1

Slide Courtesy:
Dr. Gleb V. Tcheslavski
Associate Professor
Dept. of Electrical Engineering
Lamar University, Texas, USA
 2

A transformer is a device that converts one AC voltage to another AC voltage at

the same frequency. It consists of one or more coil(s) of wire wrapped around a
common ferromagnetic core. These coils are usually not connected electrically
together. However, they are connected through the common magnetic flux
confined to the core.

Assuming that the transformer has at

least two windings, one of them
(primary) is connected to a source of
AC power; the other (secondary) is
connected to the loads.
 3

Power transformers
Core form Shell form

Windings are wrapped around two Windings are wrapped around the
sides of a laminated square core. center leg of a laminated core.

Usually, windings are wrapped on top of each other to decrease flux leakage
and, therefore, increase efficiency.
 4

Lamination
types

Laminated steel cores Toroidal steel cores

Efficiency of transformers with toroidal cores is usually higher.

 5

We consider a lossless transformer

with an input (primary) winding
having Np turns and a secondary
winding of Ns turns.

The relationship between the voltage

applied to the primary winding vp(t)
and the voltage produced on the
secondary winding vs(t) is

v p (t ) Np
 a (4.8.1)
vs (t ) Ns

Here a is the turn ratio of the transformer.

 6

The relationship between the primary ip(t) and secondary is(t) currents is

i p (t ) 1
 (4.9.1)
is (t ) a

In the phasor notation:

Vp
a (4.9.2)
Vs

Ip 1
 (4.9.3)
Is a

The phase angles of primary and secondary voltages are the same. The phase
angles of primary and secondary currents are the same also. The ideal
transformer changes magnitudes of voltages and currents but not their angles.
 7

Assuming that p and s are the angles between voltages and currents on the
primary and secondary windings respectively, the power supplied to the
transformer by the primary circuit is:

Pin  V p I p cos  p (4.11.1)

The power supplied to the output circuits is

Pout  Vs I s cos  s (4.11.2)

Since ideal transformers do not affect angles between voltages and currents:

 p  s   (4.11.3)

Both windings of an ideal transformer have the same power factor.

 8

Since for an ideal transformer the following holds:

Vp
Vs  ;I s  aI p (4.12.1)
a

Therefore: Vp
Pout  Vs I s cos   aI p cos   V p I p cos   Pin (4.12.2)
a
The output power of an ideal transformer equals to its input power – to be
expected since assumed no loss. Similarly, for reactive and apparent powers:

Qout  Vs I s sin   Vp I p sin   Qin (4.12.3)

Sout  Vs I s  Vp I p  Sin (4.12.4)

 9

The impedance is defined as a following ratio of phasors:

Z L  VL I L (4.13.1)

A transformer changes voltages and currents and, therefore, an apparent

impedance of the load that is given by
Z L  Vs I s (4.13.2)

The apparent impedance of the primary

circuit is:
Z L '  Vp I p (4.13.3)

which is
Vp aVs 2 Vs
ZL '   a  a2 Z L (4.13.4)
I p Is a Is
It is possible to match magnitudes of impedances (load and a transmission line) by
selecting a transformer with the proper turn ratio.
 10

A simple method to analyze a circuit containing an ideal transformer is by

replacing the portion of the circuit on one side of the transformer by an
equivalent circuit with the same terminal characteristics.

Next, we exclude the transformer from the circuit and solve it for voltages
and currents.

The solutions obtained for the portion of the circuit that was not replaced
will be the correct values of voltages and currents of the original circuit.

Finally, the voltages and currents on the other side of the transformer (in
the original circuit) can be found by considering the transformer’s turn ratio.

This process is called referring of transformer’s sides.

