M.tech Thesis 22122015

An Integrated Approach to Remaining Life Prediction of
Rotating Machines
by
Tarun Chugh
[Enrolment No. ENGG01201301084]
Bhabha Atomic Research Centre
A thesis submitted to the
Board of Studies in Engineering Sciences
In partial fulfilment requirements
for the Degree of
MASTER OF TECHNOLOGY
of
HOMI BHABHA NATIONAL INSTITUTE
December, 2015
AA
i
Homi Bhaba National institute
Recommendations of the Thesis Examining Committee
As members of the Thesis Examining Committee, we recommend that the thesis prepared by
Mr. Tarun Chugh entitled ‘An Integrated Approach to Remaining Life Prediction of
Rotating Machines’ may be accepted as fulfilling the thesis requirement for the Degree of
Master of Technology.
Designation Name Signature
Member-1
Member-2
Member-3
Expert-1
Expert-2
Technical Advisor
Guide/Convenor
Chairman
Final approval and acceptance of this dissertation is contingent upon the candidate’s
submission of the final copies of the thesis to HBNI.
I hereby certify that no modifications was suggested by the M.Tech Committee

/modifications suggested by the M. Tech Committee are incorporated in the final thesis and
recommend that it may be accepted as fulfilling the thesis requirement.
Place: Mumbai
Date: Prof. P.V.Varde

OS & Head, RRSD
BARC, Mumbai-85
ii
DECLARATION
I, hereby declare that the investigation presented in the thesis has been carried out by me. The
work is original and has not been submitted earlier as a whole or in part for a degree /
diploma at this or any other Institution / University.
Tarun Chugh
[Enrolment No: ENGG01201301084]
iii
DEDICATIONS
To
My Parents & Brother
iv
ACKNOWLEDGEMENTS
I would like to express my gratitude to Prof. P.V.Varde, OS & Head, RRSD, BARC.
Throughout the whole project, his patient guidance, constant encouragement and meticulous
attention to detail provide me with tremendous motivation.
I am also indebted to Sh. Alex Mathew, MS (Electrical), Dhruva, BARC for his precious time
and invaluable input into my research.
It would not have been possible to come to a fruitful and mutually beneficial conclusion
without the wise guidance and suggestions from Sh. V. Sankaranarayanan (SO/H+, NPCIL),
Sh. Ashish Khandelwal (SO/G, NPCIL).
I also wish to thank Sh. Nishant Mishra (SO/D, BARC), Smt. Rejani Jayaprakashan (SA/F,
BARC), Sh. Sanju Verma, Sh. Vyas and Sh. Ajay for giving their precious time in order to
make this project a success.
TARUN CHUGH
[Enrolment No: ENGG01201301084]
v
ABSTRACT
Induction motors are workhorses of the industry. Motor life depends on electrical, thermal,
mechanical and environmental stresses. This objective of this study is to examine and analyze the
stressors that cause degradation and ageing in induction motors. Reliability studies like FMEA, FTA,
etc. are conducted to study the various failure modes of an induction motor and it is seen that one of
the failure modes i.e. Winding Failure, accounts for 25-40 % failures of motors.
This study defines loss of motor life as the loss of stator winding insulation life due to thermal and
environmental stresses .From the thermal point of view, the stator winding insulation is the weakest
part of a squirrel cage induction motor, and equations are developed to estimate the insulation life
and, hence, motor life. In this regard, an integrated model consisting of an electrical model, thermal
model and insulation ageing model is developed to evaluate the effect of various anomalies/stressors
on the life of motor. This model is used to quantify the loss of life of a 2.3 kW, 415 V induction
motor. The electrical model is developed using Open Circuit and Blocked Rotor Test on the motor.
Using the model, the stator winding losses are calculated and given as an input to the thermal model
to find the temperature rise. This work presents a simple technique for calculating thermal parameters
based on motor testing rather than from motor design data. The thermal ageing model is also
developed based on Eyring Equation considering temperature and humidity as stressors. This work
examines the loss of life when the motor is supplied by over-voltages and unbalanced voltages. A
Fuzzy Logic Approach to determine the remaining useful life (RUL) using the stator current is also
discussed. Further, a reliability model of the induction motor is developed using the data of Coolant
Pump Motors of a typical Research Reactor taken over a period of 30 years and Mean Time to Failure
(MTTF) of the motor is found. This model is then used to obtain the Weibull Plot for insulation
failure. Also, the activation energy for humidity is calculated using the above data which is
subsequently used in the Insulation Ageing Model (Eyring Model) to estimate the remaining life of
motor.
vi
Government of India
Department of Atomic Energy
BARC Training School
MUMBAI
Certificate
This is to certify that the thesis titled, ‘An Integrated Approach to Remaining Life
Prediction of Rotating Machines ’ is a bonafide record of the project work carried out by
Mr. Tarun Chugh of NPCIL Mumbai, OCES 2013 during the period 2014 -2015 for his
partial fulfilment of M.Tech programme.
Guide:
Prof. P.V. Varde

OS & Head, RRSD
BARC, Mumbai-85
vii
CONTENTS
ABSTRACT ........................................................................................................................ vi
LIST OF TABLES ................................................................................................................ x
LIST OF FIGURES ............................................................................................................ xii
NOMENCLATURE ........................................................................................................... xiv
CHAPTER 1 ......................................................................................................................... 1
INTRODUCTION ........................................................................................................... 1
1.1 GENERAL............................................................................................................... 1
1.2 MAINTENANCE OF INDUCTION MOTORS ....................................................... 3
1.3 RESIDUAL USEFUL LIFE ..................................................................................... 5
1.4 SCOPE OF WORK.......................................................................................................6
CHAPTER 2 ......................................................................................................................... 7
LITERATURE REVIEW................................................................................................ 7
CHAPTER 3 ....................................................................................................................... 12
INDUCTION MOTOR RELIABILITY STUDIES...................................................... 12
3.1 CONSTRUCTION OF INDUCTION MOTOR ..................................................... 12
3.2 INSULATION SYSTEM OF INDUCTION MOTOR...............................................14
3.3 STRESSORS, FAILURE MECHANISMS, AND CAUSES .................................. 18

