View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

provided by UTHM Institutional Repository

PREDICTING TRANSFORMER END OF LIFE USING TRANSFORMER

THERMAL LIFE SIMULATION TECHNIQUE

FERASS A. MUFTAH SAAD

A thesis is submitted of the fulfillment of the requirements

for the award of the degree of Master of Electrical
Engineering

Faculty of Electrical and Electronics Engineering

University Tun Hussein Onn Malaysia (UTHM)

April 2011
 V

ABSTRACT

Large power transformers are key components in power system networks and their
correct functioning is vital to system operation. Transformer failures can have enormous
impact on security and reliability of supply and cost. In addition, Power transformers
ageing are one of the critical issues utilities are facing, since a large number of units are
approaching or have exceeded their designed lifetime. Their replacement will involve a
considerable amount of time and cost. Therefore, developing a replacement strategy for
aged transformer populations is crucial.

This theses presents simulation technique for life assessment of the insulation of
the distribution transformer. Load and ambient temperatures are two important factors
that influence the life of insulation in transformers. The estimated load factors and
ambient temperatures are input to the IEC life consumption models to assess the
consumed life of insulation. The simulation is based on IEC 60076-7 Revision 2005
thermal model. Sumer and winter load profiles and ambient temperatures of the
transformer in Libya Electricity Company are used as the input for the simulation
together with transformer parameters from the heat run test. Hottest spot temperature
and loss-of-life are calculated. Besides the temperature, the moisture factor also has
been introduced in loss of life calculation.
 VI

ABSTRAK

Transformer merupakan komponen yang utama dalam jaringan elektrik kuasa.

Keupayaan transformer untuk beroperasi dengan baik sangat penting dalam
memastikan sistem operasi elektrik berfungsi dengan baik. Kerosakan kepada
transformer boleh memberi kesan kepada sekuriti dan reliabiliti kepada bekalan elektrik
serta juga kos untuk menggantinya. Masalah penuaan kepada transformer merupakan
isu kritikal yang dihadapi oleh utiliti. Ini adalah kerana hampir keseluruhan transformer
akan tamat jangka hayat dalam beberapa tahun lagi dan ada sebahagiannya telah
melebihi jangka hanyat yang telah ditetapkan. Keadaan ini akan memberi kesan kepada
utiliti kerana penggantian transformer-transformer ini akan melibatkan kos yang tinggi
dan juga mengambil masa yang lama. Oleh sebab itu, sangat penting bagi utility untuk
membina satu kaedah ataupun strategi untuk menukarkan kesemua transformer ini
supaya ia tidak menyebabkan gangguan bekalan elektrik kepada pelanggan.

Tesis ini menerangkan teknik simulasi untuk menentukan jangka hayat untuk
transformer yang digunakan dalam pengedaran ekektrik. Dalam simulasi ini, beban
elektrik dan suhu udara merupakan dua faktor yang penting dalam menjangka hayat
untuk transformer. Simulasi teknik menggunakan transformer model yang terdapat
didalam standard antarabangsa IEC 60076-7. Suhu pada musim panas dan musim sejuk
untuk transformer di Libya Electircity Company telah digunakan sebagai parameter
kepada simulasi yang dijalankan bersama parameter dari heat run test. Suhu transformer
yang paling panas digunakan untuk mengira jumlah jangka hayat yang tinggal bagi
sesebuah transformer. Selain daripada suhu, faktor air juga telah diambil kira dalam
proses mengira jangka hayat untuk transformer bagi mendapat keputusan yang lebih
baik
 VII

CONTENTS

TITLE i
DECLARATION ii
DEDICATION iii
ACKNOWLEDGEMENT iv
ABSTRACT v
ABSTRAK vi
CONTENTS vii
LIST OF TABLES x
LIST OF FIGURES xi
LIST OF SYMBOLS xiii
LIST OF APPENDIX xv
CHAPTER 1 INTRODUCTION 1
1.1 Project Background 1
1.2 Problem Statement 2
1.3 Project Objectives 3
1.4 Project Scopes 3
1.5 Outline of thesis 4
CHAPTER 2 LITERATURE REVIEW 6
2.1 Transformer insulation life 6
2.1.1 Ageing factors 7
2.1.1.1 Hydrolysis (water) 8
2.1.1.2 Oxidation (oxygen) 8
2.1.1.3 Pyrolysis (heat) 9
 VIII

2.2 IEC thermal ageing equation 9

2.3 End of life criteria 10
2.4 Transformer thermal 13
2.4.1 Introduction 13
2.4.2 Transformer thermal diagram 15
2.5 Direct hotspot measurement 18
2.6 Transformer cooling system 19
2.6.1 Type of cooling transformer 20
2.6.1.1 Oil Natural and Air Natural Cooling (ONAN) 20
2.6.1.2 Oil Natural and Air Forced Cooling (ONAF) 21
2.6.1.3 Oil Forced and Air Forced Cooling (OFAF) 22
2.6.1.4 Oil Direct and Air Forced Cooling (ODAF) 23
2.7 IEC 60076- 7 thermal model 24
CHAPTER 3 METHODOLOGY 28
3.1 Introduction 28
3.2 Methodology flowchart 29
3.3 Information collection 30
3.4 The simulation model 30
3.4.1 The top oil rise model 30
3.4.2 The hotspot rise model 31
3.4.2.1 The hotspot rise component 1 model 32
3.4.2.2 The hotspot rise component 2 model 33
3.4.3 Hotspot temperature model 33
3.4.4 Loss of life model 34
3.5 Life time estimation 35
CHAPTER 4 RESULT AND ANALYSIS 36
4.1 Introduction 36
4.2 Calculation the initial conditions 37
4.2.1 Calculation The summer initial condition 37
4.2.2 Calculation The winter initial condition 38
 IX