 11

Example 4.1: A single-phase power system consists of a 480-V 60-Hz generator

that is connected to the load Zload = 4 + j3  through the transmission line with
Zline = 0.18 + j0.24 . a) What is the voltage at the load? What are the
transmission line losses? b) If a 1:10 step up transformer and a 10:1 step down
transformer are placed at the generator and the load ends of the transmission
line respectively, what are the new load voltage and the new transmission line
losses?
a) Here:
V
I G  I line  I load 
Zline  Zload
4800

0.18  j 0.24  4  j 3
4800
  90.8  37.8 A
5.2937.8
 12

Therefore, the load voltage:

Vload  Iload Zload   90.8  37.8  (4  j3)   90.8  37.8 536.9   454  0.9V

The line losses are:

Ploss  Iline
2
Rline  90.82  0.18  1484W
b) We will
1) eliminate transformer T2
by referring the load
over to the transmission
line’s voltage level.
2) Eliminate transformer T1
by referring the
transmission line’s
elements and the equivalent load at the transmission line’s voltage over to the
source side.
 13

The load impedance when referred to the transmission line (while the
transformer T2 is eliminated) is:
2
 10 
'
Zload  a22 Zload     4  j3  400  j300
1
The total impedance on the
transmission line level is
Z eq  Zline  Zload
'

 400.18  j 300.24
 500.336.88

The total impedance is now referred across T1 to the source’s voltage level:
2
1
Z'eq  a12 Zeq     500.336.88   5.00336.88
 10 
 14

The generator’s current is

V 4800
IG  
Z'eq 5.00336.88
 95.94  36.88 A
Knowing transformers’ turn ratios, we
can determine line and load currents:
Iline  a1IG  0.1  95.94  36.88   9.594  36.88 A
Iload  a2Iline  10   9.594  36.88   95.94  36.88 A
Therefore, the load voltage is:
Vload  Iload Zload   95.94  36.88  5  36.87   479.7  0.01V
The losses in the line are: Ploss  Iline
2
Rline  9.5942  0.18  16.7W
Note: transmission line losses are reduced by a factor nearly 90, the load voltage is
much closer to the generator’s voltage – effects of increasing the line’s voltage.
 15

Real transformers approximate ideal ones to some degree.

The basis transformer operation can

be derived from Faraday’s law:
d
eind  N (4.19.1)
dt

(4.19.2)

(4.19.3)

(4.19.4)
 16

If the source voltage vp(t) is applied to the primary winding, the average flux in the
primary winding will be:
1
 
Np  v p (t )dt (4.20.1)

A portion of the flux produced in

the primary coil passes through
the secondary coil (mutual flux);
the rest is lost (leakage flux):

 p  m  Lp (4.20.2)

average primary flux mutual flux

Similarly, for the secondary coil:

s  m  Ls (4.20.3)

Average secondary flux

 17

Therefore:

e p (t ) Np
 a (4.22.1)
es (t ) Ns

That is, the ratio of the primary voltage to the secondary voltage both caused by
the mutual flux is equal to the turns ratio of the transformer.

For well-designed transformers:

m Lp ;m Ls (4.22.2)

Therefore, the following approximation normally holds:

v p (t ) Np
 a (4.22.3)
vs (t ) Ns
 18

Even when no load is connected to the secondary coil of the transformer, a current
will flow in the primary coil. This current consists of:
1. The magnetization current im needed to produce the flux in the core;
2. The core-loss current ih+e hysteresis and eddy current losses.

Flux causing the

magnetization current

Typical magnetization curve

 19

Ignoring flux leakage and assuming time-harmonic primary voltage, the average
flux is:
1 1 Vm
sin tWb
Np  Np 
  v (t ) dt  V cos  tdt  (4.24.1)
p m
N p
If the values of current are comparable to the flux they produce in the core, it is
possible to sketch a magnetization current. We observe:
1. Magnetization current is not sinusoidal: there are high frequency components;
2. Once saturation is reached, a small increase in flux requires a large increase
in magnetization current;
3. Magnetization current (its fundamental component) lags the voltage by 90o;
4. High-frequency components of the current may be large in saturation.
Assuming a sinusoidal flux in the core, the eddy currents will be largest when
flux passes zero.
 20

total excitation current in a transformer

Core-loss current

Core-loss current is:

1. Nonlinear due to nonlinear effects of hysteresis;

2. In phase with the voltage.

The total no-load current in the core is called the excitation current of the
transformer:

iex  im  ih e (4.25.1)
 21

The last approximation is valid for well-designed unsaturated cores. Therefore:

ip Ns 1
N pi p  N sis    (4.27.1)
is Np a
An ideal transformer (unlike the real one) can be
characterized as follows:

1. The core has no hysteresis or eddy currents.

2. The magnetization curve is
3. The leakage flux in the core is zero.
Magnetization curve of
4. The resistance of the windings is zero. an ideal transformer
 22

To model a real transformer accurately, we need to account for the

following losses:

1. Copper losses – resistive heating in the windings: I2R.

2. Eddy current losses – resistive heating in the core: proportional to
the square of voltage applied to the transformer.
3. Hysteresis losses – energy needed to rearrange magnetic domains
in the core: nonlinear function of the voltage applied to the
transformer.
4. Leakage flux – flux that escapes from the core and flux that passes
through one winding only.
5. Magnetizing current – Needed to produce flux in the core
 23

Cooper losses are modeled

by the resistors Rp and Rs.
Leakage flux in a primary
winding produces the
voltage:
dLp
eLp (t )  N p (4.29.1)
dt
Since much of the leakage flux pass through air, and air has a constant reluctance
that is much higher than the core reluctance, the primary coil’s leakage flux is:

Lp  P N p i p (4.29.2)
permeance of flux path
Therefore:
eLp (t )  N p P N p i p   N pP
d 2
di p
(4.29.3)
dt dt
 24

Recognizing that the self-inductance of the primary coil is

Lp  N p2P (4.30.1)

The induced voltages are:

di p
Primary coil: eLp (t )  Lp (4.30.2)
dt
dis
Secondary coil: eLs (t )  Ls (4.30.3)
dt
The leakage flux can be modeled by primary and secondary inductors.
The magnetization current can be modeled by a reactance XM connected across
the primary voltage source.
The core-loss current can be modeled by a resistance RC connected across the
primary voltage source.
Both currents are nonlinear; therefore, XM and RC are just approximations.
 25

The transformer’s
equivalent circuit
However, the exact circuit is not
very practical.

Therefore, the equivalent circuit is usually

referred to the primary side or the secondary
side of the transformer.

Equivalent circuit of the transformer

referred to its primary side.

Equivalent circuit of the transformer

referred to its secondary side.
 26

For many practical applications,

approximate models of
transformers are used.

Referred to the primary side.

Referred to the secondary side.

Without an excitation branch

referred to the primary side.

The values of components of the

transformer model can be Without an excitation branch
determined experimentally by an referred to the secondary side.
open-circuit test or by a short-circuit test.
 27

The open-circuit test.

Full line voltage is applied to the primary
side of the transformer. The input voltage,
current, and power are measured.

From this information, the power factor of the input current and the magnitude and
the angle of the excitation impedance can be determined.
To evaluate RC and XM, we determine the conductance of the core-loss resistor is:
1
GC  (4.33.1)
RC
The susceptance of the magnetizing inductor is:
1
BM  (4.33.2)
XM
 28

Since both elements are in parallel, their admittances add. Therefore, the total
excitation admittance is:
1 1
YE  GC  jBM  j (4.34.1)
RC XM
The magnitude of the excitation admittance in the open-circuit test is:

I oc
YE  (4.34.2)
Voc
The angle of the admittance in the open-circuit test can be found from the circuit
power factor (PF):
Poc
cos   PF  (4.34.3)
Voc I oc
 29

In real transformers, the power factor is always lagging, so the angle of the current
always lags the angle of the voltage by  degrees. The admittance is:
I oc I
YE      oc   cos 1 PF (4.35.1)
Voc Voc

Therefore, it is possible to determine values of RC and XM in the open-circuit test.

 30

The short-circuit test.