3.3 FAILURE DATA ANALYSIS ............................................................................... 25
3.4 FAILURE MODES AND EFFECTS ANALYSIS ................................................ 26
3.5 FAULT TREE ANALYSIS OF INSULATION BREAKDOWN............................ 33
3.5 INFERENCE FROM PRELIMINARY STUDIES.................................................. 34
CHAPTER 4 ....................................................................................................................... 35
INTEGRATED APPROACH TO REMAINING LIFE ESTIMATION..................... 35
4.1 INTRODUCTION.................................................................................................. 35
4.2 ELECTRICAL MODEL OF AN INDUCTION MOTOR. ...................................... 36
4.3 THERMAL MODEL OF AN INDUCTION MOTOR ............................................ 44
4.4 INSULATION AGEING MODEL ........................................................................ 46
viii
CHAPTER 5 ....................................................................................................................... 57
EXPERIMENTAL STUDY........................................................................................... 57
5.1 GENERAL DETAILS...............................................................................................57
5.2 DETERMINATION OF EQUIVALENT CIRCUIT OF MOTOR...........................60
5.3 TEMPERATURE MEASUREMENT.......................................................................64
5.4 EFFECT OF OVER-VOLTAGE...............................................................................65
5.5 EFFECT OF VOLTAGE UNBALANCE.................................................................69
CHAPTER 6 ....................................................................................................................... 76
CASE STUDY................................................................................................................ 76
6.1 GENERAL..................................................................................................................76
6.2 COOLAT PUMP MOTOR.........................................................................................77
6.3 DATA ANALYSIS.....................................................................................................80
6.4 MTTF FOR WINDING FAILURE.............................................................................83
6.5 WEIBULL DISTRIBUTION FOR WINDING FAILURE........................................85
6.6 CALCULATION OF ACTIVATION ENERGY FOR HUMIDITY..........................88
6.7 REMAINING LIFE OF INSULATION SYSTEM.....................................................89
CHAPTER 7 ....................................................................................................................... 91
CONCLUSION .............................................................................................................. 91
FUTURE SCOPE OF WORK...........................................................................................94
REFERENCES .............................................................................................................. 95
ANNEXURE A: MAJOR MAINTENANCE ACTIVITIES/EVENTS
IN THE LIFETIME OF COOLANT PUMP MOTORS....................97
ix
LIST OF TABLES
Table 3.1: NEMA Insulation Class of Motor (MG-1)........................................................ 15
Table 3.2.: Origin of electric motor stresses ………………………………………………21
Table 3.3: Percentage of failure by component Failed Component ……………………… 25
Table 3.4: FMEA of an Induction Motor ………………………………………………… 28
Table 4.1 Equivalent circuit parameters of induction motor ...............................................36
Table 4.2 No-load Equivalent circuit parameters of induction motor .................................37
Table 4.3 Blocked rotor equivalent circuit parameters of induction motor ........................39
Table 4.4 Parameters of induction motor ............................................................................40
Table 5.1: Motor Name Plate Data ......................................................................................57
Table 5.2: Auto-Transformer Name Plate Data ..................................................................57
Table 5.3 : No-load Test Results ........................................................................................60
Table 5.4 : Blocked-Rotor Test Results ..............................................................................62
Table 5.5: Equivalent circuit parameters of induction motor ..............................................63
Table 5.6: Parameters of induction motor ...........................................................................63
Table 5.7: Induction Motor- Temperature measured at various intervals ...........................64
Table 5.8: Thermal Model parameters of induction motor ................................................64
Table 5.9: Thermal Model parameters of induction motor-2 ..............................................65
Table 5.10 : Input Data for over-voltage test ......................................................................65
Table 5.11: Observed temperatures for various voltages ....................................................66
Table 5.12: Estimated life of life for i) V=100% ii) V=107% ii) V=112% .......................68
Table 5.13: Sequence Voltages for different unbalance conditions ....................................71
Table 5.14: Sequence Voltages and Currents for different unbalance conditions ..............71
Table 5.15 : Input Data for unbalance test ..........................................................................72
Table 5.16: Observed temperatures for various % voltage unbalance ................................72
x
Table 5.17 Loss of life v/s No. of operating hours for Voltage Unbalance i) 0% ii) 0.6%
iii) 1.77% iv) 2.78 % .........................................................................................................74
Table 6.1: Coolant Pump Motor Data ................................................................................77
Table 6.2: Coolant Pump Motor operating duration data ..................................................84
Table 6.3: Input Data for Eb calculation ...........................................................................88
Table 6.4: Input data for remaining life calculation of Coolant Pump Motors .................90
Table A.1: Maintenance data for Motor-0 ........................................................................97
Table A.2: Maintenance data for Motor-1 ......................................................................100
xi
LIST OF FIGURES
Figure 1.1: Types of Maintenance .............................................................................................4
Figure 3.1: Block Diagram of a Motor ....................................................................................12
Figure 3.2: Diagram of a stator of an Induction Motor ..........................................................13
Figure 3.3: Factors causing insulation failure in a Motor ……………………………….......24
Figure 3.4: Failure Pareto of a motor ………………………………………………………..26
Figure 3.5 : FTA of Induction motor winding failure ……………………………………….33
Figure.4.1 : Motor Life Prediction Flowchart .........................................................................35
Figure 4.2 Equivalent Circuit (Per Phase) of induction motor ................................................36
Figure 4.3: Equivalent Circuit at No-load ...............................................................................37
Figure 4.4: Equivalent Circuit at Blocked-Rotor ....................................................................39
Figure 4.5: Positive Sequence Equivalent Circuit ..................................................................43
Figure.4.6: Heat Transfer model with single thermal time constant ......................................44
Figure.4.7: Variation of stator winding temperature with time with ambient temperature 30 C
.................................................................................................................................................46
Figure.4.8 : Fuzzy Logic Block Diagram ...............................................................................52
Figure.4.9 : Fuzzy logic –Linguistic variables .......................................................................54
Figure.4.10 : Membership Function of Ia,Ib and Ic ...............................................................54
Figure.4.11 : Membership Function for RUL ........................................................................55
Figure.4.12 : Relationship of Ia with RUL ............................................................................56
Figure 5.1: Experimental Setup- Motor with Eddy Current Dynamometer (Load) ...............59
Figure 5.2 Separation of losses……………………………………………………………...61
Figure 5.3: Induction motor equivalent circuit ......................................................................62
Figure 5.4: Measured Temperature v/s Time for 2.2kW, 415 V Induction Motor running at
full load .................................................................................................................................64
xii
Figure 5.5 Test cycle for Over-voltage Test…………………………………………………66
Figure 5.6: Measured Temperature v/s Voltage (Percentage) for 2.2kW, 415 V Induction
Motor running at full load ..................................................................................................... 67
Figure 5.7: Loss of life v/s No. of operating hours for i) V=100% ii) V=107% ii) V=112%
................................................................................................................................................68
Figure 5.8: Schematic of voltage-unbalance test ..................................................................70
Figure 5.9 Test cycle for voltage unbalance test…………………………………………...70
Figure 5.10: Positive Sequence Equivalent Circuit .............................................................71
Figure 5.11: Measured Temperature v/s Voltage Unbalance (Percentage) for 2.2kW, 415 V
Induction Motor ..................................................................................................................72
Figure 5.11: Measured Temperature v/s Negative sequence current for 2.2kW, 415 V
Induction Motor ..................................................................................................................73
Figure 5.13: Loss of life v/s No. of operating hours for Voltage Unbalance i) 0% ii) 0.6%
iii) 1.77% iv) 2.78 % .........................................................................................................74
Figure 6.1: RTD position in stator winding of Coolant Pump Motor .................................78
Figure 6.2 : Coolant Pump Motor -1 Average Winding Temperature= 94.98 0C ...............78
Figure 6.3 : Coolant Pump Motor -2 Average Winding Temperature= 93.27 0C................79
Figure 6.4 : Coolant Pump Motor -3 Average Winding Temperature= 85.26 0C ...............80
Figure 6.5: Causes of Coolant Pump Motor mal-operation .................................................82
Figure 6.6: Chi-square distribution ......................................................................................84
Figure 6.7 : Cumulative Distribution Function for Weibull Distribution for
= 60.79 and = 1 ...................................................................................87
Figure 6.8: Probability Density Function for Weibull distribution = 60.79
and = 1 ............................................................................................................................87
xiii
NOMENCLATURE
V Voltage (Phase to ground) [V]
I1 Stator Current [A]
I0 Magnetising current [A]
I1’. Total current [A]
R0 No Load resistance [Ohms]
X0 No Load leakage reactance [Ohms]
R1 Stator resistance (Per phase) [Ohms]
X1 Stator reactance (Per phase) [Ohms]
R2’ Rotor resistance (referred to stator) (Per phase) [Ohms]
X2’ Rotor reactance (referred to stator) (Per phase) [Ohms]
RL Load resistance [Ohms]
s Slip
Voltage (No-Load) [V]
Current (No-Load) [A]
. . Power Factor (No-Load)
Power (No-Load) [W]
_ Stator Winding Resistance [Ohms]
xiv
0 Equivalent Impedance (No-Load) [Ohms]
0 No Load resistance [Ohms]
0 No Load leakage reactance [Ohms]
Voltage (Blocked Rotor) [Volts]
Current (Blocked Rotor) [A]
. . Power Factor (Blocked Rotor)
Power (Blocked Rotor) [W]
01 Equivalent Impedance [Ohms]
01 Equivalent reactance [Ohms]
01 Equivalent resistance [Ohms]
Rotor Speed [RPM]
Synchronous Speed [RPM]
Shunt Impedance [Ohms]
Effective Impedance [Ohms]
θ Ambient Temperature [0C ]
θ Temperature [0C ]
τth Thermal time constant
Cth Thermal capacitance [Farad]
Rth Thermal resistance [Ohms]
xv
Imax Maximum permissible current through the stator winding [A]
θ Maximum permissible temperature through the stator winding [0C ]
Lx Percent lifetime at temperature Tx [0C ]
L100 percent lifetime at rated temperature Tc [0C ]
Tc Hot-spot temperature for insulation class [0C ]
Tx Total allowable temperature for insulation class [0C ]
HIC Halving interval
Ea Activation energy for temperature [eV]
Eb Activation energy for humidity [eV]
K Boltzman Constant [eV/K]
tr Time at temperature Tr [Hours]
ti Time at temperature Ti [Hours]
Relative Humidity at temperature Tr
Relative Humidity at temperature Ti
Ia, Ib, Ic Phase currents [A]
µia(Iaj) Membership Function
Torque [N.m]
W Load Value Set [N.m]
K Dynamometer Constant
xvi
Failure Rate [1/years]
Chi-square
′ Confidence level
Degrees of freedom
Mean Time to Failure [years]
( ) Weibull Cumulative Distribution Function
Characteristic life [years]
β Weibull Slope
xvii
CHAPTER 1
INTRODUCTION
1.1 GENERAL
Rotating machines like induction motors are complex electro-mechanical devices utilized in
most industrial applications for the conversion of power from electrical to mechanical form.
Induction motors are used worldwide as the workhorse in industrial applications. Such
motors are robust machines used not only for general purposes, but also in hazardous
locations and severe environments. General purpose applications of induction motors include
pumps, conveyors, machine tools, centrifugal machines, presses, elevators, and packaging
equipment. Additionally, induction motors are highly reliable, require low maintenance, and
have relatively high efficiency. Moreover, the wide range of power of induction motors,
which is from hundreds of watts to megawatts, satisfies the production needs of most
industrial processes.
Depending upon their application, electric motors and other electrical equipment used in
nuclear power plants are qualified to the requirements of various regulations and standards. In
a typical nuclear power plant, these include the environmental qualification requirements, the
Class IE electrical equipment qualification requirements governed by IEEE standards (e.g.,
IEEE Standards. 323-1974 and 334-1974) etc. These qualification requirements are intended
to ensure that the electrical equipment that is relied upon to maintain the integrity of the
reactor coolant pressure boundary, to shut down the reactor, to keep the reactor safely
shutdown, to mitigate the consequences of accidents, and to monitor certain post-accident
conditions, will remain functional during and following design basis events, at any time over
the life of the plant. The qualification requirements consider the extremes of the
environmental conditions that electrical equipment will encounter during and following
1
design basis events, including seismic, temperature, pressure, humidity, chemical sprays,
radiation, vibration, submergence, and synergistic effects.[2]
Although induction motors are constructed, tested, and qualified to rigorous standards,
failures of electric motors in nuclear power plants continue to occur. Operating anomalies,
failures of other equipment, and other unforeseen circumstances can all contribute to aging
degradation in motors. Recent studies regarding the operating experience of electric motors
and the effects of aging on electrical equipment in nuclear power plants have indicated that
many electric motor failures can be attributed to the aging and degradation of insulating
materials and bearings caused by high temperature, vibration, moisture and other stressors.
Electric motors operating both inside and outside of the reactor building at a nuclear power
plant are exposed to special environmental conditions of radiation, elevated temperatures, and
high humidity. During design basis events, these conditions can reach the extremes postulated
during environmental qualification type testing. Electric motors may also then be exposed to
high vibration, containment spray, moisture impingement and/or submersion, high pressure,
and other environmental stresses.
Induction motors in balance of plant (BOP) locations may also face unique operating
environments. Condensate pump and feed-water pump motors, for example, may be exposed
to high temperature and humidity, water or chemical spray impingement, submergence, and
vibration during operation. These environmental factors may reach extreme levels during
operating transient conditions. Service water pumps and circulating water pumps are located
in intake structures adjacent to rivers or the ocean. They can be exposed to humid and salt-
laden atmosphere throughout their service life, and the possibility of submergence exists in
these locations.
2
Healthiness of the machines contributes to the production, down time reduction, reliability
and revenues. Monitoring of the healthiness of the machines, therefore, is very important and
essential. The ability to accurately predict changes in properties/parameters of electrical
machines is of critical importance in optimizing the maintenance schedule of the plant. Thus,
there is a continuous need to device test methods or to find more searching and sensitive
parameters to predict the machine health. In view of this, it becomes quite important for a
maintenance engineer to be able to predict the health of induction motor leading to
appropriate usage of the machine(s), reduction in downtime, enhanced operational reliability
& safety and revenues. Thus, the maintenance action can be optimized by diagnostics and
prognostics methods which form a part of Condition Based Maintenance (CBM).
1.2 MAINTENANCE OF INDUCTION MOTORS
Condition-based maintenance entails continuous monitoring of system data to provide an
accurate assessment of the health, or status, of a component/system and performing
maintenance based on its observed health. It involves using real-time system monitoring and
data processing. Another capability that may form part of a CBM system is an ability to
provide an estimate of the remaining useful life (RUL) of the system or component being
monitored. This type of functionality is known as prognostics, as opposed to diagnostics
which is used to assess the current condition of a monitored system.
A condition-based maintenance approach promises a range of improvements over existing
approaches, with a potential reduction in overall maintenance costs being one of the primary
drivers for developing such approaches. The cost associated with each of the various
maintenance approaches is depicted in Figure 1.1. A corrective maintenance approach has a
relatively low maintenance cost (minimal preventative actions), but high performance costs
associated with the high cost of operational failures. In contrast, preventative maintenance
generally has a low operating cost, associated with reduced instances of in-service failures,
3
but often uses very conservative estimates regarding the probability of component failures
and so has a high maintenance cost, associated with the removal of components before they
have reached the end of their useful lives. It would seem, therefore, that the most cost
efficient approach is to undertake maintenance when there is objective evidence of need, i.e.
condition-based maintenance
Figure 1.1: Types of Maintenance
The development of CBM approaches has been enabled by developments and advancements
in sensor technologies, data collection, storage and processing capabilities, and continuous
improvements in algorithms and data analysis techniques. CBM systems are founded upon
the ability to infer equipment condition using data collected from monitored systems. Ideally,
a complete CBM system will incorporate both diagnostic and prognostic capabilities. The
distinguishing factor between diagnostic and prognostic capabilities is the nature of the
analysis. Diagnostics involves posterior event analysis (i.e. identifying the occurrence of an
event which has already happened), while prognostics is concerned with prior event analysis
(i.e. predicting the future behavior of a system under observation)
4
1.3 RESIDUAL USEFUL LIFE
Definition as per IEEE Guide 1205 Residual Life [16]: The remaining period of time during
which a system, structure, or component is expected to perform its essential function under
specified service conditions.
For an induction motor, the end-of- life signifies that there is a need for replacement of
induction motor and is beyond any kind of refurbishment. It signifies the possibility of a
serious problem like change is the alignment beyond the tolerances, damage to stator core,
breaking of rotor bars, etc.
Remaining life is a sub-set of end of life. It usually signifies the remaining life that the
equipment possess such that the operator can plan the maintenance activities accordingly. For
example , for motor insulation, remaining life signifies the time after which the motor may
require some maintenance like rewinding.
The remaining useful life (RUL) of an asset or system is defined as the length from the
current time to the end of the useful life. The concept of the RUL has been widely used in
operational research, reliability and statistics literature with important applications in other
fields such as material science, biostatistics and econometrics.
RUL estimation is one of the key factors in condition based maintenance (CBM) and
prognostics and health management.
It is critically important to assess the RUL of an asset while in use since it has impacts on the
planning of maintenance activities, spare parts provision, operational performance, and the
profitability of the owner of an asset RUL estimation has also an important role in the
management of product reuse and recycle which has strategic impacts on energy
consumption, raw material use, pollution and landfill The reused products must have
sufficient long lives left among others to be able to be reused.
5
A new concept of Ageing Management has recently come up and it helps the operator to take
a conscious decision pertaining to replacement/refurbishment of motor.
1.4 SCOPE OF WORK
This study defines loss of motor life as the loss of stator winding insulation life due to
thermal and environmental stresses. From the thermal point of view, the stator winding
insulation is the weakest part of squirrel cage induction motor, and equations are developed
to estimate the insulation life and, hence, motor life. Since the rotor is of squirrel cage type,
there is no rotor insulation, the rotor structure is more robust and hence, motor life is focused
on stator winding insulation. [9]
This work examines the loss of life when the motor is supplied by over-voltages and
unbalanced voltages. In order to estimate motor life, the electrical and the thermal models are
developed. This work presents a simple technique for calculating thermal parameters based
on motor testing rather than from motor design data. The thermal ageing model (based on
Eyring Equation) is also developed.
Additionally, a reliability model of a motor is developed using the data of Coolant Pump
Motors of a typical Research Reactor taken over a period of 30 years and Mean Time to
Failure (MTTF) of the motor is found. This model is then used to obtain the Weibull Plot for
insulation failure. Also, the activation energy for humidity is calculated using the above data
which is subsequently used in the Insulation Ageing Model (Eyring Model) to estimate the
remaining life of motor.
6
CHAPTER 2
LITERATURE REVIEW
2.1 Xiao-Sheng Si, Wenbin Wang, Chang-Hua Hu, Dong-Hua Zhou paper on Remaining
useful life estimation – A review on the statistical data driven approaches reviews the recent
modeling developments for estimating the Remaining Life. The review is centred on
statistical data driven approaches which rely only on available past observed data and
statistical models. The approaches are classified into two broad types of models, that is,
models that rely on directly observed state information of the asset, and those do not. [1]
2.2 M. Vfflaran and M. Subudhi paper on Aging Assessment of Large Electric Motors in
Nuclear Power Plants examined the stressors that cause degradation and aging in large
electric motors operating in various plant locations and environments. This is supplemented
by a review of motor designs, and their nuclear and balance of plant applications, in order to
characterize the failure mechanisms that cause degradation, aging, and failure in large electric
motors. A generic failure modes and effects analysis for large squirrel cage induction motors
is shown to identify the degradation and aging mechanisms affecting various components of
these large motors, the failure modes that result, and their effects upon the function of the
motor. The effects of large motor failures upon the systems in which they are operating, and
on the plant as a whole, are analyzed from failure reports in the databases.[2]
2.3 Electrical machines by J.B. Gupta- Induction Motor Test-No Load and Blocked Rotor
Tests- explains the way how the induction motor equivalent circuit is derived . The
magnetising circuit is derived using the No-Load Test taking into consideration the stator
winding losses. Stator and rotor impedance is derived from the results of the Blocked Rotor
7
Test, where the magnetising circuit is neglected. The equivalent circuit is further used to find
the stator winding loss [3]
2.4 Amit J Modak and Dr. H.P. Inamdar paper on Computer-aided Simulation Model of
Stator Ground-wall Insulation of Induction Motor based on Current Park’s Vector Approach
consists of design of computer based simulation model to generate the stator current data
pattern for the simulation of degree of degradation of stator ground-wall insulation based on
Current Park vector approach. The generated stator current data pattern can be further applied
to any suitable artificial neural network classifier for the artificial intelligence based fault
diagnosis process. [4]
2.5 M. Zeraoulia A. Mamoune H. Mangel M.E.H. Benbouzid paper on A Simple Fuzzy
Logic Approach for Induction Motors Stator Condition Monitoring applies fuzzy logic to
induction motors fault detection and diagnosis. The motor condition is described using
linguistic variables. Fuzzy subsets and the corresponding membership functions describe
stator current amplitudes. A knowledge base, comprising rule and data bases, is built to
support the fuzzy inference. The induction motor condition is diagnosed using a
compositional rule of fuzzy inference.[5]
2.6 Pedro Vicente Jover Rodríguez and Antero Arkkio paper on Detection of Stator Winding
Fault In Induction Motor Using Fuzzy Logic presents a reliable method for the detection of
stator winding faults (which make up 38% of induction motor failures) based on monitoring
the line/terminal current amplitudes. In this method, fuzzy logic is used to make decisions
about the stator motor condition. The fuzzy system is based on knowledge expressed in rules
and membership functions, which describe the behaviour of the stator winding. [6]
8
2.7 R.SaravanaKumar , K.Vinoth Kumar and Dr. K.K.Ray paper on Fuzzy Logic based fault
detection in induction machines using Lab view IJCSNS International Journal of Computer
Science and Network Security, VOL.9 No.9, September 2009 is concerned with modeling of
three-phase induction motor in three phase reference frame using Matlab/Simulink and
designing an intelligent system for condition monitoring of the motors.[7]
2.8 O. Uyar and M. Cunkas paper on Design of Fuzzy Logic Based Motor Protection
System 6th International Advanced Technologies Symposium (IATS’11), 16-18 May 2011,
Elazığ, Turkey presents a Simulink model of protection system against six different fault
parameter. A delay time is computed by fuzzy logic when one of the faults is detected such
as over voltage, over current, winding temperature, voltage unbalance, current unbalance and
low voltage. If the fault persists at the end of this delay, the system sends a stop signal to the
motor.[8]
2.9 Pragasen Pillay and Marubini Manyage paper on Loss of Life in Induction Machines
Operating with Unbalanced Supplies estimates motor life when a motor is supplied with a
combination of over- or under-voltages with unbalanced voltages. The motor life is predicted
by estimating the stator winding insulation life of squirrel cage motors using Arrhenius’
equations. Electrical and thermal models are used to calculate motor losses and temperatures,
respectively. The thermal model parameters are obtained from simple motor testing
techniques rather than from complex methods requiring motor design data. [9]
2.10 E.Anbarasu and M.Karthikeyan paper on Modeling of Induction Motor and Fault
Analysis reviews motor current signature analysis as the reference method for the diagnosis
of induction machines faults. [10]
9
2.11 Zhi Gao dissertation titled Sensorless Stator Winding Temperature Estimation for
Induction Machines elucidates the single time constant thermal model for an induction motor.
This model is subsequently used to estimate the temperature at the stator winding which
inturn gives the remaining life.[11]
2.12 Jose Policarpo , G. de Abreu and Alexander Eigeles Emanuel, paper on Induction Motor
Thermal Aging Caused by Voltage Distortion and Imbalance: Loss of Useful Life and Its
Estimated Cost deals with effect of voltage distortion and imbalance on the thermal ageing of
the insulation of low- voltage induction motors.[12]
2.13 Emanuel L. Brancato paper Estimation of Lifetime Expectancies of Motors discusses
methods for evaluating life expectancies of motors under various factors motor temperature
rise under varying load, percentage lifetime expectancy at specific load levels under various
temperature rise conditions, etc. The effect of frequent start/stops is also discussed here. [13]
2.14 Arfat Siddique, G.S.Yadava and Bhim Singh paper Effects of Voltage Unbalance on
Induction Motors discusses effects of different voltage magnitude unbalances with the same
voltage unbalance factor on the stator loss and rotor copper loss and total copper loss of three
classes of three-phase induction motor. The analysis shows the effects of resulting losses
during unbalanced voltage supply. The increase in losses under unbalanced voltages leads to
the necessity of derating the motor. It has also been shown that under-voltage unbalance
cases cause the more copper losses. [14]
10
2. 15 Sandesh Kumar Mishra , Pratibha Tiwari & Navneet Kumar Mishra paper titled
Comparative Analysis Of Three-Phase Induction Motor Fed From Balanced And Unbalanced
Power Supply deals with a comparative study made on three-phase induction motor under
balanced and unbalanced voltage supply. Plots of rotor speed, electromagnetic torque, stator
and rotor current for balanced and different unbalance cases are analysed and discussed. [15]
2.16 IEEE Guide 1205 -2014 IEEE Guide for Assessing, Monitoring, and Mitigating Aging
Effects on Class 1E Equipment Used in Nuclear Power Generating Stations gives an
overview of the concept of remaining life.[16]
2.17 Paul Ellerman report on Calculating Reliability using FIT and MTTF: Arrhenius
HTOL Mode establish a method for calculating the standard reliability values Failure Rate
(λ), Failures in Time (FIT) and Mean Time to Failure (MTTF) using the Arrhenius High
Temperature Operating Life (HTOL) model. Chi-Square analysis is used to find the failure
rate and MTTF using data like number of rejects and total operational durationwhile
considering generally accepted values of confidence interval. This model is subsequently
used to find the parameters of the Weibull distribution. [17]
2.18 EPRI Report on Life Expectancy of Motors in Mild Nuclear Plants Environments
elaborates a way to find of the Weibull Distribution Parameters of Winding Failure using
temperature. The empirical relations to determine the characteristic life have been statistically
derived using regression analysis on the failure data of a large number of motors. The model
also takes into account the load cycle of the motor.[18]
11
CHAPTER 3
INDUCTION MOTOR RELIABILITY STUDIES
3.1 CONSTRUCTION OF INDUCTION MOTOR
An electric motor includes the stator, rotor, frame, shaft, load coupling, bearings, motor
housing, motor mounting, cooling air fans and filters, lubricating oil system, bearing cooling,
stator and rotor cooling, terminations, component cooling lines at the machine, heaters, and
instrumentation sensors. The components, support systems and subsystems of an electric
motor are shown in the Figure 3.1 [2]
Figure 3.1: Block Diagram of a Motor
The most commonly used motor in nuclear power stations, and all of industry in general, is
the three phase induction motor. The induction motor is an synchronous machine, running at
1 % to 10% below synchronous speed, depending on design specifications and load torque.
As in all electric motors, the induction motor contains two major parts: a stator and a rotor.
12
Stator - The stator is a cylindrical-shaped, stationary component within which the rotor
rotates. It is made up of a three-phase winding around an iron core formed from laminated
steel punching. The stator core and windings are mounted in, and enclosed by, the motor
frame. The three phase stator windings, physically located in slots in the stator core
laminations, are spatially arranged and distributed to produce a rotating magnetic field when
the three phase voltages are applied. Figure 3.2 illustrates a typical stator showing the
arrangement of the three phase stator windings, winding connections, the stator core, and the
enclosing housing.
Figure 3.2: Diagram of a stator of an Induction Motor [2]
Rotor of an induction motor can be of two types squirrel cage or wound rotor. For a squirrel
cage induction motor, the major elements of all induction motor rotors are: the iron core
(formed from slotted, laminated steel punching); the rotor shaft that supports the iron core,
windings, bearing surfaces, and slip rings (if applicable); and the rotor winding, either wound
rotor type or squirrel cage type. In the squirrel cage induction motor, instead of a coil
13
winding, the electrical circuit of the rotor consists of highly conductive copper, copper alloy,
or aluminum alloy bars located in the slots of the rotor core. The bars are connected at each
end of the rotor by a heavy, annular-shaped, conductive end ring.
3. 2 INSULATION SYSTEM OF AN INDUCTION MOTOR
3.2.1 General [19]
Any rotating machine e.g. induction motor has two types of insulation; groundwall insulation
and conductor insulation. Groundwall insulation separate those components that may not be
in galvanic contact with each other. E.g. Groundwall insulation galvanically separates coil
from iron core of machine. Conductor insulation separates wires and turns of a coil.
Insulation class A
This includes organic fibrous materials on a cellulose base such as paper, pressboard, cotton,
cotton cloth and natural silk etc., impregnated with lacquers or immersed in an insulating
liquid. The impregnation or immersion ensures that the oxygen content of the air does not
affect the insulating properties or enhance the thermal ageing of the insulating material.
Typical materials in this class are varnished cloth and oil-impregnated paper.
Insulation class E
This includes wire enamels on a base of polyvinyl formal, polyurethane or epoxy resins as
well as moulding powder plastics on phenol-formaldehyde and similar binders, with cellulose
fillers, laminated plastics on paper and cotton cloth base, triacetate cellulose films, films and
fibres of polyethylene terephthalate.
14
Insulation class B
This includes inorganic materials such as mica, glass fibre and asbestos etc., impregnated or
glued together with varnishes or compositions comprising ordinary organic substances for
heat resistance such as oil-modified synthetic resins, bitumen, shellac and Bakelite
Insulation class F
This comprises composite materials on mica, glass fibre and asbestos bases, impregnated or
glued together with silicone resins or silicone elastomer. These materials must not contain
any organic fibrous materials such as paper or cloth backing, which is covered under class B
and even F insulation systems.
NEMA Insulation Class [19]