4.3 Summer and winter daily load profiles 39

Summer and winter daily ambient temperature 40
4.4
profiles
4.5 Simulation results 40
Summer and winter daily top oil temperature 41
4.5.1
rise simulation
Summer and winter daily winding hotspot 41
4.5.2
temperature rise simulation
Summer and winter daily hotspot temperature 43
4.5.3
simulation
4.5.4 Summer and winter daily loss of life simulation 44
Simulation study on variations of daily loss of 45
4.6
life
Lifetime estimation for the 20MVA, 66/11KV 48
4.7
power transformer
4.7.1 Secondary ageing factors 50
4.7.1.1 Derivation of moisture ageing factor 51
4.7.1.2 Derivation of acidity ageing factor 55
Lifetime estimation with the secondary ageing 58
4.7.2
factors in calculation
CHAPTER 5 CONCLUSIONS AND RECOMMANDATIONS 60
5.1 Conclusions 60
5.2 Recommendations 62
REFERENCES 63
APPENDIX 67
 X

LIST OF TABLES

2.1 Normal insulation life value for a well dried oxygen free system at 12
the reference temperature 110 0C
2.2 Exponents used in temperature calculation 18
2.3 IEC Recommended thermal parameters for differential equations solution 26
4.1 the transformer parameters 37
4.2 Loss of life comparison for summer profile 46
4.3 Loss of life comparison for winter profile 47
4.4 Lifetime expectancy for the 20MVA, 66/11KV transformer 49
4.5 pre-exponential factor (A) estimated for activation energy of 111 52
KJ/mol from Emsley’s results
4.6 pre-exponential factor (A) estimated for activation energy of 111 52
KJ/mol from Lundgaard’s results
4.7 Estimation of moisture ageing factor with respect to the moisture 53
content in the paper
4.8 The acidity factor with acidity content in paper 56
4.9 Transformer expected life based on DP=200 with operation 58
temperature of 600C
4.10 Transformer expected life based on 50 % tensile strength retention 58
with operation temperature of 600C
 XI

LIST OF FIGURES

2.1 Cellulose molecule structure 7

2.2 Transformer losses classification 14
2.3 Transformer Thermal Diagram 16
2.4 ONAN cooling diagram 20
2.5 ONAF cooling diagram 22
2.6 OFAF cooling diagram 23
2.7 ODAF cooling diagram 24
2.8 Block diagram representation of differential equations 25
3.1 Methodology flowchart 29
3.2 Block diagram of top oil rise model 31
3.3 Block diagram of the sum of two differential hotspot 32
equations
3.4 Block diagram of HSR1 model 32
3.5 Block diagram of HSR2 model 33
3.6 Block diagram of Hotspot temperature model 34
3.7 Block diagram of Loss of life model 34
4.1 Summer and winter daily load profiles for the 20MVA, 39
66/11KV transformer
4.2 Summer and winter daily ambient temperature for the 40
20MVA, 66/11KV transformer
4.3 Summer and winter daily top oil temperature rise for the 41
20MVA, 66/11KV transformer
4.4 Summer and winter daily winding hotspot temperature rise 42
 XII

component1 for the 20MVA, 66/11KV transformer

4.5 Summer and winter daily winding hotspot temperature rise 42
component2 for the 20MVA, 66/11KV transformer for grade
C25 and C30
4.6 Summer and winter daily winding hotspot temperature rise for 43
the 20MVA, 66/11KV transformer
4.7 Summer and winter daily hotspot temperature for the 20MVA, 44
66/11KV transformer
4.8 Summer and winter daily loss of life for the 20MVA, 45
66/11KV transformer
4.9 Loss of life comparison for summer data 48
4.10 Loss of life comparison for winter data 48
4.11 Moisture content in paper with time for 60 0C and 70 0C 54
temperature
4.12 Moisture ageing factor with time for 60 0C and 70 0C 55
temperature
4.13 Acidity in oil versus time at different temperature 56
4.14 Acidity in oil with time for temperature of 60 0C and 70 0C 57
0 0
4.15 Acidity ageing factor with time for 60 C and 70 C 57
temperature
4.16 Comparison between moisture factor and acidity factor with 59
time at average temperature of 60 0C
 XIII

LIST OF SYMBOLS AND ABBREVIATIONS

AC - Acidity content in paper

AF - Acidity ageing factor
AFLL - Acidity factor for lower limit
AFUL - Acidity factor for upper limit
B - An ageing rate constant
DP - Degree of molecular polymerization
DPold - Final DP value
DPnew - Initial DP value
FAA - Ageing acceleration factor
g - Is the average winding to average oil temperature rise at rated load
H - Hotspot factor
HS - Hotspot temperature, 0C
HSR - Winding hotspot rise over top oil temperature, 0C
HSRR - Winding hotspot temperature rise at rated load, 0C
HSR1 - Winding temperature rise component 1, 0C
HSR2 - Winding temperature rise component 2, 0C
I - Current, Amps
K - Load factor, per unit
K11,k22,k21 - Constant
L - Lifetime
m - Mass flow rate kg/s
m - Winding constant (IEEE)
M - A constant
 XIV

MC - Moisture content in the paper

MF - Moisture ageing factor
MFUL - Moisture ageing factor for upper limit
MFLL - Moisture ageing factor for lower limit
n - Oil constant (IEEE)
OF - The oxygen ageing factor
ONAN - Oil Natural and Air Natural
ONAF - Oil Natural and Air Forced
OFAF - Oil Forced and Air Forced
ODAF - Oil Direct and Air Forced
P - Material constant
PEC - Eddy current losses, Watts
PT - The total losses in transformer, Watts
P0 - No-load losses, Watts
PS - Load losses, Watts
PSL - Stray losses, Watts
Q - Heat flow, Watts
R - Ratio of load loss to no load loss
S - The laplace variable
TA - Ambient temperature, 0C
TO - Top oil temperature, 0C
TOR - Top oil temperature rise over ambient, 0C
TORR - The top oil temperature rise over ambient at the rated load, 0C
V - The relative ageing rate
x - The oil constant (IEC)
y - The winding constant (IEC)
Ѳ - The absolute temperature, k
τo - Oil time constant, minute
τw - Winding time constant, minute
 XV

