

Radiator Radiator
Oil flow tank Oil flow tank

(a) (b) Fan

Main tank Main tank
Tube

Fan

(c) (d)

Radiator
tank
Transformer plain tank

Tank with ribs and

corrugations

(e)

[Link].
3.37 | Cooling
3.37 methods:
| Cooling (a)(a)
methods: OFAN,
OFAN,(b)(b)
OFAF, (c)(c)
OFAF, ONAF, (d)(d)
ONAF, ONAN with
ONAN tubes
with and
tubes
(e) ONAN
and (e) ONAN
3.74 Design of Transformer

Air Blast (AB)

•• In this method, the transformer is cooled by a continuous blast of cool air forced
through the cores and the windings.
•• The air blast is produced by external fans. The improvement in heat dissipation
caused by air blast allows higher specific loadings to be used in air-type transformers.
Oil Natural Air Natural (ONAN)
•• Oil is a better conductor of heat than air.
•• Oil has a high co-efficient of volume expansion with temperature.
•• Oil and air are naturally circulated in this cooling
Oil Natural Air Forced (ONAF)
•• In this method, the oil circulating under natural circulation, transfers the heat to the
tank walls.
•• The transformer tank is made hollow and air is blown through the hollow space to
cool the transformer.
•• However, the normal way of cooling the transformers by air blast is to use radiation
tanks of corrugated or elliptical tubes separated from the transformer tank and cooled
by air blast produced by fans.
Oil Natural Water Forced (ONWF)
•• In this method, copper cooling coils are mounted above the transformer and below
the surface of oil.
•• Water is circulated through the cooling coils to cool the transformer.
•• This method proves to be cheap where a natural water head is already available.
Oil Forced Air Natural (OFAN)
•• In large transformers, the natural circulation of oil is insufficient for cooling the
transformer and forced circulation is employed.
•• Oil is circulated by a motor-driven pump from the top of a transformer tank to an
external cooling plant where the oil is cooled.
Oil Forced Air forced (OFAF)
•• In oil natural–air forced method, the oil circulating under natural heat transfers heat
to tank walls.
•• The air is blown through the empty space to cool the transformer.
•• The oil is cooled in external heat exchanger using air blast produced by fans.
•• Therefore, mixed cooling conditions are used the transformer working with oil
natural air natural conditions upto 50% rating and oil forced air forced conditions at
higher loads.
Oil Forced Water Forced (OFWF)
•• In this method, the pressure of oil is kept higher than that of water.
•• This cooling method is suitable for banks of transformers, but from the system
reliability considerations not more than say three tanks should be connected in one
cooling pump circuit.
•• OFWF is used for transformers designed for hydroelectric plants.
 Temperature Rise in Tank with Tubes 3.75

3.20 Temperature Rise and Design of Cooling System

Tanks with plain walls are used for smaller rating transformers up to 20–30 kVA rating, as
they have sufficient surface area to dissipate the heat. But with the increase in the rating
of the transformer, maintaining the temperature rise within the safe working limit is not
possible with simple plain walled tanks. Hence, tubes and radiators are used. The tank
walls, tubes and radiators dissipate heat by radiation and convection. The modes of heat
transfer from various regions of transformer are given in Table 3.10.

Table 3.10 | Types of heat transfer from various regions of transformer

From region To region Type of heat transfer
Core and winding Outer surface, yoke Conduction
Outer surface, yoke Oil Convection
Oil Tank or cooler walls Convection
Tank or cooler walls Cooling medium (air Convection and radiation
or water)

3.21 Temperature Rise in Plain-walled Tank

The temperature rise should not exceed 55oC for natural cooling, 60oC for forced air cooling
and 65oC for forced water cooling and 45oC for oil cooling. The cooling system dissipates
6 W by radiation and 6.5 W by convection per m2 of tank surface per oC. Generally for
heat dissipation, the bottom surface is not considered. Similarly, the top surface has fittings,
bushings, etc. and is not considered to dissipate heat. So, if the cooling surface of the
transformer tank neglecting the top and bottom surfaces is taken as St m2, then 12.5St watts
per m2 per oC will be dissipated.
Hence, for a plain-walled tank, the temperature rise,