Insulation systems are rated by standard NEMA (National Electrical Manufacturers
Association) classifications according to maximum allowable operating temperatures
Allowable
Allowable
Temperature
Temperature
Rise at full
Temperature Maximum Operation Rise
load
Tolerance Temperature Allowed 1.15 service
1.0 service
Class factor motor
factor motor
1)
1)
o o o o
C F C C
A 105 221 60 70
B 130 266 80 90
F 155 311 105 115
H 180 356 125 -

Table 3.1: NEMA Insulation Class of Motor (MG-1)
T(oF) = [T(oC)](9/5) + 32 (3.1)
15
Allowable temperature rises are based upon a reference ambient temperature of 40 oC.
Operation temperature is reference temperature + allowable temperature rise + allowance for
"hot spot" winding.
In general a motor should not operate with temperatures above the maximum. Each 10 oC rise
above the rating may reduce the motor lifetime by one half. It is important to be aware that
insulation classes are directly related to motor life.
3.2.2 Practices of Insulation systems
LV Motors
Wound stator
By simple impregnation in a recommended insulating varnish, normally synthetic or epoxy,
followed by baking (curing), in a temperature-controlled oven, at a specified temperature for
a specific period.
For powerhouse insulation treatment the stator may be dipped in varnish for a minimum two
to three times, each dipping being followed by backing. Sometimes one immersion of the
entire stator and two additional immersions of the overhangs followed by backing may also
be sufficient.
Formed wound machines
For large motors, the practice is to wind the stator with formed coils. The coils are pre-
formed and cured before insertion into the stator slots. They are insulated with resin-rich
glass and mica paper tapes. The process of impregnation is therefore termed ‘resin-rich’
insulation. The completed formed wound stator is then heated to remove trapped moisture
16
and finally impregnated in varnish class F or H as required. It may be simple dipping, roll
impregnation (rotating in impregnating varnish for (1/2–1 hr) or vacuum pressure
impregnation (VPI). While VPI is the preferred method being more reliable, as noted below,
other methods also providing satisfactory results and being economical are practiced by many
manufacturers up to 1000 kW or so. The stator is then cured in an oven as described above.
The process of insulation and curing conforms to powerhouse insulation requirements. This
practice facilitates easy removal of an individual coil at site in case of a damage and
replacement with a spare coil. The user can stock spare coils for such eventualities.
MV Motors
To wind MV motors two methods are adopted:
Resin rich system:
The stator is wound with pre-formed coils on similar lines as noted above. After thorough
testing on each coil for the polarization index and dissipation factor (tan delta) and the
impulse voltage withstand test, the individual coils are completely cured and toughened
before inserting them into the slots.
Vacuum pressure impregnation (VPI)
For MV motors, the recommended practice is to have the stator vacuum pressure impregnated
(VPI) in insulating resins as a standard procedure, not only to meet the requirements of
‘powerhouse insulation’ but to also develop a more simplified insulating process, to cure and
toughen the stator windings and to meet the severities of all operating conditions a motor may
have to encounter. As described later, this is termed a resin-poor insulating process because
the insulating tapes now have low resin content as they are later to be impregnated in resin.
Performance and field data of this insulating system have revealed excellent results,
17
surpassing those of the normal impregnating process and even the process of resin-rich
formed coils. As an economy measure, the general practice of leading manufacturers is to
adopt a resin-rich formed coil system for frame sizes smaller than 630 and resin poor VPI,
simple dipping or roll impregnation for frame sizes 630 and higher. But it is always
recommended to adopt a resin-poor system for all MV motor windings, irrespective of frame
sizes. In a pressure-vacuum impregnation system, since the whole stator iron bulk and the
stator windings form a solid mass, removal of one coil and its replacement is impossible,
unlike in the previous case. But in view of the excellent properties of a post vacuum
impregnated insulating system, the chances of any part of the stator winding developing an
operational defect are remote. In all probability no such localized damage would arise over
the life span of the motor. The windings may fail on account of a failure of the protective
system to clear a fault or isolate the machine on a fault, but if the motor fails, the whole stator
is scrap and a totally new stator has to be requisitioned.
The coil ends must be rigidly supported and adequately braced with binding rings or tapes to
prevent their movement and also absorb shocks and vibrations during excessive over-loads,
starting inrush currents (Ist), and voltage surges.
3.3 STRESSORS, FAILURE MECHANISMS, AND CAUSES
The basic stressors that affect the operating life of electric motors are well known throughout
the industry. The most significant are heat, mechanical vibration, and wear.[2]
The stressors that affect large electric motors are:
1. Heat
2. Chemicals
3. Pressure
4. Steam
18
5. Radiation
6. Mechanical Cycling/Rubbing
7. Humidity/Water Spray
8. Electromagnetic Cycling
9. Vibration/Seismic
10. Foreign Object Ingestion
The stressors act independently and/or synergistically to cause failures in the major
subcomponents of large electric motors, such as the stator windings, electrical terminations,
bearings, and rotor cage. All of the stressors listed above contribute to the gradual or
catastrophic degradation of the insulation system. Mechanical and electromagnetic cycling,
ingestion of foreign objects, and vibration-related stressors act upon the mechanical integrity
of the machine. They can cause bearing and lubrication system problems, rotor breakage,
mounting/enclosure failures, and failures of the shaft/couplings
3.3.1 Sources of Motor Stressors
The sources or origins of the stressors may be grouped into four categories: 1) operational,
component level, 2) operational, system level, 3) environmental, and 4) human factors.
By identifying the nature and origin of large motor stressors, a determination may be made as
to the best approach to mitigate the effects of each stressor. The effects of some stressors can
only be counteracted by incorporating features into the original specification and design of
the motor. Others may be mitigated by system level design, good operating and maintenance
practices, and the proper surveillance, monitoring, and testing activities.
Component level stressors, such as heat originating from electrical and mechanical losses,
can never be eliminated, however their effects are predictable and can be mitigated by design,
good manufacturing, and monitoring/testing. System level operational stressors originate
from a variety of electrical, mechanical, and operational conditions, both transient and
19
steady-state. Many of the effects of these kinds of stressors can be mitigated or arrested by
good plant and system design, electric power system quality, protective relaying, and
efficient plant operation and maintenance.
Environmental factors, including both normal ambient conditions for operation, as well as
accident conditions, are significant stressors for electric motors. Most of these environmental
considerations are location and application specific. They can be well defined and,
consequently, accounted for in the specification and design of the electric motor that will be
used in a particular application and plant location. The geographic location of the plant, the
time of the year, and the operating status of the plant (full power, startup, shutdown, etc.) will
also contribute to the ambient temperature and humidity of some motor applications.
Finally, human factors are the last source of stressors on large motors. These problems can
never be fully eliminated, but their impact can be lessened through improved procedures and
training, adherence to manufacturers' recommendations, good maintenance and operating
practices, and thorough design engineering. Administrative and quality controls in
maintenance, modification, and operating activities can also help to reduce human factors
errors affecting large electric motors.
Category/Origin of Stressor Stressors