LIST OF APPENDICES

APPENDIX TITLE PAGE

A The model block input parameters designed to enter the 67

transformer parameters to calculate the hot spot temperature
B Simulink/ Matlab model for transformer hotspot temperature 68
calculation
C Simulink/ Matlab model for transformer loss of life calculation 69
 1

CHAPTER 1

INTRODUCTION

1.1 Project Background

A power transformer is a static piece of apparatus with two or more windings which,
by electromagnetic induction, transforms a system of alternating voltage and current
into another system of voltage and current usually of different values and at same
frequency for the purpose of transmitting electrical power (GECOL, 2007).

Power transformers represent the largest portion of capital investment in

transmission and distribution substations. In addition, power transformer outages
have a considerable economic impact on the operation of an electrical network
(Dejan Susa, 2007). The life of a transformer depends mainly on the life of its
insulation system and it is primarily depend on transformer operating temperature
and time.

In fact, most of power transformers use paper and oil as the main of
insulation. There are three possible mechanisms that contribute to the insulation
degradation are hydrolysis, oxidation and pyrolysis. The agents responsible for the
respective mechanisms are water, oxygen and heat (temperature) (M.T.Ishak, 2009).
Indeed the temperature has been considered as the main parameter affecting the loss-
of- life of insulation. Hence, the heat produced (internal temperature) in the
transformers as a result of loading and the effect of ambient temperature is the
important factor that affecting the life of the transformer.
 2

Since the temperature distribution is not uniform in a transformer, the part

that is operating at the highest temperature will usually be considered in estimating
transformer insulation life since it will undergo the greatest deterioration. This
temperature is referred as hotspot temperature. Therefore, accurate prediction of the
hotspot temperature is important for both manufacturers and utilities.

There are a few methods to measure the hotspot temperature, one of which is
using fibre optical temperature sensors positioned at the predicted hotspot of the
windings. The thermal sensors, attached to the end optical fibre, are usually placed
between the insulated conductor and spacer, and their signals via optical fibre
transmitted out of the tank. However due to the cost which may be difficult to justify
in terms of cost for every new transformer. It is not practical for retro-fitting the
existing transformers. The main difficulty with direct measurement technique is how
to accurately locate the hotspot and possible the sensors.

Another method to identify the hotspot temperature is by using transformer

thermal model or calculation method. The calculation of the internal transformer
temperature (Hotspot temperature) is a very complicated and difficult task.
However, engineers made simplifying assumptions in the generally accepted
methods for calculating the temperature of power transformers as reported in the
IEEE and IEC standards (IEEE Standard C57.19, 1995 and IEC Standard, 1991).
By using these standards, a computer simulation can be used to calculate the hotspot
temperature given the historical loading and ambient temperature profiles are
available. Then the life of transformer can be predicted from the calculated
temperature.

1.2 Problem Statements

Most of power transformers in Libya today were installed during the 1960s.
Although some of these transformers may still be operating satisfactorily, they are
approaching or past the designed lifetime. Ageing equipment is a serious
 3

contributing factor to poor system reliability and high operating costs in many
utilities. Moreover, simultaneous transformer installation will probably lead to
simultaneous failure and replacement in the future. The replacement requires a lot of
capital investment and it represents a financial burden for utilities over coming
years. Therefore it is important for utilities to know when to replace ageing
transformers so that the replacement could be scheduled in a manner to lower the
cost and give minimize impact on customers.

1.3 Project Objectives

The main objectives of this research are listed as follows:

 To carry out literature review on transformer insulation ageing and lifetime

assessment.
 To compile data for operation and specifications of transformer (load,
ambient temperature, design data).
 To create a simple Simulink/Matlab software for calculate hot-spot
temperature and the transformer loss of life. This will be done with the help
of IEC standard equations.
 To take in consideration of secondary ageing factors such as moisture and
acidity to calculate the transformer loss of life.
 To predict the lifetime of transformers base on their actual loading profiles
and ambient temperatures.

1.4 Project Scopes

These scopes of this study are:

 The transformer located in Libya.

 The ambient temperature taken in two period time ( summer and winter)
 4

 The data are collected from General Electricity Company of Libya which use
scada control program.
 IEC 60076-7:2005 loading guide standard will be used with computer
simulation to estimate the transformer end-of-life.

1.5 Outline of thesis

The remainder of this is organized as follows:

 Chapter 2: literature review
In this chapter, an intensive literature review on transformer insulation ageing has
been made in this chapter to understand the insulation ageing process. This chapter
also presents the simple thermal diagram used to describe the thermal performance
of a transformer and the hotspot temperature of winding where the severest ageing
process would occur. Different cooling modes available in are also discussed. The
last part of this chapter explain IEC60076-7 version 2005 thermal model in details.

 Chapter 3: Methodology
In this chapter the real data of transformer data, loading and ambient temperature
from the Libya electricity company and the meteorology office are used in this
study. The last IEC60076-7 thermal model is established in Simulink/Matlab by
solving the top oil differential equation, hotspot differential equation and loss of life
of the transformer using Runge-kutta numerical method.