Total loss in watts

θ=
12.5 St

3.22 Temperature Rise in Tank with Tubes

As the capacity of the transformer increases, the losses and the temperature rise increases.
In order to keep the temperature rise within limits, air may have to be blown over the
transformer. This is not advisable as the atmospheric air containing moisture, oil particles, etc.,
may affect the insulation. To overcome the problem of atmospheric hazards, the transformer
is placed in a steel tank filled with oil. The oil conducts the heat from core and coil to the tank
walls. From the tank walls, the heat goes dissipated to the surrounding atmosphere due to
radiation and convection. Further as the capacity of the transformer increases, the increased
losses demand a higher dissipating area of the tank or a bigger sized tank. This calls for more
space, more volume of oil and increases the cost and transportation problems. To overcome
these difficulties, the dissipating area is to be increased by using auxiliary radiator tanks or
by using tubes without increasing the size of the tank.
3.76 Design of Transformer

Let the cooling surface of the tank be St m2. This will dissipate 12.5St W/m2/oC. By
provision of tubes, the total surface area becomes x St m2.
Convection from tube walls = x St 6.5 × 1.35 W = x St 8.8 W ( ∵ 1.35 is due to 35%
increase in heat transfer due to convection by syphoning action of the tubes)

Total dissipation = 12.5 St + x St 8.8 W

The number of tubes can be determined by knowing the surface area of each tube. The
diameter of tubes normally used is 5 cm and spaced 7 to 7.5 cm centre to centre apart.

3.23 Design of Tank with Tubes

Dissipating surface area of tank = St

Dissipating surface area of tubes = xSt

Total loss dissipation in surface area of tank = 12.5 St W/°C

Total loss dissipation in surface area of and cooling tubess = (12.5 + 8.8 x )St W/°C

Total loss dissipition by tubes due to convection = 1.35× 6.5 xSt

= 8.8 xSt W/°C

Total surface area of walls of tank and tubes = St + x St

= (1 + x ) St

Total loss dissipation by in surface area of tank and cooling tubes

Total loss dissipation =
Total surface area of wallls of tank and tubes
(12.5 + 8.8 x ) St
=
(1 + x )St

Temperature rice with tubes,

Total loss
(θ ) =
Specific heat dissipation × surface area
Pi + Pc
=
(12.5 + 8.8 x )St

From the above equation, area of tubes can be found as follows.

Pi + Pc
θ=
12.5St + 8.8 xSt

Pi + Pc
⇒ 12.5 St + 8.8 x St =
θ
 Design of Tank with Tubes 3.77

Pi + Pc
⇒ 8.8 xSt = − 12.5 St
θ

1  Pi + Pc 
Total area of tubes, xSt =  − 12.5St 

8.8  θ 

Length of a tube = lt
Diameter of a tube = dt
Radius of a tube = rt
Area of a tube =πdt lt
Total area of tubes
Total number of tubes, nt =
Area of a tube
1  Pi + Pc 
=  − 12.5St 
8.8 π dt lt  θ 

The dimensions of tank depends on the type and capacity of transformer, voltage rating
and electrical clearance to be provided between the transformer and the tank, clearance to
accommodate the connections and taps, clearance for base and oil above the transformer, etc.
The clearance between different parts depends on rating of the transformer.
Let D = distance between adjacent limbs
De = external diameter of outer winding
Cw = width wise clearance between the outer winding and the tank wall
Cl = length wise clearance between the tank wall and the outer winding
Ch = height wise clearance between the tank wall and the outer winding
The various clearance values for a typical transformer of rating 1000 to 5000 kVA and 11
to 33 kV voltage rating is given in Table 3.11.

Table 3.11 | Clearance values for a typical transformer

Clearances Cw Cl Ch
Values 40 to 80 mm 50 to 125 mm 400 to 600 mm

The height wise clearance includes the clearance of 50 to 60 mm at the base, clearance of
150 to 250 mm above oil and space of about 200 to 250 mm for leads. Figure 3.38 shows the
transformer main dimensions with tank along with various clearances.

3.23.1 Flow Chart for Design of Cooling System

The design procedure for design of the cooling system is given below as a flowchart in
Fig. 3.39. This involves the calculation of number of cooling tubes required and suitably
arranging them around the tank.