Operational 1. I2R Heating Heat, Vibration/Seismic, Mechanical
(Component 2. Friction Cycling
Level) 3. Windage Heat, Vibration/Seismic, Mechanical
4. Mechanical Imbalance Cycling
Vibration/Seismic ,Mechanical Cycling
Heat, Vibration/Seismic, Mechanical
Cycling
Operational 1. Mechanical Overload Heat, Vibration/Seismic, Mechanical
(System Level) 2. Frequent Starting Cycling
Heat, Mechanical Cycling, Electromagnetic
3. Over-voltage Cycling
4. Voltage Imbalance Heat, Electromagnetic Cycling
5. Electrical Transients Cycling, Electromagnetic Cycling
6. Trips Cycling, Electromagnetic Cycling
20
7. Support Equipment Problems Cycling, Electromagnetic Cycling
(air,water,lube oil system, I&C, circuit Heat, Vibration/Seismic, Mechanical
breakers ) Cycling, Electromagnetic Cycling, Steam,
Foreign Objects
8. Associated Equipment (driven loads,
discharge, suction, bypass Heat, Vibration/Seismic, Steam, Foreign
valves/dampers) Objects
Environmental 1. Ambient Conditions Heat, Pressure, Radiation. Humidity,

Chemicals
2. Moisture and water impingement Humidity
3. Steam Heat, Pressure, Humidity, Vibration,
Steam
4. Submersion/Immersion Pressure, Humidity, Chemicals, Vibration
5. Chemical spray Humidity, Chemicals, Vibration
6. Vibration Vibration, Mechanical Cycling
Human Factors 1. Misapplication/Undersized Heat, Vibration/Seismic, Mechanical
Cycling, Electromagnetic Cycling
2. Mis-operation (Excessive starting) Heat, Mechanical Cycling, Electromagnetic
Cycling
3. Poor Maintenance (Lube Oil/Grease) Heat, Vibration/Seismic, Chemicals,
Mechanical Cycling, Foreign Objects
4. Manufacturing Defect Heat, Vibration/Seismic, Chemicals,
Mechanical Cycling, Electromagnetic
Cycling, Foreign Objects
5. Installation Error Heat, Vibration/Seismic, Chemicals,
Mechanical Cycling, Foreign Objects
Table 3.2.: Origin of electric motor stresses
The stator winding system plays an important role in induction motors. A well designed
stator winding insulation system can prevent the electrical short. There are several
components and features in a stator winding insulation system, such as strand (subconductor)
insulation, turn insulation and ground-wall (or ground or earth) insulation. Turn insulation is
used for preventing shorts among the turns in the coil.
The main cause for stator failures can be divided into four groups: thermal, electrical,
mechanical and environmental-stress.[2]
Thermal Stress: One of the thermal stresses the insulation is subject to is the thermal aging
process. An increase in temperature accelerates the aging process and thus reduces the
lifetime of the insulation significantly. As a rule of thumb, a 10 degrees increase in
21
temperature decreases the insulation life by 50%. Under normal operating conditions the
aging process itself does not cause a failure, but makes the insulation more vulnerable to
other stresses, which then produce the actual failure. In order to ensure a longer lifetime and
reduce the influence of the aging process one can either work at low operating temperatures
or use an insulation of higher quality, i.e. use a higher insulation class.
Another thermal stress that has a negative effect on the insulation lifetime is thermal
overloading, which occurs due to voltage variations, unbalanced phase voltages, cycling,
overloading, obstructed ventilation or ambient temperature. For example, even a small
increase in the voltage unbalance has an enormous effect on the winding temperature. As a
rule of thumb, the temperature in the phase with the highest current will increase by 25% for
a voltage unbalance of 3.5% per phase.
Voltage Unbalance: The greatest effect of voltage unbalance is on three-phase induction
motors. Three phase induction motors are one of the most common loads on the network and
are found in large numbers especially in industrial environments.
Negative phase sequence in induction motors is caused due to unbalanced voltages in the
supply voltage applied on the stator terminals or unbalanced windings.
Negative phase sequence components create a rotating magnetic field in the stator which
moves in the opposite direction. This causes a decrease in the torque developed by the
motor. The motor will thus have to draw a higher current for the same mechanical load.
It should be ensured that the flow of air through the motor is not obstructed since the heat
cannot be dissipated otherwise the winding temperature will increase. If this is not possible
22
however, this should be taken into account by upgrading the insulation system or restricting
the winding temperature.
Electrical Stress: There are different reasons why electrical stresses lead to failure of the
stator insulation. These can usually be broken down into problems with the dielectric
material, the phenomena of tracking and corona and the transient voltages that a machine is
exposed to. The type of dielectric material that is used for phase-to ground, phase-to-phase
and turn-to-turn insulation as well as the voltage stresses applied to the insulating materials,
influence the lifetime of the insulation significantly. Thus, the materials for the insulation
have to be chosen adequately in order to assure flawless operation and desired design life.
Tracking and corona are phenomena that only occur at operating voltages above 600V and
5kV respectively.
The negative influence of transient voltage conditions on the winding life has been observed
in recent years. These transients, that either cause deterioration of the winding or even turn-
to-turn or turn-to-ground failures, can be caused by line-to-line, line-to-ground or multiphase
line-to-ground faults in the supply, repetitive restriking, current limiting fuses, rapid bus
transfer, opening and closing of the circuit breakers, capacitor switching (power factor
improvement), insulation failure in the power system or lightning strike. Variable frequency
drives are subject to permanent voltage transients. Especially during the starting and stopping
process high voltage transients can occur.
Mechanical Stress: The main causes for insulation failure due to mechanical stresses are coil
movement and strikes from the rotor. The force on the winding coils is proportional to the
square of the motor current and reaches its maximum value during the startup of the motor.
This force causes the coils to move and vibrate. The movement of the coils again can cause
severe damage to the coil insulation or the conductor.
23
There are different reasons that will cause the rotor to strike the stator. The most common are
bearing failures, shaft deflection and rotor-to-stator misalignment. Sometimes the contact is
only made during the start but it can also happen that there will be a contact made at full
speed of the motor. Both contacts can result in a grounded coil. There are other mechanical
stresses, which the windings are exposed to, like loose rotor balancing weights, loose rotor
fan blades, loose nuts or bolts striking the motor or foreign particles that enter the motor.
Environmental Stress: Environmental stress can also be called contamination. The presence
of foreign material can lead to reduction in the heat dissipation, premature bearing failure or
even the breakdown of the insulation system by causing shorts. If possible the motor should
be kept clean and dry internally as well as externally, to avoid the influence of moisture,
chemicals and foreign particles on the insulation condition.[20]
Figure 3.3: Factors causing insulation failure in a Motor
24
3.4 FAILURE DATA ANALYSIS
According to published surveys, induction motor failures include bearing failures, inter-turn
short circuits in stator windings, and broken rotor bars and end ring faults. Bearing failures
are responsible for approximately two-fifths of all faults. Inter-turn short circuits in stator
windings represent approximately one-third of the reported faults. Broken rotor bars and end
ring faults represent around ten percent of the induction. This table presents the surveys
conducted by the Electric Power Research Institute (EPRI), which surveyed 6312 motors, and
the survey conducted by the Motor Reliability Working Group of the IEEE-IAS, which
surveyed 1141 motors.
Table 3.2 – Percentage of failure by component Failed Component
IEEE-IAS EPRI
Bearings Related 44 41
Windings Related 26 39
Rotor Related 8 9
Others 22 14
Table 3.3: Percentage of failure by component Failed Component
Several alternatives have been used in industry to prevent severe damage to induction motors
from the above mentioned faults and to avoid unexpected production shutdowns. Schedule of
frequent maintenance is implemented to verify the integrity of the motors, as well as to verify
abnormal vibration, lubrication problems, bearings conditions, and stator windings and rotor
cage integrity. Most maintenance must be performed with the induction motor turned off,
which also implies production shutdown. Usually, large companies prefer yearly maintenance
in which the production is stopped for full maintenance procedures. Redundancy is another
25
way to prevent production shutdowns, but not induction motor failure. Employing
redundancy requires two sets of equipment, including induction motors. The first set of
equipment operates unless there is a failure, in which case the second set takes over. This
solution is not feasible in many industrial applications due to high equipment cost and
physical space limitations
It has been well established that stator winding failure such as stator turn-to-turn insulation
failure, stator ground-wall insulation failure, etc. constitute a major reason of induction motor
failure as shown in Figure 3.5
Figure 3.4: Failure Pareto of a motor
3.5 FAILURE MODES AND EFFECTS ANALYSIS .[2]
In order to understand the relationships of the various stressors to large motor
operationalperformance, a failure modes and effects analysis (FMEA) was performed. The
FMEA provides a systematic procedure for determining how each component of a device or
26
system can fail the mechanisms that cause it to fail, and how it can affect the overall
performance of the device or system. The means for detection of the identified failure
mechanisms are established along with methods for mitigating the effects of the failure
mechanisms. The criticality of individual component failures can then be determined in order
to prioritize inspection, surveillance, maintenance and mitigation activities and to allocate
maintenance resources. FMEAs can also indicate the usefulness of design improvements or
modifications.
The FMEA Analysis of a squirrel cage induction motor is shown in Table 3.4
27
COMPON COMPONENT FAILURE FAILURE MECHANISMS FAILURE EFFECTS MITIGATING ACTIVITIES
ENT FUNCTION MODE
NAME
Stator  Produces a  Winding to  Thermal degradation of insulation due to high  Electrical Trip  Good maintenance practices to
Winding sinusoidally ground fault ambient temperature, restricted ventilation, over-  Damage to motor keep motor clean and
distributed, voltages, low frequency, mechanical overload, winding requiring ventilation clear and
rotating magnetic voltage imbalance, single phasing, too frequent rewind unrestricted
field in the stator starting, high process fluid temperature, dust or dirt  Good operating practices to
when 3-phase ac accumulation. reduce number of starts.
voltage is applied  Mechanical degradation of insulation due to  Monitor and trend vibration.
to the 3-phase vibration and rubbing  Monitor and trend ambient
stator winding.  Breakdown of insulation due to electrical transients temp, winding temp, motor
and surges amps, rpm, process fluid temp,
 Degradation of insulation due to moisture, cooling water temp.
lubricant, chemical reactions, or dirt  Monitor and trend
 Manufacturing defect in insulation insulation condition
 Mechanical damage from loose part or ingested part parameters
 Improve quality of station
electric power.
 Consider upgrading to higher
insulation class on next rewind
 Winding-to  Same as above  Same as  Same as above
winding above
28
Fault
 Turn-to-turn  Same as above  Same as above  Same as above