 Chapter 4: Result and Analysis

In this chapter the hotspot temperature have been calculated as the top oil
temperature added to the hotspot temperature rise and ambient temperature. It also
calculated the loss of life of the transformer from the hotspot temperature. Besides
the temperature, secondary ageing factors such as water and acidity also have been
introduced in loss of life calculation, using proposed formulation to represent
secondary ageing factor (acidity and water ) which increases with life.
 5

 Chapter 5: Conclusion and Recommendation

This chapter gives conclusions of this thesis research and recommendations for
future study.
 5

CHAPTER 2

LITERATURE REVIEW

2.1 Transformer insulation life

A transformer cannot wear out in the same way as a rotating machine since it has
practically no moving parts, except tap changers or cooling fan and pump. The
copper windings, the electric steel core and other metal parts of the transformer can
have a very long life if an adequate protection against corrosion is applied. However
this is not the case for insulating materials such as both Kraft papers which are
mostly made from cellulose materials and insulating oils (M.T. Ishak, 2009).

Since the invention of the power transformer, the conventional conductor

insulation has been some from of paper or cloth. The main constituent of these
fibrous materials is cellulose, an organic compound whose molecule is made up of
long chain of glucose rings, typically numbering in the range from 1000-1400
(S.V.Kulkani and S.A. Khaparde, 2004). It is may be represented simply as
[C5H10O5]n . Degree of molecular polymerization (DP) is the average number of
glucose rings in the molecule. The chemical structure of a portion of a cellulose
molecule is shown in Figure 2.1.
 6

Figure 2.1: Cellulose molecule structure (W.J. Mcnutt, 1993)

A single cellulose fiber is made up of many of these long chains of glucose

ring. Since the mechanical strength of the cellulose materials depends on the length
and condition of the fibers, the degree of polymerization gives a good measure of
retained functionality as transformer insulation ages in service.

Since the cellulose materials cannot be replaced once the winding are
manufactured, the transformer life can be defined as the total time between the
initial state for which the cellulose paper is considered new to the final state for
which dielectric stress, short circuit stress or mechanical movement which could
occur in normal service would cause an electrical failure of cellulose paper. In a
brief, when paper insulation fails the transformer life has ended.

2.1.1 Ageing factors:

Insulation paper ageing is a complex and irreversible phenomenon. As cellulose

ages thermally in an operating transformer, three possible mechanisms that
contribute to the insulation degradation are: hydrolysis, oxidation and pyrolysis
(heat) .The agents responsible for the respective mechanisms are water, oxygen and
heat (temperature) (M.T.Ishak, 2009). They will be explained in more details.
 7

2.1.1.1 Hydrolysis (water)

The water in the transformer comes from two sources: ingress of water from the
atmosphere, and the degradation or ageing of cellulose and oil. Hence, the oxygen
bridge between glucose rings is affected by water, causing a rupture of the chain and
the formation of two –OH groups, each attached to its monomer. The result is
reduction of DP and shortening and weakening of the fiber (W.J. Mcnutt, 1993).

In general, the mechanical life of the non upgraded paper insulation is reduced
by half for doubling in water content and in earlier studies by Fabre and Pichon it
was found that the rate of thermal deterioration of paper is proportional to its water
content. For example, decreasing the water content of the paper from 1.0% to 0.5%
doubles the life of that paper. However, later studies found that the rate of
degradation for later stages of degradation is in the form of logarithm relationship to
moisture level. Recent studies by Lundgaard of SINTEF Energy Research have
shown if the normal life is defined as ageing under dry, oxygen free conditions, then
a water control of 1% in non upgraded insulation paper can reduce life expectancy to
30% of normal life, whereas for 1% water in the thermal upgraded paper, it will
reduce the life expectancy to 60% of the normal life. If the water constant increases
to 3-4%, the life expectancy will drop to approximately 10% and 25% of normal life
for non upgrade paper insulation and upgraded paper insulation respectively
(M.T.Ishak, 2009).

2.1.1.2 Oxidation (oxygen)

Oxygen attacks the carbon atoms in the cellulose molecule to form aldehydes and
acids, releasing water, carbon monoxide, and carbon dioxide. The bonds between
rings are weakened, leadings to lower DP. Water released by this mechanism can
also contribute to the hydrolysis effect previously mentioned (W.J. Mcnutt, 1993).

According to Fabre and Pichon, oxygen increases the rate of degradation of

paper in oil containing 0.3%-5% water by a factor of 2.5 for the free breathing
 8

transformers. A free breathing transformer is the transformer has vents above the oil
that allow air to enter and exit as the oil expands and contracts due to variations in
the operation temperature. Reducing the oxygen level from 30,000ppm in oil to less
than 300ppm will reduce the ageing factor (M.T. Ishak, 2009).

2.1.1.3 Pyrolysis (heat)

Heat in the extreme will result in charring of the fibers, but at lesser levels it
contributes to the breakdown of individual monomers in the cellulose chain. A solid
residue is formed and gases are liberated, namely water vapor, carbon monoxide,
carbon dioxide, and hydrogen. Again DP is reduced (W.J. Mcnutt, 1993).

The paper insulation deteriorates rapidly if its temperatures are more than
90oC.The winding made of copper can hold their mechanical strength up to several
hundred degrees Celsius without deterioration and the transformer oil does not
significantly degrade below 140% (M.T. Ishak, 2009).

2.2 IEC thermal ageing equation

The IEC 354 (IEC Standard, 1991) chooses to use the Montsinger rule of thermal
degradation which is a simplified expression of the Arrhenius law of thermal
degradation. The simple exponential of Montsinger is given as
(2.1)

Where L is lifetime, P is material constant and Ѳ is the absolute temperature.