Fairly low input voltage is applied to the
primary side of the transformer. This voltage
is adjusted until the current in the secondary
winding equals to its rated value.
The input voltage, current, and power are again measured.
Since the input voltage is low, the current flowing through the excitation branch is
negligible; therefore, all the voltage drop in the transformer is due to the series
elements in the circuit. The magnitude of the series impedance referred to the
primary side of the transformer is:
VSC
Z SE  (4.36.1)
I SC
The power factor of the current is given by:
PSC
PF  cos   (4.36.2)
VSC I SC
 31

Therefore:
VSC 0 VSC
Z SE     (4.37.1)
I SC     I SC
Since the serial impedance ZSE is equal to

Z SE  Req  jX eq (4.37.2)

Z SE   Rp  a 2 RS   j  X p  a 2 X S  (4.37.3)

it is possible to determine the total series impedance referred to the primary side
of the transformer. However, there is no easy way to split the series impedance
into primary and secondary components.
The same tests can be performed on the secondary side of the transformer. The
results will yield the equivalent circuit impedances referred to the secondary
side of the transformer.
 32

Example 4.2: We need to determine the equivalent circuit impedances of a 20

kVA, 8000/240 V, 60 Hz transformer. The open-circuit and short-circuit tests led to
the following data:
VOC = 8000 V VSC = 489 V
IOC = 0.214 A ISC = 2.5 A
POC = 400 W PSC = 240 W

The power factor during the open-circuit test is

POC 400
PF  cos     0.234lagging
VOC I OC 8000  0.214
The excitation admittance is
I OC 1 0.214 1 1 1
YE    cos PF    cos 0.234  0.0000063  j 0.0000261   j
VOC 8000 RC XM
 33

Determining the values of components:

Example
1 1
Therefore: RC   159k ; X M   38.3k 
0.0000063 0.0000261
The power factor during the short-circuit test is
PSC 240
PF  cos     0.196lagging
VSC I SC 489  2.5
The series impedance is given by
VSC 489
Z SE   cos 1 PF  78.7
I SC 2.5
 38.4  j192
Therefore:
Req  38.3; X eq  192
The equivalent circuit
 34

Another approach to solve circuits containing transformers is the per-unit system.

Impedance and voltage-level conversions are avoided. Also, machine and
transformer impedances fall within fairly narrow ranges for each type and
construction of device while the per-unit system is employed.

The voltages, currents, powers, impedances, and other electrical quantities are
measured as fractions of some base level instead of conventional units.
actualvalue
Quantity perunit (4.40.1)
basevalueof quantity
Usually, two base quantities are selected to define a given per-unit system. Often,
such quantities are voltage and power (or apparent power). In a 1-phase system:
Pbase ,Qbase ,orSbase  Vbase I base (4.40.2)

Vbase Vbase 
2

Z base   (4.40.3)
I base Sbase
 35

I base
Ybase  (4.41.1)
Vbase

Ones the base values of P (or S) and V are selected, all other base values can
be computed form the above equations.
In a power system, a base apparent power and voltage are selected at the
specific point in the system. Note that a transformer has no effect on the
apparent power of the system, since the apparent power into a transformer
equals the apparent power out of a transformer. As a result, the base apparent
power remains constant everywhere in the power system.
On the other hand, voltage (and, therefore, a base voltage) changes when it
goes through a transformer according to its turn ratio. Therefore, the process
of referring quantities to a common voltage level is done automatically in the
per-unit system.
 36

Example 4.3: A simple power system is given by the circuit:

The generator is rated at 480 V and 10 kVA.

a) Find the base voltage, current, impedance, and apparent power at every
points in the power system;
b) Convert the system to its per-unit equivalent circuit;
c) Find the power supplied to the load in this system;
e) Find the power lost in the transmission line (Region 2).
 37

a. In the generator region: Vbase 1 = 480 V and Sbase = 10 kVA

Sbase1 10000
I base1    20.83 A
Vbase1 480
Vbase1 480
Z base1    23.04
I base1 20.83
The turns ratio of the transformer T1 is a1 = 0.1; therefore, the voltage in the
transmission line region is
Vbase1 480
Vbase2    4800V
a1 0.1
The other base quantities are
 38