Fault
 Open  Breakdown of insulation and melting of conductor  Same as above  Same as above
winding due to electrical transients and surges
 Broken winding conductor due to vibration,
electromagnetic transients, and/or cyclic fatigue.
Stator  Connect motor  Phase-to-  Same as above  Same as above  Good maintenance practices to
Leads and line terminations ground keep motor ventilation clean
Coil to individual fault and unrestricted.
Cross-ties stator winding  Phase-to-  Same as above  Same as above  Use surge capacitors.
coils phase  Good operating practices to
Fault reduce number of starts.
 Open circuit  Breakdown of insulation and melting of conductors  Same as above  Monitor and trend vibration.
due to electrical transients and surges  Monitor and trend ambient
 Broken conductor due to vibration, electromagnetic temp, winding temp, motor
transients, and/or cyclic fatigue. Mechanical amps, rpm, process fluid temp,
damage from loose part or ingested part cooling water temp.
 Mechanical damage from contact with rotating part  Improve quality
of station electric power
29
 Degradation
 Loose  Loosening of leads, coil crossties, and fasteners due and damage to  Same as above
leads or to vibration, electromagnetic transients, and/or insulation and
coil cross- cyclic fatigue. conductors
ties  Mechanical damage from loose
part or ingested part
Stator Core  Insulated,  Loose  Loosening of stator core assembly due to vibration  Increased losses  Periodic visual or borescope
magnetic iron laminations  Loosening of stator core assembly due to (heat) due to larger inspection
alloy and locking electromagnetic transients heat flux.  Periodic infrared
laminations, bars in stator  Misalignment of core assembly during manufacture  Increased motor thermography surveys and
bound core current trending
together by Assembly  Monitor and trend vibration
locking bars to  Monitor and trend motor
form stator core; amps, rpm, winding temp
magnetic flux  Lamination  Thermal degradation and wear of lamination  Same as above  Same as above
path for stator overheating insulation
Rotor  Rotating  Rotor bars  Fatigue due to vibration and mechanical cycling  Increased rotor  Periodic visual or borescope
Squirrel magnetic field cracked at  Fatigue due to electro-magnetic cycling and cage resistance and inspection
Cage produced by 3 end ring transients heating  Periodic infrared
Assembly phase voltage  Defective welds or brazed Joints thermography surveys and
applied to stator trending
winding induces  Rotor bars  Loosening due to vibration and mechanical cycling  Increased vibration  Monitor and trend vibration
30
currents in the loose in core  Loosening due to electromagnetic cycling and and wear of core  Monitor and trend motor
squirrel cage slots transients laminations amps, rpm, winding temp
rotor  Loosening due to thermal cycling and excessive insulation
circuit that starting
develop  Defective swaging during Manufacture
the same number
of
rotor poles as  Broken rotor  Same as above  Same as above
there bar
are stator poles.
Rotor Core  Insulated,  Loose  Loosening of rotor core assembly due to vibration  Increased losses  Periodic visual or borescope
magnetic iron laminations  Loosening of rotor core (heat) due to larger inspection
alloy and locking assembly due to electromagnetic leakage flux  Periodic infrared
laminations, bars in stator transients  Increased motor thermography surveys and
bound core  Misalignment of core Current trending
together by Assembly assembly during manufacture  Monitor and trend vibration
locking  Monitor and trend motor amps,
bars to form rpm, winding temp
stator  Lamination  Thermal degradation and  Same as above  Same as above
core; magnetic overheating wear of lamination insulation
flux
path for stator
31
 Support and  Wear of  Insufficient or excessive lubrication  Excessive vibration  Periodic inspection and cleaning
Antifrictio provide bearing  Dirt, moisture, or other contamination in lubricant  Thermal of bearings
n Bearings for movement of rollers  Wrong lubricant breakdown  Periodic inspection and
rotating elements and race  Lube oil cooling insufficient; high bearing or burning of  maintenance of lube oil system,
of temperature lubricant including oil sample
the motor  Unbalanced or misaligned rotating elements  Reduced bearing analysis
 Transverse mechanical Life  Monitor and trend vibration
 Loading Material degradation due to  Monitor and trend bearing
corrosion temperature
 Material degradation due to circulating current
 Failure of  Same as above  Same as above

 Same as above
rollers, roller
cage, or
race; bearing
seizure
Bearing  Maintain lube oil  Degradation  Wrong lubricant  Loss of lubricant  Periodic inspection of bearings
Seals or grease within and  Lube oil cooling  Increased bearing and seals
bearing housing deformation insufficient; high bearing temperature temperature  Periodic replacement of bearing
and prevent entry of seals  Unbalanced or misaligned Entry of dirt, dust, seals
of dirt, moisture, rotating elements moisture, or other
and other  Normal aging contaminants into
contaminants  Installation error lubricant
32
3.6 FAULT TREE ANALYSIS OF INSULATION BREAKDOWN
One of the most critical component of an induction motor and also one of the main sources of
their failure is the stator winding insulation system. Various surveys on motor reliability have
been carried out over the years where the percentage of motor failures due to problem with
the insulation is about 26%. The unscheduled process downtime caused by a failure of the
insulation system can cause enormous costs. [20]
Figure 3.5 : FTA of Induction motor winding failure
33
3.7 INFERENCE FROM PRELIMINARY STUDIES
Based on the Failure data analysis, failure mode and effect analysis and fault tree analysis, it
can be concluded that winding insulation failure forms a major component in the total failures
of an induction motor.
Thus, the ambit of this report is to study and analyse the effect of insulation degradation on
the remaining useful life of an induction motor.
34
CHAPTER 4
INTEGRATED APPROACH TO REMAINING LIFE ESTIMATION
4.1 INTRODUCTION
Motor life depends on electrical, thermal, mechanical, and environmental stresses. This
chapter defines the loss of motor life as the loss of stator winding insulation life due to
thermal and voltage stresses. From a thermal point of view, the stator winding insulation is
the weakest part of the squirrel cage induction motors, and equations have been developed to
estimate the insulation life and, hence, motor life. Since the rotor is of squirrel cage type,
there is no rotor insulation, the rotor structure is more robust and hence, motor life is focused
on stator winding insulation.
In order to estimate motor life, an integrated approach is used where electrical, thermal and
insulation ageing model are developed. This is shown in Figure 4.1 [9]
Figure.4.1 : Motor Life Prediction Flowchart
From the flowchart it can be seen that the first step to determine the remaining life is
development of an electrical model using motor design data and use it to calculate the losses.
The losses that eventually cause the heating of an induction motor are an input to the thermal
model which is based on a single time constant thermal model of an induction motor which
gives the temperature rise that a motor is subjected to. The temperature rise depends on the
thermal resistance, thermal capacitance, ambient temperature and losses. The thermal model
gives the temperature which is the input to the thermal ageing model. The thermal ageing
35
model is based on Arrhenius Equation/Eyring Equation using constants for a typical class F
insulation system. This model which assumes that insulation degradation is a function of
temperature only, gives the remaining life of the insulation as the output.
Another novel way to estimate the Remaining Life by Fuzzy Logic theory is also discussed.
4.2 ELECTRICAL MODEL OF AN INDUCTION MOTOR.
The electrical model is used to simulate the motor’s electrical performance characteristics.
The per phase electrical model of the induction motor, with rotor referred with respect to the
stator can be represented as [3]
Figure 4.2 Equivalent Circuit (Per Phase) of induction motor
Where the various parameters are explained below
Parameter Symbol Parameter

V Voltage (Phase to ground)
I1 Stator Current
I0 Magnetising current
I1’ Total current
R0 No Load resistance
X0 No Load leakage reactance
R1 Stator resistance (Per phase)
X1 Stator reactance (Per phase)
R2’ Rotor resistance (referred to stator)
X2’ Rotor reactance (referred to stator)
RL Load resistance
S Slip
Table 4.1 Equivalent circuit parameters of induction motor
Determination of equivalent circuit of motor using No-Load and Blocked Rotor Test [3]
No-Load Test
36
The no load test on an induction motor gives information with respect to exciting current and
no-load losses. The test is performed at rated frequency and with balanced poly-phase
voltages applied to the stator terminals. Readings are taken at the rated voltage, after the
motor runs long enough for the bearings to be properly lubricated.
The behaviour of the machine may be judged from the equivalent circuit of Fig. 4.3. The
current drawn by the machine causes a stator-impedance drop and the balance voltage is
applied across the magnetizing branch. However, since the magnetizing branch impedance is
large, the current drawn is small and hence the stator impedance drop is small compared to
the applied voltage (rated value). This drop and the power dissipated in the stator resistance
are therefore neglected and the total power drawn is assumed to be consumed entirely as core
loss. The current drawn is at low power factor. This test enables us to compute the resistance
and inductance of the magnetizing branch in the following manner.
Figure 4.3: Equivalent Circuit at No-load
S.No. Parameter Symbol

1 Voltage (No-Load)
2 Current (No-Load)
3 Power Factor (No-Load)
4 Power (No-Load)
5 Stator Winding Resistance
6 Magnetising Current
7 No Load resistance
8 No Load leakage reactance
Table 4.2 No-load Equivalent circuit parameters of induction motor
37
= 3∗ ∗ ∗ . (4.3)
√3
√3 ∗
0= (4.4)
( ∗ . )
= (( ) −( ∗ . ) ) (4.5)
√3 ∗
0= (4.6)
( )
Blocked Rotor Test [3]
Like the short-circuit test on a transformer, the Blocked-Rotor test on an induction motor
gives information with respect to the leakage impedances. The rotor is blocked so that it
cannot rotate (hence the slip is equal to unity), and balanced polyphase voltages are applied to
the stator terminals.
The equivalent circuit for blocked-rotor conditions is identical to that of a short circuited
transformer. An induction motor is more complicated than a transformer, however, because
its leakage impedance may be affected by magnetic saturation of the leakage-flux paths and
by rotor frequency. The guiding principle is that the blocked-rotor test should be performed
under conditions for which the current and rotor frequency are approximately the same as
those in the machine at the operating condition for which the performance is later to be
calculated. For example, if one is interested in the characteristics at slips near unity, as in
starting, the blocked-rotor test should be taken at normal frequency and with currents near the
values encountered in starting. If, however, one is interested in normal running
characteristics, the blocked-rotor test should be taken at a reduced voltage which results in
approximately rated current; the frequency also should be reduced, since the values of rotor
effective resistance and leakage inductance at the low rotor frequencies corresponding to
38
small slips may differ appreciably from their values at normal frequency, particularly with
double-cage or deep-bar rotors.
Figure 4.4: Equivalent Circuit at Blocked-Rotor
The total leakage reactance at normal frequency can be obtained from this test value by
considering the reactance to be proportional to frequency. The effects of frequency often are
negligible for normal motors of less than 25-hp rating, and the blocked impedance can then
be measured directly at normal frequency. The importance of maintaining test currents near
their rated value stems from the fact that these leakage reactances are significantly affected
by saturation. Based upon blocked-rotor measurements, the blocked-rotor resistance
01 can be found from the blocked-rotor voltage and current as shown below. Similarly, the
blocked rotor reactance can be found. Once these parameters have been determined, the
equivalent circuit parameters can be determined

1 Voltage (Blocked Rotor)
2 Current (Blocked Rotor)
3 Power Factor (Blocked Rotor) . .
4 Power (Blocked Rotor)
5 Equivalent Impedance 01
6 Equivalent reactance 01
7 Equivalent resistance 01
Table 4.3 Blocked rotor equivalent circuit parameters of induction motor
ℎ ℎ = (4.7)
√
√ ∗
01 = (4.8)
39
01 = (4.9)
01 = ( 01) − ( 01) (4.10)
01 = 1 + 2 (4.11)
01 = 1 + 2 (4.12)
1 = 0.5 ∗ 01 (4.13)
2 = 05 ∗ 01 (4.14)
Thus the induction motor parameters can be found using No-Load and Blocked rotor test.
Determination of winding copper loss [3]

1 Rotor Speed
2 Synchronous Speed
3 Slip
4 Load resistance ′
5 Shunt Impedance
6 Effective Impedance
Table 4.4 Parameters of induction motor
( )=
ℎ ( )= 1500 (4.15)
−
( )= (4.16)
1
= 2 −1 (4.17)
= 0|| 0 (4.18)
= ( 1+ 2 + )+ ( 1+ 2 ) (4.19)
1 = (4.20)
0 = (4.21)
= 0 + 1 (4.22)
=3∗ ∗ (4.23)
40
Effect of Unbalance
An excessive level of voltage unbalance can have serious impacts on mains connected
induction motors. The level of current unbalance that is present is several times the level of
voltage unbalance. Such an unbalance in the line currents can lead to excessive losses in the
stator and rotor that may cause protection systems to operate causing loss of production.
Although induction motors are designed to tolerate a small level of unbalance they have to be
derated if the unbalance is excessive. If operated at the nameplate rated capacity without
derating the useful life of such induction motors can become quite short. If an induction
motor is oversized to a given application then some level of protection is built into its
operation although the motor does not operate at the best efficiency and power factor. [14,
15]
Causes of voltage unbalance include unequal impedances of three-phase transmission and
distribution system lines, large and/or unequal distribution of single-phase loads, phase to
phase loads and unbalanced three-phase loads. When a balanced three-phase load is
connected to an unbalanced supply system the currents drawn by the load also become
unbalanced.
NEMA (National Electrical Manufacturers Association of USA) standard definition that is
given by equation
Maximum deviation from mean of( V , V , V )