Many investigators have not always agreed on the criteria for which L is
representative of lifetime. Therefore, the possible way is by changing the equation
(2.1) to the rate of ageing. This can be done by inversing of the lifetime, that is
(2.2)
 01

Where V is the rate ageing, M is a constant which is depended on many factors but
normally moisture content and oxygen of the insulation. Most important is the fact
that the coefficient of temperature variation, P, can be general assumed as a constant
over the temperature range of 80oC to 140oC. In this temperature range, the rate of
ageing doubles for every 6K temperature rise. This means that if an insulation
service life of N years applies for a temperature ѲoC, the temperature of (Ѳ +6) oC
will reduce the life by (N/2) years (M.T.Ishak, 2009).

The Montsinger relation can be used to obtain the relative rate of thermal ageing any
hotspot temperature over the reference temperature. For transformers designed in
accordance with IEC 60076-2 (IEC 60076-2 standard, 1997), the relative rate of
thermal ageing is taken to be equal to unity for a hotspot temperature of 98oC. This
corresponds to operation at an ambient temperature of 200C and hotspot temperature
rise of 78oC. The relative ageing rate for non thermal upgraded paper is given by
equation (2.3).

(2.3)

For ambient temperature other than 20oC, the hotspot temperature rise has to be
modified accordingly. For example, when the ambient temperature is 30 oC, the
allowable hotspot rise is 68oC

2.3 End of life criteria

The general definition of “end of life” for an insulation system is the point at which
the insulation no longer performs reasonably; from the viewpoint of insulation’s
purpose, it seems only right to consider this point at which the insulation system no
longer maintains the majority of its original dielectric strength. However as paper
insulation ages, the dielectric strength of the paper does not decrease significantly
even well after the paper has become brittle. Therefore, the point at which the paper
loses enough strength to withstand the mechanical forces is regarded as the practical
point of end of life for insulation system.
 00

To what extent the retention of strength to withstand mechanical forces needs

be, is an option in making the definitions of the end of life point and often it is
expressed in terms of a measureable physical property. This could be either a
mechanical property such as tensile strength, burst strength, elongation to break or
abrasion resistance; or it could be a chemical property such as degree of
polymerization. The end of life point could be in percentage deterioration or an
absolute value of these properties.

Another possible way to define an end point for transformer life is by doing
functional life tests on the actual transformers. However this method is only suitable
for small transformers not the power transformers because it is not economically
practical.

In order to properly predict insulation life, it is critical to understand the basis

from which any life reference values are derived. Over the years there have been
many studies on insulation ageing which generated significant but often conflicting
results and definitions of end of life (M.T .Ishak, 2009).

Table 2.1, in (IEEE Standard C57.19, 1995 and M.T .Ishak, 2009), lists
various end of life criteria or normal insulation life for a transformer with thermally
upgraded paper, in a well-dried and oxygen free condition at the reference
temperature of 110 0C.
 01

Table 2.1: Normal insulation life value for a well dried oxygen free system at the
reference temperature 110 0C (IEEE Standard C57.19, 1995 and M.T.Ishak, 2009)

Normal insulation life

Basis

Hours Years
50 % retained tensile strength of insulation
65,000 7.42
(former IEEE std C57.92-1981)

25% retained tensile strength of insulation 135,000 15.41

Retained degree of polymerization in

insulation, DP=200 150,000 17.12

Interpretation of distribution transformer

functional life test data (former std C57.91- 180,000 20.55
1981)

Table 2.1, shows that the IEEE (IEEE Standard C57.19, 1995) standard gives
the user the freedom to choose end of life criteria base on their requirement which is
most applicable to their needs. Therefore we can say that there are no standard
values for determining the transformer insulation life end point.

All these normal insulation life definition associate the loss of life with paper
losing its ability to withstand mechanical stresses. The main point here is that if the
unit is run for 20.55 years (for example) at 1100C hotspot temperature, the life of the
insulation will expire. If it runs hotter, the life will expire sooner, and if it runs
cooler, the life will extend.

In these theses DP equal 200 is chosen as the end of life criterion and would
give the insulation life of 150,000 hours under full load condition for IEEE loading
Guide. This method has been suggested by some researches based on the
measurement of the degree of polymerization. A new Kraft paper has DP in the ring
1000-1200 after going through the factory drying process. The DP drops to lower
value with time.
 02

On contrary, the IEC Loading Guide gives no value for expected life. However,
based on design and experience, the life of a transformer has been estimated as 30
years service life of the insulation system with 980C for hotspot temperature for
Kraft paper by assuming that the insulation system is well preserved (i.e. without
oxygen, water or contaminations) (M.T. Ishak, 2009).

2.4 Transformer thermal

2.4.1 Introduction

Transformer losses are produced by the electrical current flowing through the coils
and the magnetic field alternating in the core (G. Pop, M. Chindris and R. Bindilu,
2009). The losses are calculated using measured voltage and current. This allows the
meter to continue registering Kwh, Kvarh and Kvah (unaffected by compensation).
Transformer losses are generally classified into no load losses and load losses as
shown in Figure 2.2. Transformer losses can be determinate with the following
mathematical expression (G. Pop, M. Chindris and R. Bindilu, 2009).
(2.4)

Where:

: total losses in transformer;

: no-load losses;
PS: load losses.
 03

Figure 2.2: Transformer losses classification

The no load losses (referred to as excitation losses, core losses, or iron losses) are a
very small part of the power rating of the transformer, usually less than 1%.
However, these losses are considered constant over the lifetime of the transformer
(do not vary with load), and thus they generally represent a sizeable operating
expense, especially if energy costs are high. Therefore, accurate measurements are
essential in order to evaluate individual transformer performance accurately. They
include losses due to magnetization of the core, dielectric losses in the insulation,
and winding losses due to the flow of the exciting current and any circulating
currents in parallel conductors. No load losses are the losses in a transformer when it
is energized but not supplying load. The no load losses contain tow main
components, hysteresis loss and eddy current loss (G. Pop, M. Chindris and R.
Bindilu, 2009).