Sbase2  10kVA
10000
I base2   2.083 A
4800
4800
Z base2   2304
2.083

The turns ratio of the transformer T2 is a2 = 20; therefore, the voltage in the
load region is
Vbase 4800
Vbase    240V
a2 20
The other base quantities are
 39

Sbase  10kVA
10000
I base   41.67 A
240
240
Z base   5.76
41.67

b. To convert a power system to a per-unit system, each component must be

divided by its base value in its region. The generator’s per-unit voltage is
4800
VG , pu   1.00 pu
480
The transmission line’s per-unit impedance is
20  j 60
Z line , pu   0.0087  j 0.026 pu
2304
 40

The load’s per-unit

impedance is

1030
Zload , pu 
5.76
 1.73630 pu
The per-unit
equivalent circuit

c. The current flowing in this per-unit power system is

Vpu 10
I pu    0.569  30.6 pu
Ztot , pu 0.0087  j 0.026  1.73630
 41

Therefore, the per-unit power on the load is

Pload , pu  I pu
2
Rpu  0.5692 1.503  0.487
The actual power on the load is

Pload  Pload , pu Sbase  0.487 10000487W

d. The per-unit power lost in the transmission line is

Pline, pu  I pu
2
Rline, pu  0.5692  0.0087  0.00282
The actual power lost in the transmission line

Pline  Pline, pu Sbase  0.00282 100008.2W

 42

Example 4.4: Sketch the appropriate per-unit equivalent circuit for the 8000/240 V, 60
Hz, 20 kVA transformer with Rc = 159 k, XM = 38.4 k, Req = 38.3 , Xeq = 192 .

To convert the transformer to per-unit system, the primary circuit base impedance
needs to be found.
Vbase1  8000V ;Sbase1  20000VA
2
Vbase 80002
Z base1  1
  3200
Sbase1 20000
38.4  j192
Z SE , pu   0.012  j 0.06 pu
3200
159000
RC , pu   49.7 pu
3200
00
X M , pu   12 pu
3200
 43

Therefore, the per-unit equivalent circuit is shown below:

 44

Since a real transformer contains series impedances, the transformer’s output

voltage varies with the load even if the input voltage is constant. To compare
transformers in this respect, the quantity called a full-load voltage regulation (VR)
is defined as follows:

Vs ,nl  Vs , fl Vp a  Vs , fl
VR  100%  100% (4.52.1)
Vs , fl Vs , fl
In a per-unit system:
Vp , pu  Vs , fl , pu
VR  100% (4.52.2)
Vs , fl , pu

Where Vs,nl and Vs,fl are the secondary no load and full load voltages.
Note, the VR of an ideal transformer is zero.
 45

To determine the VR of a transformer, it is necessary to understand the voltage

drops within it. Usually, the effects of the excitation branch on transformer VR can
be ignored and, therefore, only the series impedances need to be considered. The
VR depends on the magnitude of the impedances and on the current phase angle.
A phasor diagram is often used in the VR determinations. The phasor voltage Vs is
assumed to be at 00 and all other voltages and currents are compared to it.

Considering the diagram and by applying the

Kirchhoff’s voltage law, the primary voltage is:

Vp
 Vs  Req I s  jX eq I s (4.53.1)
a
A transformer phasor diagram is a graphical
representation of this equation.
 46

A transformer operating at a lagging power factor:

It is seen that Vp/a > Vs, VR > 0

A transformer operating at
a unity power factor:
It is seen that VR > 0

A transformer operating at a
leading power factor:
If the secondary current is leading,
the secondary voltage can be higher
than the referred primary voltage;
VR < 0.
 47

The efficiency of a transformer is defined as:

Pout Pout
 100%  100% (4.55.1)
Pin Pout  Ploss
Note: the same equation describes the efficiency of motors and generators.
Considering the transformer equivalent circuit, we notice three types of losses:
1. Copper (I2R) losses – are accounted for by the series resistance
2. Hysteresis losses – are accounted for by the resistor Rc.
3. Eddy current losses – are accounted for by the resistor Rc.