Voltage unbalance = (4.24)
Mean of ( V , V , V )
The greatest effect of voltage unbalance is on three-phase induction motors. Three phase
induction motors are one of the most common loads on the network and are found in large
numbers especially in industrial environments.
41
The rotating magnetic field which rotates in the opposite direction induces voltages in the
rotor. These voltages have a frequency that is double the system frequency. Since the
frequency of this rotor voltage is higher, it flows on the surface of the rotor due to the skin
effect and causes surface heating which can lead to motor damage.
As an example, a motor with a locked rotor current that is 6 times the running current would
give rise to a very significant 30% unbalance in the motor line current if the voltage
unbalance is 5%.
If the motor is fully loaded some stator phase windings and the rotor will carry more current
than that is permitted thus causing extra motor losses. This will lead to a reduction in motor
efficiency while reducing the insulation life caused by overheating.
In addition to reduced efficiency, overheating and loss of insulation life, induction motors
operating with unbalance will be noisy in their operation caused by torque and speed
pulsations. Obviously in such situations the effective torque and speed will be less than
normal.
To study the effect of unbalance voltage on the life of winding, sequence component theory is
used to derive the positive and negative sequence component of voltages as shown below.
42
1 1 1
2 = 1 (4.25)
3
0 1 1 1
Where, = 1120
V1, V2 and V0 are the positive, negative and zero sequence components of the line voltages
Vab, Vbc and Vca. The positive sequence voltage V1 is used to find the positive sequence
current I1.
Using the positive sequence component of voltage, the positive sequence circuit is used to
analyse the motor behaviour under voltage unbalance. The positive sequence circuit is shown
below.
Figure 4.5: Positive Sequence Equivalent Circuit
The negative sequence current (I2) is found as:
= 1 + 2 (4.26)
Also, the ratio of positive sequence impedance to negative sequence impedance equivalent to
ratio of starting current to full load current. Thus the negative sequence current can be
calculated by multiplying the negative sequence voltage with the ratio of starting to full load
current.
The stator winding loss is found as:
=3∗ ∗ (4.27)
43
4.3 THERMAL MODEL OF AN INDUCTION MOTOR. [11]
In the second stage, the output of the electrical model i.e. the I2R losses is used to estimate
the winding temperature of the stator. The thermal model of an induction motor is based on
single thermal capacitance and single thermal resistance i.e. with single thermal time
constant. The thermal capacitance and thermal resistance are normally predetermined by a set
of parameters for a given class of motors, classified by their full load current (FLC), service
factor (SF) and trip class (TC). Otherwise, these constants can be determined experimentally.
Thermal models with a single thermal capacitor and single thermal resistor are derived from
the heat transfer of a uniform object, as shown in Figure 4.4.
Figure.4.6: Heat Transfer model with single thermal time constant
The quantities θ and θ , in 0C, are temperatures of the uniform object and its ambient,
respectively. The power input into this uniform object is determined by the power losses from
the current, I (Unit: A), on the resistor, R (Unit: Ohms). Heat is dissipated through the
boundry of the uniform object (the shaded region in Figure 4.4) to the ambient. The thermal
resistance, Rth, in 0C/W, models this heat transfer. The thermal capacitance, Cth, in J/0C, is
defined to be the energy needed to elevate temperature by one degree Celsius for the object.
It represents the total thermal capacity of the object.
44
The difference between the input power and the output power is used to elevate the
temperature of the uniform object.
d(θ − θ )
Pin − Pout = Cth ∗ (4.28)
dt
The input power is the heat, I2R, generated by the current in the resistor. The output power is
the heat transfer, θ − θ /Rth, across the boundry of the object to its ambient. Therefore, the
above equation can be written as:
θ−θ d(θ − θ )
I R− = Cth ∗ (4.29)
Rth dt
Solving the above equation as a first order differential equation, a closed form solution is
obtained.
θ(t) = I R. Rth 1 − e τ + θ (4.30)
Where,
τth = Cth ∗ Rth which is the thermal time constant of the motor.
If the constant current, I, flows through this uniform object for a sufficiently long time, i.e.
for t, the final temperature of this uniform object is
θ() = I R. Rth + θ (4.31)
For a specific motor, it is designed to work under some permissible temperature, θ ,
determined by the stator winding insulation material. This maximum temperature determines
the maximum permissible current through the stator winding,
Imax = (θ − θ )/R. Rth (4.32)
For a motor, if the stator current exceeds a predetermined value for a certain time, the stator
winding temperature will rise above its maximum permissible value.
Thus the stator winding temperature as a function of time and current can be written as :
θ(t) = (θ − θ )(I/I_max) 1−e τ + θ (4.33)
45
For a typical induction motor, the temperature rise in the stator windings as a function of time
is written as:
Figure.4.7: Variation of stator winding temperature with time with ambient temperature 30 C
4.4 INSULATION AGEING MODEL [9,12,13]
With time, the insulation becomes brittle and shrinks, leading to cracks. The insulation at the
point of cracks weakens gradually as surrounding pollutants find their way through these
cracks. The weakening of insulation with time is called ‘ageing’.
The life of the insulation will also be affected by an excessive operating temperature. Ageing
occurs when a machine is occasionally over-loaded. Sometimes the size of the machine may
be only marginal when it was initially chosen and with the passage of time, it may be
required to perform duties that are too arduous. Every time the machine over-heats, the
insulation deteriorates, and this is called thermal ageing of insulation.
46
The insulating system is one of the main parts of electrical rotary machine. High operating
and reliability demands require the proper technological steps to manufacture the insulating
system. Much effort has been invested over several decades in studying the aging
characteristics of the various electrical insulating system designs and insulating materials
employed in high-voltage equipments. These studies aim mainly to allow reasonable
estimates of the service life expectancies of such equipment and to assess their reliability in
operating conditions after a given number of years in service. The aging of a polymeric
material, or of any other material for that matter, inherently involves the alterations of the
material’s physical and/or chemical structure which is expected to be related to the changes in
the physical and chemical properties of the material. When the aging of a dielectric material
is evoked, it usually implies that the alterations in the properties of the material are
detrimental to its service operation and service reliability. When these properties have
deteriorated to the point where the material can no longer operate safely under normal stress
conditions, it implies that it has reached the end of its useful life. The causes of aging are yet
to be fully understood, but obviously, the degree of aging strongly depends on the nature of
the involved material and on the nature and duration of the applied stresses. Indeed, aging can
be induced by a combination of the various stresses (electrical, mechanical, thermal or
environmental) to which the insulation system is subjected. The simultaneous application of
these stresses leads to the interaction of aging mechanisms.
This project deals with estimation of insulation age based on the following approaches:
 Arrhenius Model
 Eyring Model
 Fuzzy Logic
47
4.4.1 Arrhenius Model
Motor winding failures due to stator insulation breakdown has been found to be one of the
major causes of motor failure. The motor insulation life is affected by thermal, electrical, and
mechanical stresses, as well as environmental conditions. In this study, the thermal stress is
considered to dominate other stresses. That is, the insulation aging process will depend on the
magnitude and duration of the operating temperature. The winding insulation has a specific
lifetime, and it deteriorates with time due to thermal stresses. Therefore, motor life can be
predicted by estimating the stator winding insulation life.
The Arrhenius Model is based on Arrhenius equation. The thermal ageing model that
modifies the 10◦ C rule (insulation life is halved for every 10◦ C rise) to a more accurate half
interval index is shown in the equation below.
= 100. 2( )/
(4.34)
Where,
Lx : Percent lifetime at temperature Tx (in degrees centigrade);
L100 :Percent lifetime at rated temperature Tc (in degrees centigrade);
Tc :Hot-spot temperature for insulation class (in degrees centigrade);
Tx :Total allowable temperature for insulation class (in degrees centigrade);
HIC halving interval (in degrees centigrade); (14, 11, 9.3, 8, and 10 for class A, B, F, H and
H’, respectively)
There is another form of Arrhenius’ equation which can be used to estimate insulation life
which takes into consideration the variation of ambient or operating temperature magnitude
with time and estimates life lost as a function of operating time. The above equation however,
assumes that the motor is operating continuously at the same temperature throughout its life
span and estimates the expected life at that temperature.
48
tr Ea −
ln = (4.35)
ti k
Where
tr time at temperature Tr
ti time at temperature Ti
= 1.05 eV for Class F insulation
K =0.8617* 10−4 eV/K
The insulation class limit is normally based on a 20000 h (2.3 years) period. That is, the
insulation is expected to operate continuously at its maximum temperature for 2.3 years
without failing. For a motor to operate for more than 2.3 years, the manufacturer will have to
use a higher insulation class, as the life will be reduced further under unbalanced supplies.
Even if a motor has class A temperature rise with class F insulation, the insulation life will
deteriorate with time. The insulation life can be estimated using class A temperature rise and
then substituted to class F insulation using above equation. The insulation life using clas sA
temperature rise is found to be 43.2 times class F insulation life at an ambient temperature of
40◦C. At a lower ambient temperature of 25 0C, class A insulation life is 58 times that of class
F. Therefore, the motor will last longer at lower ambient temperatures. For example, if class
F insulation life is 2.3 years, this motor will last for 43.2 * 2.3 = 99.4 years or 58 * 2.3 = 133
years at an ambient temperature of 400 0C and 250 0C, respectively. The motor is expected to
last for these years if operated continuously at full load at constant temperature.
Thus using the above model the remaining life of the insulation can be predicted.
4.4.2 Eyring Model

Chemical reactions in progress between water ions and insulation (hydrolysis) may lead to
insulation destruction. Insulation mechanical properties can disrupt with moisture. Electrical
properties with moisture absorption changing significantly (material dielectric strength
significantly change, dielectric losses increase, electric stress structure redistribution). This
49
leads to sparkover, to insulation surface breach due to partial discharges followed by erosion
of insulation surface. Synergistic effects may occur between an electric field and humidity,
which can lead on some polymers to creating water trees. Dynamics of condensation,
capillarity and absorption and resulting material changes depends on exposure to moisture
mode (steam or liquid).
Therefore, it has been seen that besides temperature, another factor that causes accelerated
degradation of motor winding insulation is humidity. Thus, there is a need to redefine the
insulation ageing model taking into consideration the effect of humidity also. This can be
done using Eyring’s Model which is an extended version of Arrhenius Model. The Eyring’s
Model is explained below:
The key to this model formulation was thorough understanding of model parameters based on
two independently changing degradation factors – humidity and temperature.
1 1 1 1
( )= ∗ − ∗ − (4.36)
tr time at temperature Tr
RHr is relative humidity at temperature Tr
RHi is relative humidity at temperature Ti
(Activation Energy for temperature)= 1.05 eV for Class F insulation