Hysteresis losses are caused by the frictional movement of magnetic domains in

the core laminations being magnetized and demagnetized by alternation of the
magnetic field. Eddy current losses are caused by varying magnetic fields inducing
eddy currents in the laminations and thus generating heat.
 04

The load losses are commonly called copper losses or short circuit losses. They
are in accordance with convention subdivided into I2R losses, Eddy current losses
and stray losses (G. Pop, M. Chindris and R. Bindilu, 2009).
I2R losses (Dc losses) occur in transformer windings and are caused by the
resistance of the conductors. The magnitude of these losses increases with the square
of the load current and are proportional to the resistance of the windings.

Eddy current losses due to magnetic fields caused by alternating current, also
occur in the windings .Stray losses are due to losses in structures other than
windings, such as clamps, tank or enclosure walls, etc.; this can be expressed as (G.
Pop, M. Chindris and R. Bindilu, 2009):
Ps= I2.R + PEC + PSL (2.5)

where:
I2.R : losses due value of the current and resistance of the transformer;
PEC : Eddy Current Losses;
PSL : Stray Losses.

All these losses cause heating in all parts of transformer and this heat must be
taken away to avoid high temperature because they will cause decrease of insulation
and the normal life of transformer (M.T. Ishak, 2009).

2.4.2 Transformer thermal diagram

The temperature distribution inside a transformer is extremely complex and difficult

to model accurately, consequently the thermal characteristics of transformers are
normally analyzed in the simple thermal diagram. A typical thermal diagram using
the IEC 354 and IEEE C57.91-1995 standards is shown in Figure 2.3, (A. Al-Nadabi
and H. al-Riyami, 2009).
 05

Figure 2.3: Transformer Thermal Diagram (A. Al-Nadabi and H. al-Riyami, 2009).

Several assumptions are made in such a thermal diagram and the assumptions are
(A. Al-Nadabi and H. al-Riyami, 2009).

 The change in the oil temperature inside and along the winding is linearly
increasing from bottom to top,
 The increase in the winding temperature from bottom to top is linear with a
constant temperature difference g,
 At the winding top the hot- spot temperature is higher than the average
temperature g rise of the winding,
 The difference in the temperature between the hot-spot and the oil at the of
the winding is defined as Hg , where H is a hot-spot factor.

For many years, the value of hot-spot factor, H was taken to be approximately
10% of the average gradient, that is, the maximum gradient was considered to be 1.1
times the average gradient.
 06

CIGRE Working Group12-09 did the thermal tests and collected the data during
the tests to quantify the hot-spot factor used in the IEC loading guide (A. A.
Elmoudi, 2006). These analysis data showed that the value of the hot-spot factor is
1.1 for distribution transformers and 1.3 for medium and large transformers. These
values for the hot-spot factor depending on the transformer size, short circuit
impedance and winding design.

The hot spot temperature is normally located at the top of winding this is due to
leakage flux concentration near to the top of the winding which increase the eddy
current losses and in consequence, the temperature. The hot spot temperature for a
transformer under any load k is equal to the sum of the ambient temperature, the top
oil temperature rise over ambient and the hot spot temperature rise over top oil. This
can be expressed by the equation below:
HS = TA +TO + HSR (2.6)
Where:
TA is the ambient temperature, (oC)
TO is the oil temperature rise over ambient, (oC)
HSR is the hot spot temperature rise over top oil temperature, (oC)
HS is the ultimate hot spot temperature, (oC)

The top oil temperature rise over ambient temperature is given by the
following equation:

(2.7)

where:

TORR is the top oil temperature rises over ambient at rated load.
R is the ratio of load losses at rated current to no load losses.
K is the load factor.
n is an empirically derived exponent that depends on the cooling method.

The hot spot temperature rise over top oil temperature is given by the
following equation:
(2.8)
where:
 07

H is the hot spot factor.

g is the average winding to average oil temperature rise at rated load.
m is an empirically derived exponent that depends on the cooling method.

The exponent n and m approximately account for changes in load loss and oil
viscosity because of the change in temperature. These exponents are different for
different cooling modes as shown in Table 2.2, (M.T. Ishak, 2009).

Table 2.2: Exponents used in temperature calculation (M.T.Ishak, 2009)

standard IEC IEEE

Type of cooling x Y=2m n m

ONAN 0.8 1.6 (0.8) 0.8 0.8

ONAF 0.9 1.6 (0.8) 0.9 0.8

OFAF 1.0 1.6 (0.8) 0.9 0.8

ODAF 1.0 1.0 (2.0) 1.0 1.0

2.5 Direct hotspot measurement

The direct temperature measurement techniques using fibre optic sensors are
currently available and are in use for some larger power transformers. Ideally, this is
the best method is to directly measure the winding hot spot temperature (A. A.
Elmoudi, 2006). In this method a sensor, made of photo-luminescent material and
attached to the end of optical fiber (S.V.Kulkani and S.A. Khaparde, 2004). These
devices are capable of indicating the temperature only at the spots where sensors are
located. The sensors are usually placed between insulated conductor and radial
spacer. CIGERE Working Group (CIGRE Working Group12-09 survey, 1995)
suggests the eight sensors would be adequate if placed in the winding location where
the highest temperatures are expected. However due to the cost which may be
difficult to justify in terms of cost for every new transformer and also it is not
practical for retro-fitting the existing transformers.
 08

2.6 Transformer cooling system

The heat produced in a transformer must be dissipated to an external cooling

medium in order to keep the temperature in a specified limit. If transformer
insulation is experienced higher temperatures than the allowed value for a long time,
it will cause rapid degradation of insulation and hence severely affect the
transformer life.