Since the output power is Pout  Vs I s cos  s

The transformer efficiency is
Vs I s cos 
 100% (4.55.3)

PCu  Pcore  Vs I s cos 

 48

Example 4.5: A 15 kVA, 2300/230 V transformer was tested to by open-circuit and

closed-circuit tests. The following data was obtained:

VOC = 2300 V VSC = 47 V

IOC = 0.21 A ISC = 6.0 A

POC = 50 W PSC = 160 W

a. Find the equivalent circuit of this transformer referred to the high-voltage side.
b. Find the equivalent circuit of this transformer referred to the low-voltage side.
c. Calculate the full-load voltage regulation at 0.8 lagging power factor, at 1.0
power factor, and at 0.8 leading power factor.
d. Plot the voltage regulation as load is increased from no load to full load at
power factors of 0.8 lagging, 1.0, and 0.8 leading.
e. What is the efficiency of the transformer at full load with a power factor of 0.8
lagging?
 49

a. The excitation branch values of the equivalent circuit can be determined as:
Poc 50
 oc  cos 1
 cos 1
 84
Voc I oc 2300  0.21
The excitation admittance is:
I oc 0.21
YE    84    84  0.0000095  j 0.0000908S
Voc 2300
The elements of the excitation branch referred to the primary side are:
1
Rc   105k 
0.0000095
1
XM   11k 
0.0000908
 50

From the short-circuit test data, the short-circuit impedance angle is

PSC 160
 SC  cos 1  cos 1  55.4
VSC I SC 47  6
The equivalent series impedance is thus
VSC 47
Z SE   SC  55.4  4.45  j 6.45
I SC 6
The series elements referred to
the primary winding are:

Req  4.45; X eq  6.45

The equivalent circuit

 51

b. To find the equivalent circuit referred to the low-voltage side, we need to divide
the impedance by a2. Since a = 10, the values will be:

RC  1050 X M  110 Req  0.0445 X eq  0.0645

The equivalent circuit will be
 52

c. The full-load current on the secondary side of the transformer is

Srated 15000
I S ,rated    65.2 A
VS ,rated 230
Vp
Since:  VS  Req I S  jX eq I S
a
1
At PF = 0.8 lagging, current I s  65.2  cos (0.8)  65.2  36.9 A
Vp
and  2300  0.0445   65.2  36.9   j 0.0645   65.2  36.9   234.850.40V
a
The resulting voltage regulation is, therefore:
V p a  VS , fl
VR  100%
VS , fl
234.85  230
 100%
230
 2.1%
 53

At PF = 1.0, current I s  65.2 cos1 (1.0)  65.20 A

Vp
and  2300  0.0445   65.20   j 0.0645   65.20   232.941.04V
a

The resulting voltage regulation is, therefore:

Vp a  VS , fl 232.94  230
VR  100%  100%  1.28%
VS , fl 230
 54

At PF = 0.8 leading, current I s  65.2 cos1 (0.8)  65.236.9 A

Vp
and  2300  0.0445   65.236.9   j 0.0645   65.236.9   229.851.27V
a

The resulting voltage regulation is, therefore:

Vp a  VS , fl 229.85  230
VR  100%  100%  0.062%
VS , fl 230
 55

e. To find the efficiency of the transformer, first calculate its losses.

The copper losses are:
PCu  I S2 Req  65.22  0.0445  189W
The core losses are:
Vp a 
2
234.852
Pcore    52.5W
RC 1050
The output power of the transformer at the given Power Factor is:
Pout  VS I S cos   230  65.2  cos 36.9  12000W
Therefore, the efficiency of the transformer is

Pout
 100%  98.03%
PCu  Pcore  Pout