(Activation Energy for humidity)= Unknown
K =0.8617* 10−4 eV/K
Effect of Frequent Start/Stops [13]
Repeated motor starts also contribute to reduction in life due to increased temperatures since
all losses in an electric motor are converted into heat. Part of this heat energy is dissipated by
the motor through the mechanisms of thermal radiation, conduction and convection, while the
remainder of this energy causes the temperature of the motor to rise. When the heat
50
absorption by the motor becomes zero, or in different words, when the rate of heat dissipated
becomes equal to the rate of heat generated by the motor, steady-state temperature is reached.
This temperature in a properly designed motor represents the temperature rating of its
insulation system (temperature rise in the insulation plus the hottest spot allowance plus the
temperature of the ambient).
It remains then to determine what magnitude of heating may be expected on motor stator
windings. Calculations of this are approximate in that all the heat generated during the
interval of acceleration is assumed to be absorbed and that the conductor cross-section
follows an average practice, namely that 500 circular mils per ampere (0.253 "'/ampere) (at
full load) is used.
With these assumptions, the temperature rise in the stator windings can be calculated, using
the following expression:
10
T = 2.3 ∗ (4.37)
D
where T is the temperature rise, OC, D is the circular in mils per ampere (5.061 x 10-4
mm2/ampere) Assuming, for example, that a motor-driven equipment requires 30 s to reach
operating speed, that the current during this period is 5 x the full-load current (on the
average), and that the motor stator conductors have a cross-section of 500 circular mils per
ampere (0.253 mm2/ampere) at normal full load, the per-ampere area is effectively reduced to
100 circular mils. Thus, the total temperature rise at the end of one start is, by above
equation, T = 2.3 x 104/1002 x 30 = 69°C.
If one assumes that this motor is started right after it has been shut off, then this temperature
rise of 69°C can be added to the operating temperature. Thus, the windings experience a
temperature 69°C above its classification temperature rating.
If two additional assumptions are made, namely that (1) the insulation ageing curve has a
10°C slope, or that for every 10°C rise the rate of insulation deterioration doubles, and that
(2) the rounded-off temperature rise is 700C, then during the acceleration period the rate of
deterioration will be increased by a factor of 270/100, or will be 128 x that for the normal
temperature. At this rate of deterioration, the 30 s accelerating period will be equivalent to 30
x 128 or 3840 s - about one hour. Each motor start, therefore, will reduce the calculated
life by one hour. This is a conservative estimate which can be more accurately determined
by calculating the changes in life expectancies incrementally for each second of temperature
rise.
51
4.4.3 Fuzzy Logic [6, 7, 8]
A stator current signal contains potential fault information. The most suitable measurements
for diagnosing the faults under consideration, in term of easy accessibility, reliability, and
sensitivity, are the stator current amplitudes Ia, Ib, and Ic.
Figure.4.8 : Fuzzy Logic Block Diagram
Fuzzy systems rely on a set of rules. These rules, while superficially similar, allow the input
to be fuzzy, i.e. more like the natural way that humans express knowledge. Thus, a power
engineer might refer to an electrical machine as “somewhat secure” or a “little overloaded”.
This linguistic input can be expressed directly by a fuzzy system. Therefore, the natural
format greatly eases the interface between the engineer knowledge and the domain expert.
Furthermore, infinite graduations of truth are allowed, a characteristic that accurately mirrors
the real world, where decisions are seldom “crisp”.
As stated, the induction motor condition can be deduced by observing the stator current
amplitudes. Interpretation of results is difficult as relationships between the motor condition
and the current amplitudes are vague. Therefore, using fuzzy logic, numerical data are
represented as linguistic information
52
In our case, root mean square (rms) the stator current amplitudes Ia, Ib, and Ic are considered
as the input variables to the fuzzy system. The stator condition in terms of the Residual
Useful Life (RUL) is chosen as the output variable. All the system inputs and outputs are
defined using fuzzy set theory.
µia(Iaj)
Ia = ϵ Ia (4.38)
Iaj
µib(Ibj)
Ib = ϵ Ib (4.39)
Ibj
µic(Icj)
Ic = ϵ Ic (4.40)
Icj
µrul(RULj)
RUL = ϵ RUL (4.41)
RULj
where , Ibj , Icj, and RULj are, respectively, the elements of the discrete universe of discourse
Ia, Ib, Ic, and RUL. µia(Iaj ) µib(Ibj) µic(Icj ) µrul(RULj) are, respectively, the
corresponding membership functions.
Basic tools of fuzzy logic are linguistic variables. Their values are words or sentences in a
natural or artificial language, providing a means of systematic manipulation of vague and
imprecise concepts. More specifically, a linguistic variable is characterized by a quintuple (x,
T(x), U, G, M), where x is the variable name; T(x) is the set of names of the linguistic values
of x, each a fuzzy variable, denoted generically by x and ranging over a universe of discourse
U. G is a syntactic rule for generating the names of x values; M is the semantic rule
associating a meaning with each value.
For instance, the term set T (RUL), interpreting stator condition, RUL, as a linguistic
variable, could be
T (RUL) ={100% , 80%, 60%,40%,20%}
Where each term in T (RUL) is characterized by a fuzzy subset, in a universe of discourse

RUL. Good might be interpreted as a stator with no faults, damaged as a stator with voltage
unbalance, and seriously damaged as a stator with an open phase.
Similarly, the input variables Ia, Ib, and Ic are interpreted as linguistic variables, with
T (Q)={Zero, Small , Medium, Big}
53
Where Q = Ia, Ib, Ic, respectively.
Fuzzy rules and membership functions are constructed by observing the data set. For the
measurements related to the stator currents, more insight into the data is needed, so
membership functions will be generated for zero, small, medium, and big. For the
measurement related to the stator condition, it is only necessary to know if the stator
condition is good, damaged, or seriously damaged. The optimized membership functions for
this problem are shown in Figs. 3 and 4. Once the form of the initial membership functions
has been determined, the fuzzy if-then rules can be derived. In this study, two faults have
been investigated: stator voltage unbalance and open phase.
Figure.4.9 : Fuzzy logic –Linguistic variables
Here Z= Zero, S= Small, M=Medium, B=Big

Figure.4.10 : Membership Function of Ia,Ib and Ic
54
Figure.4.11 : Membership Function for RUL
These rules have been optimized so as to cover all the healthy and the faulty cases. For our
study, we have obtained the following 14 if-then rules.
Rule (1) : If Ia is Z Then RUL is 20 %
Rule (2) : If Ib is Z Then RUL is 20 %
Rule (3) : If Ic is Z Then RUL is 20 %
Rule (4) : If Ia is B Then RUL is 20 %
Rule (5) : If Ib is B Then RUL is 20 %
Rule (6) : If Ic is B Then RUL is 20 %
Rule (7) : If Ia is S and Ib is S and Ic is M Then RUL is 60%
Rule (8) : If Ia is S and Ib is M and Ic is M Then RUL is 60%
Rule (9) : If Ia is M and Ib is S and Ic is M Then RUL is 60%
Rule (10) : If Ia is M and Ib is M and Ic is M Then RUL is 100%
Rule (11) : If Ia is S and Ib is S and Ic is S Then RUL is 100%
Rule (12) : If Ia is S and Ib is M and Ic is S Then RUL is 60%
Rule (13) : If Ia is M and Ib is S and Ic is S Then RUL is 60%
Rule (14) : If Ia is M and Ib is M and Ic is S Then RUL is 60%
55
Figure.4.12 : Relationship of Ia with RUL
Thus it can be seen that the remaining life (RUL) can be found by continuous monitoring of
the stator currents. The if-then rules can be implemented using a Fuzzy Logic Controller in
simulation software. It is assumed that a very low or a very high stator current indicates a
fault/mal-operation, and thus the motor life is lesser. A medium value of stator current
indicates a healthy motor with high remaining life.
56
CHAPTER 5
EXPERIMENTAL STUDY
5.1 GENERAL DETAILS
MOTOR NAME PLATE DATA
Make Siemens
Connection Delta
Voltage 415 V
HP 3 (2.2 kW)
Full Load Ampere 4.8 A
RPM 1430
Power Factor 0.79
Insulation Class B
Table 5.1: Motor Name Plate Data
AUTO-TRANSFORMER
Voltage 1. 0-415 V for A1,

A2,A3
2. 0-460 V for
B1,B2,B3
Phase 3
Frequency 50/60 Hz
Table 5.2: Auto-Transformer Name Plate Data
EDDY CURRENT DYNAMOMETER
Eddy current dynamometer is a swinging stator absorption dynamometer . The dynamometer
works on electromagnetic principle. The dynamometer mechanical unit is composed of two
sections, the stator and rotor. A toothed rotor is centrally placed leaving controlled equal
facial gap on its either side. The rotor disc is bolted to two half shafts. The rotor has a
cylindrical roller bearing at its driving end and a ball bearing at its non driving end. The stator
is also supported (cradled) on two ball bearings. The whole system is mounted on a base
frame. The rotor is coupled to the prime mover usually by a double universal joint propeller
shaft.
The stator is composed of casing, an excitation coil, two end covers, two cooling chambers
with water passages for cooling water circulation. These parts are supported on two non-
57
magnetic adopters. The stator has also cooling water pipelines, inlet and out pipes connected
to end covers.
When D.C. current is passed in the excitation coil, a torroidal magnetic flux is set up axially
around the rotor through casing, end covers and annular face of cooling chambers. When
rotor rotates , the reluctance changes due to toothed construction of rotor. This action
generates eddy currents in both the cooling chambers. These eddy currents oppose the motion
of the rotor creating ‘braking action’ on the prime mover. This ‘braking action’ is transferred
equally on the stator, as action is equal to reaction. If DC current is increased, braking torque
also increases and vice-versa.
As the stator is cradled and attached to the load cell (force measuring device), this reaction
force gets registered on the digital load indicator through load cell and electronic circuitry.
The torque arms are attached to the stator. The torque radius is accurately controlled in length
during manufacture. Pans are suspended at this distance. Dead weights are placed in one pan
for torque calibration.
A 60 teeth gear wheel and speed sensor is used to measure speed in r.p.m.
Figure 5.1 shows the experimental setup comprising the 2.2kW, 415V motor and load i.e.
Eddy current dynamometer.
58
Figure 5.1: Experimental Setup- Motor with Eddy Current Dynamometer (Load)
Power calculation is made using the following relation,
( . )= ℎ ℎ ( )∗ ( ) (5.1)
. .= 2∗ ∗ ∗ (5.2)
4500
Modes of Operation
1. I Mode- Excitation Current Constant

2. N Mode- Speed Constant
3. T Mode- Torque Constant
Torque Calculation:
= . (5.3)
Where,
P= Rated Power of Induction Motor
W= Speed
T= Torque
59
2 ∗ 1500
2.2 ∗ 10 = .
60
Solving,
T=14.3 N-m
Full Load Operation
Load Value Set, W (Torque)= 14
Dynamometer Constant, K (From Name Plate Data) = 9550
N = 1440
= ∗ (5.4)
On Solving,
P= 2.11 kW
5.2 DETERMINATION OF EQUIVALENT CIRCUIT OF MOTOR

Stator Winding Resistance = 3.5 Ohms
Insulation Resistance= 4 M –ohms
No-Load Test Results
S.No Parameter Symbol Value

.
1 Voltage 415 V
2 Current 3.1 A
3 Power Factor . . 0.134
4 Power 296.95 W
5 Stator Winding Resistance _ 3.5 Ohms
6 Magnetising Current 3.02 A

7 No Load resistance 0 1730
ohms
8 No Load leakage reactance 0 233.98
ohms
Table 5.3 : No-load Test Results
60
Separation of friction and windage losses
Separation of losses
500
450
400
350
300
No Load Losses 250
(W) 200
150
100
50
0
0 20 40 60 80 100
Voltage (% of V)
Figure 5.2 : Separation of losses
No Load Losses= Core Loss+ Friction& Windage Loss+ Stator Copper Loss
By extrapolating the graph to V= 0,
Friction & Windage Loss= 170 W
ℎ = = 1.78 (5.5)
√3
_ 3 ∗ (1.78) ∗ 3.5 = 33.26 (5.6)
+ + = 750 − 33.26 = 716.73 (5.7)
= 3 ∗ 415 ∗ 1.78 ∗ 0.134 = 296.95 (5.8)
√3 ∗
0= = 1730.38 ℎ (5.9)
( ∗ . )
= (( ) −( ∗ . ) ) = 3.02 (5.10)
√3 ∗
0= = 233.98 ℎ (5.11)
( )
61
Blocked Rotor Test Results
S.No. Parameter Symbol Value

1 Voltage 50 V
2 Current 4.4 A
3 Power Factor . . 0.4
4 Power 150 W
5 Equivalent Impedance 01 19.68 Ohms
6 Equivalent reactance 01 18.08 Ohms
7 Equivalent resistance 01 7.75 Ohms
Table 5.4 : Blocked-Rotor Test Results
ℎ ℎ = = 2.54 (5.12)
√3
√3 ∗
01 = = 19.68 ℎ (5.13)
01 = / = 7.75 Ohms ( 5.14)
01 = ( 01) − ( 01) = 18.089 Ohms (5.15)
Figure 5.3: Induction motor equivalent circuit
01 = 1 + 2 (5.16)
01 = 1 + 2 (5.17)
1 = 0.5 ∗ 01 (5.18)
2 = 05 ∗ 01 (5.19)
62
Parameters Value
0 1730.38 Ohms
0 233.98 Ohms
1 3.5 Ohms
2 4.25 Ohms
1 9.04 Ohms
2′ 9.04 Ohms
Table 5.5: Equivalent circuit parameters of induction motor
Determination of winding copper loss
Parameter Symbol Value

Rated Speed 1430 RPM
Synchronous Speed 1500 RPM
Slip 0.046
Load Resistance 88.14
Shunt Impedance 31.07 + 229.78
Effective Impedance 95.89 + 18.08
Table 5.6: Parameters of induction motor
( ) = 1430
ℎ ( )= 1500
−
( )= = 0.046 (5.20)
1
= 2 − 1 = 88.14 ℎ (5.21)
= 0|| 0 = 31.07 + 229.78 (5.22)
= ( 1+ 2 + ) + ( 1 + 2 ) = 95.89 + 18.08 (5.23)
= 4.8
= 3∗ ∗ = 80.73 (5.24)
63
5.3 TEMPERATURE MEASUREMENT
The motor was coupled with the load and run at full load. Temperature of the stator winding
was found and the following observations were made.
Time (Minutes) Temperature

0 30
5 39
10 44.7
15 47.2
20 48.8
25 49.5
30 49.7
40 50
50 50.1
60 50.1
Table 5.7: Induction Motor- Temperature measured at various intervals
55
50
45
Measured
Temperature
(0C) 40
35
30
0 10 20 30 40 50 60
Time (In Minutes)
Figure 5.4: Measured Temperature v/s Time for 2.2kW, 415 V Induction Motor running at full load
Using the equation,
θ(t) = (θ − θ )(I/I_max) 1−e τ + θ (5.25)
Parameter Value
θ 130 0C
θ 30 0C
Loading 100%
I 4.5 A
Table 5.8: Thermal Model parameters of induction motor
64
Imax = (θ − θ )/R. Rth (5.26)
ζ th = Rth * Cth (5.27)
Determination of Rth and Cth for the induction motor
The temperature rise of the motor under full load conditions is fit using the equation (5.25)
given below,
θ ( t) = ( θ − θ )(I/I_max) 1−e τ + θ
The following parameters are determined using the above equations
Parameter Value
I_max 10.03 A
ζ th 7.637 minutes
Rth 0.0989
Cth 77.21
Table 5.9: Thermal Model parameters of induction motor-2
5.4 EFFECT OF OVER-VOLTAGE
The effect of overvoltage on the loss of life is calculated based on the following parameters:
Parameter Value
Ambient Temperature 32.5 0C
Loading 77 %
Acceleration Factor for Humidity 1
Ea (For Class B Insulation) 0.8 eV
K 0.8614* 10-4 eV/K
Table 5.10 : Input Data for over-voltage test
65
The test cycle for overvoltage test is explained in the Figure 5.5. After subjecting the motor to
the 100% voltage temperature was measured till 30 minutes until the motor reached a steady
state. Then the motor was allowed to cool down for 40 min till the temperature reached to
that of the ambient. This was repeated for voltage of 107% and 112%.
Test Cycle- Overvoltage