A cooling system of transformers increases the load capacity of a transformer

by improving its ability to dissipate the heat generated by electric current. In other
words, good cooling systems allow a transformer to carry more of a load than it
otherwise could without reaching critical hot spot temperatures.

In oil cooled transformers, the oil provides a medium for both cooling and
insulation. Heat from core, windings and structural components is dissipated by the
process of the oil circulation. The heat is finally transmitted either to atmospheric air
or water. The process of transferring heat from a transformer involves three different
heat transfer mechanisms which are conduction, convection and radiation
(M.T.Ishak, 2009). In the oil cooled transformers, convection plays the most
important role and conduction the least important (S.V.Kulkani and S.A. Khaparde,
2004). The conduction is the transfer of heat through a material by direct contact,
whereas the convection is the transfer of heat in a fluid (oil) as a result of the
movement of the fluid itself. The radiation is the transfer of heat via electromagnetic
waves through space. There are four common types of cooling are:

1- Oil Natural and Air Natural cooling (ONAN)

2- Oil Natural and Air Forced Cooling (ONAF)
3- Oil Forced and Air Forced Cooling (OFAF)
4- Oil Direct and Air Forced cooling (ODAF)

They have been used in the industry and will be explained in more details.
 11

2.6.1 Type of cooling transformer

2.6.1.1 Oil Natural and Air Natural Cooling (ONAN)

This type of cooling is most common used in the practice. The ONAN cooling
transformers have their windings and core cooled by oil naturally circulation
(thermosiphon effect). The oil is then cooled by air which is naturally cooled.
Moreover, Oil is kept in circulation by the gravitational buoyancy in the closed-loop
cooling system as shown in Figure 2.4.

Figure 2.4: ONAN cooling diagram (S.V.Kulkani and S.A. Khaparde, 2004)

The heat developed in active parts is passed on to the surrounding oil through
the surface transfer (convection) mechanism. The oil temperature increases and its
specific gravity drops, due to which it flows upwards and then into the coolers. The
oil heat gets dissipated along the colder surfaces of the coolers which increase its
specific gravity, and it flows downwards and enters the transformer tank from the
inlet at the bottom level (S.V.Kulkani and S.A. Khaparde, 2004).

In ONAN cooling system of transformers there is a big difference between the

temperatures of top and bottom oil because the oil velocity is slow between the
 10

transformer and radiators (M.T. Ishak, 2009). The ONAN cooling has the following
advantages. They are (S.V.Kulkani and S.A. Khaparde, 2004).

 It is more reliable as no cooler controls are involved and it requires less

maintenance.
 The cost increase due to extra radiators is, to a large extent, compensated by
the reduction in cost due to the absence of fan and control system.
 It is particularly useful when low noise transformers are required. Absence of
fans makes it easier to achieve the required low noise level.
 There is no cooler loss.
 Winding losses also reduce because of lower winding.

2.6.1.2 Oil Natural and Air Forced Cooling (ONAF)

As the transformer losses increases, the number and size of the radiators that are
required to cool the oil must increase. Eventually, a point is reached where air and
natural convection are not adequate to remove the heat and air must be forced
through the radiators by motor- driven fans (John J. Winder, Jr, 2002), This type of
cooling is termed as ONAF (Oil Natural and Air Forced) cooling, as shown in
Figure 2.5.

These fans generate forced air flow in the radiators to increase the oil circulation
rate from tank to radiators. This type of cooling improves the efficiency of the
transformer to increase the capability to operate at a high load, but this flow rate is
relatively low. Because of this, the heat carrying (or dissipating) capacity of the oil
is low. The heat carrying capacity can be defined as (S.V.Kulkani and S.A.
Khaparde, 2004).

Q = m C P ( T Out - T In ) (2.9)
 11

where Q is heat flow in W, m is mass flow rate in kg/s, Cp is specific heat in J/ (kg °
C) and temperatures Tout and Tin are in °C. For the given transformer oil inlet (Tin)
and top oil (Tout) temperatures,

Figure 2.5: ONAF cooling diagram (S.V.Kulkani and S.A. Khaparde, 2004)

2.6.1.3 Oil Forced and Air Forced Cooling (OFAF)

Although the cooling capacity is greatly increased by the use of forced air,
increasing the loading to take advantage of the increased capacity will increase the
temperature gradients within the transformers. A point is reached where the internal
temperature gradient limit the ability to increase load any further. The solution is to
increase the oil velocity by pumping oil as well as forcing air through the radiators.
This type of cooling called OFAF (Oil Forced and Air Forced) cooling. The usual
pump placement is at the bottom of the radiators as shown in Figure 2.6, forcing oil
from the radiator outlets into the bottom of the transformer tank in the Sam direction
as natural circulation but at a much higher velocity (John J. Winder, Jr, 2002). By
 12

direction the flow of oil within the transformer windings, reduce considerably the
temperature between the top and the bottom of the radiators (M.T. Ishak, 2009).

Figure 2.6: OFAF cooling diagram (S.V.Kulkani and S.A. Khaparde, 2004)

2.6.1.4 Oil Direct and Air Forced Cooling (ODAF)

In OFAF cooling type the oil forced from the pump into the transformer, its flow is
governed by the least resistance path as well as the buoyancy. Hence, part of the oil
may not enter either windings or core, and may from parallel path outside these two.
Thus, the top oil temperature may reduce because of the mixture of hot oil coming
from the windings and the cool oil coming from the pump. This in turn reduces the
effectiveness of radiators. The solution is to improve the heat dissipation rate by the
oil forced and directed in the windings through the paths. This type of cooling called
 13

ODAF (Oil Direct and Air Forced) cooling, as shown in Figure 2.7, (S.V.Kulkani
and S.A. Khaparde, 2004).

Figure 2.7: ODAF cooling diagram (S.V.Kulkani and S.A. Khaparde, 2004)

ODAF type of cooling is used in most of the large rating transformer .one
disadvantages of ODAF cooling is the increased pressure loss because of the ducting
system used for directing the oil flow (S.V.Kulkani and S.A. Khaparde, 2004).

2.7 IEC 60076- 7 thermal models

Most of transformers around the world including Libya are designed according to
the IEC standard. Therefore, in order to investigate the temperature related loss of
life for transformers designed according to IEC standard, it is appropriates to use the
IEC thermal model. The latest IEC 60076-7 thermal model (IEC 60076-2, 2005),
 36

RFFRERENCES

Ahmed Al-Nadabi and Hisham Al-Riyami (2009). “Design A Simulation Model to

Estimate the Hot Spot Temperature in Power Transformer”, DJERBA-TUNISIA

A. M. Emsley and G. C. Stevens (1994). “Review of chemical indicators of

degradation of cellulosic electrical paper insulation in oil-filled transformers,”
Proc. Inst. Elect. Eng., Sci. Meas. Technol., vol. 141, no. 5, pp. 324–334.

Asaad A. Elmoud (2006). “Evaluation of Power System Harmonic Effects On

Transformers”.

CIGRE Working Group 12-09 (1995)." A Survey of Fact and Opinions On The
Maximum Safe Operating Temperature of Power Transformers Under
Emergency Conditions ", ELECTRA, NO.129, P.23-63.

Damian Giaouris (2008). “Matlab/Simulink Tutorial” School of Electrical, Electronic

and computer engineering, Newcastle University. Release 14 - version 7.0
seventh Edition.

Danilo Šćepanović (2010). “Introduction to Programming in MATLAB” Lecture 5:

Symbolics, Simulink, file I/O Building GUIs.

Dejan Susa, Member IEEE, and Hasse Nordman (2009). “A Simple Model for
Calculation Transformer Hot-Spot Temperature”, IEEE Transactions On
Power Delivery, Vol. 24, No. 3.

D. Susa, M.Lehtonen (2006). “Dynamic thermal Modeling of Power Transformers:

Future Development- Part I”, IEEE Transaction on power Delivery, Vol.21,
No.4, p. 1961-1970.
 36

D. Susa, M.Lehtonen, and H.Nordman (2005). “Dynamic Thermal Modelling Of

Power Transformers” IEEE Transaction on Power Delivery, Vol.20, No.1,
p. 18-20.

Gabriel Pop, Mircea Chindris, Radu BindiIu (2009). “Determination of Power Losses
In Transformers Working In Unbalanced and Harmonic Polluted Networks”.

General Electricity Company Of Libya (2007).” Design Standard Of

GECOL”.Version 1.0.

H.Yoshida, Y.Ishioka, and T. Suzuki (1987). “Degradation of Insulation Materials of

Transformers”, IEEE Transaction on Electrical Insulation, Vol.El-22,
No.6, p.795-800.

http://www.jcmiras.net/wiki/transformer-testing-temperature-test-or-heat-run-33.htm

IEC (1991).” Guide for loading for Oil Immersed Power Transformers”
International Elctrotechnical Commission Standard, Geneve.

IEC 354 (1991).”Guide to Loading of Oil-Immersed Power Transformers”,

IEC 60076-2 (1997). “Power Transformers- Part 2 : Temperature raise “.

IEC 60076-2 (2005). “Power Transformers- Part 7 :Loading Guide for oil-Immersed
Power Transformers”.

IEEE Standard C57.19 (1995).” Guide for loading Mineral Oil Immersed
Transformers”.

James H. Harlow.”Electric Power Transformer Engineering”

John J. Winder, Jr (2005).” Power Transformers Principles and Applications”

.New York, Marcel Dekker.
 36

L.E Lundgaard, W. Hansen, T.J. Painter and D. Linhjell (2004). “Aging of Oil-
Impregnated Paper in Power Transformers”, IEEE Transaction on Power
Delivery. Vol.19.

L.E .Lundgaared, W. Hansen, S. Ingebrigtsen, D. Linhjell, and M. Dahlund (2005).

“Aging of Kraft paper by Acid Catalyzed Hydrolysis” IEEE International
Conference on Dielectric Liquids. Coimbra.

Mohad Taufiq Ishak (2009)."Simulation Study on Influencing Parameters Of

Thermal Ageing For Transformer Lifetime Prediction", University of
Manchester.

Mohammad Tolou Askari, Mohad Zainal Abidin Ab. Kader, Wan

Fatinhamamah Wan Ahmad and Mehdi Izadi (2009). “Investigate the Effect of
Ambient Temperature on HST of Transformer” Proceedings of 2009 IEEE
Student Conference on Research and Development (SCORed 2009), 16-18
Nov’09, UPM Serdang, Malaysia.

Steven T. Karris (2006). “Introduction to Simulink with Engineering Applications”

Orchard Publications.

S.V. Kulkani and S.A Khaparde (2004). Transformer Engineering: Design and
Practice. New York, Marcel Dekker.

V.A. Pantic, D.V. Pantic, and B. Mirosavljevic (2002). “Extension of the Lifetime and
Increase of The transformer Operation Safety on the Grid” , CIGRE Session
2002, 12-113.

V.G Davydov and O. Roizman (2002). “Moisture Assessment in Power Transformers:

Lessons Learned”, in VAISALA News.

W.J. MCNUTT (1993)." Insulation Thermal Life Consideration for Transformer

Loading Guides".
 33

Y. Du, M. Zahn, B.C Lesiuetre, A. V. Mamishev, and S.R Lindgren (1999).

“Moisture Equilibrium in Transformer paper-oil system,” IEEE Electrical
Insulation Magazine, Vol.15, No.1 ,p. 11-20.