120
Voltage (in percentage)
100
80
60
40
20
0
0 50 100 150 200
Time (in minutes)
Figure 5.5: Test cycle for overvoltage test
The following steady state temperatures were observed for the abovementioned conditions:
Voltage Steady State Temperature

100% 48.5 0C
107% 52.12 0C
112% 54.82 0C
Table 5.11: Observed temperatures for various voltages
66
56
55
54
53
Measured
Temperature 52
(Degree Celcius)
51
50
49
48
98% 100% 102% 104% 106% 108% 110% 112% 114%
Voltage (Percentage)
Figure 5.6: Measured Temperature v/s Voltage (Percentage) for 2.2kW, 415 V Induction Motor
running at full load
The following temperatures were input to the ageing model
1 1 1 1
( )= ∗ − ∗ − (5.28)
The plot between the Loss of Life and No. of operating hours for various voltages for the
motor is given below.
Estimated Loss of life (%)

Duration Duration
(In (In
hours) Years) V=100% V=107% V=112%
0.5 0 0.000207 0.000212 0.000232
1 0 0.000413 0.000425 0.000464
12 0 0.004959 0.005098 0.005566
24 0 0.009917 0.010195 0.011132
48 0 0.019835 0.020391 0.022263
168 0 0.069421 0.071368 0.077922
8760 1 3.619835 3.721325 4.06308
17520 2 7.239669 7.442651 8.12616
26280 3 10.8595 11.16398 12.18924
35040 4 14.47934 14.8853 16.25232
43800 5 18.09917 18.60663 20.3154
67
52560 6 21.71901 22.32795 24.37848
68320 7.8 28.2314 29.02294 31.68831
70080 8 28.95868 29.7706 32.50464
87600 10 36.19835 37.21325 40.6308
146400 16.7 60.49587 62.19201 67.90353
219000 25 90.49587 93.03314 100
Table 5.12: Estimated loss of life for i) V=100% ii) V=107% ii) V=112%
100
90
80
70
60
50 Voltage
Life Loss (Percentage) =100%
40
30 Voltage
= 107%
20
Voltage
10
= 112%
0
0 50000 100000 150000 200000
Number of operating hours
Figure 5.7: Loss of life v/s No. of operating hours for i) V=100% ii) V=107% ii) V=112%
By the above results, increased voltage gives rise to higher steady state temperature. It is also
seen that in all the 3 cases, the initial trend is similar, indicating that time for which a stress is
applied is an important factor. Thus, if the over-voltage is for a shorter duration, then its
effect on the life loss is lesser. But as the duration increases, the effect on life loss increases
sharply. By increasing the voltage by 7 %, life loss is faster as can be seen by comparing the
graphs for V= 100% and V= 107% . An additional 5% increase to 112 % causes more
deterioration than the initial 7 % increase. It can be also be seen that for V= 112%, 100% life
68
loss is achieved much earlier than for V= 100% and V= 107% where the life loss is 90% and
93% respectively. Also for initial rise of voltage, the ageing is slower and as the voltage is
subsequently increased, the effect is more . Thus, rate of ageing is higher at higher voltages.
5.5 EFFECT OF VOLTAGE UNBALANCE
NEMA (National Electrical Manufacturers Association of USA) standard definition that is
given by equation
Maximum deviation from mean of( V , V , V )

Voltage unbalance = (5.29)
Mean of ( V , V , V )
To study the effect of unbalance voltage on the life of winding, sequence component theory is
used to derive the positive and negative sequence component of voltages as shown below.
1 1 1
2 = 1 (5.30)
3
0 1 1 1
Where, = 1120
V1, V2 and V0 are the positive, negative and zero sequence components of the line voltages
Vab, Vbc and Vca.
For the study of unbalance voltages, a rheostat of 12 Ohms, 8.5 A capacity was put in the
series of one phase and thus a reduced voltage at one of the phase was observed.
69
The schematic diagram of the circuit is shown below.
Figure 5.8: Schematic of voltage-unbalance test
The test cycle for voltage unbalance test is given in Figure 5.9. After subjecting the motor to
the 0% voltage unbalance, temperature was measured till 30 minutes until the motor reached
a steady state temperature value. Then the motor was allowed to cool down for 40 min till the
temperature reached to that of the ambient. This was repeated for voltage unbalance of 0.6% ,
1.77% and 2.78%.
Test Cycle- Voltage Unbalance

3
Voltage Unbalance (in percentage)
2.5
1.5
0.5
0
0 50 100 150 200 250 300
Time (in minutes)
Figure 5.9: Test cycle for voltage-unbalance test
70
The positive sequence, negative sequence voltages and voltage unbalance for various cases
are tabulated below in Table 5.13.
Vab (Volts) Vbc (Volts) Vca (Volts) Voltage Positive Negative

Unbalance (%) Sequence Sequence
Voltage (V1) Voltage
(V2)
4100 4100 4100 0 410 0
4050 411-120 410.6120 0.6 408.86 -1.83
400.70 412-120 412120 1.77 408.23 -2.33
3950 412-120 412120 2.78 406.33 -5.66
Table 5.13: Sequence Voltages for different unbalance conditions
The positive sequence voltage V1 is used to find the positive sequence current I1.
Using the positive sequence component of voltage, the positive sequence circuit is used to
analyse the motor behaviour under voltage unbalance. The positive sequence circuit is shown
below.
Figure 5.10: Positive Sequence Equivalent Circuit
The negative sequence current (I2) is found as:
= 1 + 2 (5.31)
Voltage Rotor Positive Negative Total Positive Negative

Unbalance Speed Sequence Sequence Current (I) Sequence Sequence
(%) Voltage Voltage (A) Current Current
(RPM) (V1) (V2) (I1) (A) (I2) (A)
0.6 1430 408.86 -1.83 4.8 4.68 1.058
1.77 1435 408.23 -2.33 4.8 4.67 1.09
2.78 1441 406.33 -5.66 4.8 4.31 2.098
Table 5.14: Sequence Voltages and Currents for different unbalance conditions
71
The effect of unbalanced voltage on the loss of life is calculated based on the following
parameters:
Parameter Value
Ambient Temperature 31 0C
Acceleration Factor for Humidity 1
Ea (For Class B Insulation) 0.8 eV
K 0.8614* 10-4 eV/K
Table 5.15 : Input Data for unbalance test

The following steady state temperatures were observed for the abovementioned conditions
% Unbalance Voltage Steady State Temperature

0% 50.1 0C
0.6% 52.12 0C
1.77% 55.82 0C
2.78% 60.62 0C
Table 5.16: Observed temperatures for various % voltage unbalance
The plot between Measured Temperature v/s Voltage Unbalance is obtained and is observed
that with increase in voltage unbalance the measured temperature increases.
65
63
61
59
57
Measured
Temperature 55
(Degree Celcius )
53
51
49
47
45
0 0.5 1 1.5 2 2.5 3
Voltage Unbalance (%)
Figure 5.11: Measured Temperature v/s Voltage Unbalance (Percentage) for 2.2kW, 415 V Induction
Motor
72
Temperature v/s Negative Sequence Current
65.00
60.00
Steady State 55.00

Temperature ( 0C)
50.00
45.00
40.00
0.00 0.50 1.00 1.50 2.00 2.50
Negative Sequence Current (A)
Figure 5.12: Measured Temperature v/s Negative sequence current for 2.2kW, 415 V Induction
Motor
The following temperatures were input to the ageing model (Eqn. 5.28)
1 1 1 1
( )= ∗ − ∗ − (5.28)
From the plot between Loss of life v/s No. of operating hours for Voltage Unbalance, it is
seen that when the Voltage unbalance is more, the loss of life is faster.
Estimated Life Loss (%)

Duration
Duration (In (In
hours) Years) 0% 0.60% 1.77% 2.78%
0.5 0 0.000204 0.000221 0.000247 0.000297
1 0 0.000408 0.000441 0.000495 0.000594
12 0 0.004902 0.005298 0.005935 0.007123
24 0 0.009804 0.010595 0.011869 0.014247
48 0 0.019608 0.02119 0.023739 0.028493
168 0 0.068627 0.074166 0.083086 0.099727
8760 1 3.578431 3.867208 4.332344 5.200047
17520 2 7.156863 7.734416 8.664688 10.40009
26280 3 10.73529 11.60162 12.99703 15.60014
73
35040 4 14.31373 15.46883 17.32938 20.80019
43800 5 17.89216 19.33604 21.66172 26.00024
52560 6 21.47059 23.20325 25.99407 31.20028
68320 7.8 27.9085 30.16069 33.78833 40.55562
70080 8 28.62745 30.93767 34.65875 41.60038
87600 10 35.78431 38.67208 43.32344 52.00047
146400 16.7 59.80392 64.63005 72.40356 86.9049
219000 25 89.46078 96.6802 100 100
Table 5.17: Loss of life v/s No. of operating hours for Voltage Unbalance i) 0% ii) 0.6%
iii) 1.77% iv) 2.78 %
100
90
80
70
0%
60
Life Loss 0.60%
50
(Percentage)
1.77%
40
2.78%
30
20
10
0
0 50000 100000 150000 200000
No. of operating hours
Figure 5.13: Loss of life v/s No. of operating hours for Voltage Unbalance i) 0% ii) 0.6%
iii) 1.77% iv) 2.78 %
By the above results, increased unbalance gives rise to higher steady state temperature. It is
also seen that in all the 4 cases, the initial trend is similar, indicating that time for which a
stress is applied is an important factor. Thus, if the unbalance is for a shorter duration, then its
effect on the life loss is lesser. But as the duration increases, the effect on life loss increases
sharply. By increasing the unbalance by 0.6 %, life loss is faster as can be seen by comparing
the graphs for Unbalance= 0% and Unbalance= 0.6% . An additional 1.1% increase to 2.77%
74
causes more deterioration than the initial 0.6 % increase. It can be also be seen that for
unbalance = 2.78%, 100% life loss is achieved much earlier than for unbalance= 0% and
Unbalance= 0.7 % where the life loss is 89 % and 96 % respectively. Also for increase in the
value of unbalance i.e. from 0% to 0.6 % , the ageing is slower and as the unbalance is
subsequently increased, the effect is more . Thus, rate of ageing is higher at higher values of
unbalance.
75
CHAPTER 6
CASE STUDY
COOLANT PUMP MOTOR – PLANT DATA ANALYSIS
6.1 GENERAL
The following is a Case Study of the plant data of the Coolant Pump Motor of a typical
research reactor. The reactor uses heavy water (deuterium) as a moderator and coolant. This
section brings forward a way to estimate the remaining life of the coolant pump motor using
plant data.
Power Supply System
The electrical power to various loads is supplied from the 22kV sub-station located in service
building ground floor. The power supply is fed by two cables running through independent
routes from 110kV sub-station. Redundant 3.3kV buses are used for feeding the power to
loads of Coolant Pump Motors. There are two number of 10MVA transformers supplying
power to 3.3kV buses. All other loads including AC motors are supplied through 4 nos. of 2
MVA transformers whereas 2 nos. of transformers will be capable of meeting the full load
requirements.
Three nos. of 500kVA Diesel Generators are kept in poised state to cater to essential and
safety related load requirements. These DGs sets start automatically in case of grid failure
and feed the loads through 3 independent buses. A tie-breaker closure is provided to feed
loads in case of failure of one DG set.
Presently uninterrupted 415 V AC is supplied by 2 nos. of 150kVA Motor Alternator (MA)
sets and 2 nos. of 20kVA DC/AC inverters.
76
Uninterrupted Class I DC power is supplied to 250V and 48V buses through dedicated
battery banks and Automatic Constant Voltage Rectifiers (ACVR).
6.2 COOLANT PUMP MOTOR
The reactor has 3 sets of Coolant Pumps which provides heat removal from the reactor core
by coolant circulation in the closed circuit as well as it provides heat transfer to the secondary
circuit. 2 sets of spare Coolant Pump Motor are also provided which can be used if any
maintenance activity is being carried out in any of the Coolant Pump Motors. The technical
specification of Coolant Pump Motor of a typical research reactor is given below:
MOTOR Voltage AMP RPM P.F. IP Ambient DUTY INSULATION

(kV) Temp CLASS
(0C)
Motor-1 3.3 113 1492 0.9 55 50 S1 F
Motor-2 3.3 113 1492 0.9 55 50 S1 F
Motor-0 3.3 113 1492 0.9 55 50 S1 F
Motor-4 3.3 113 1495 0.9 55 S1

Motor-3 3.3 113 1495 0.9 55 S1
Table 6.1: Motor Data
Maintenance of Coolant Pump Motors
The following parameters of the motor are periodically monitored:
 Temperature (Measured at 8 Points including Driving End and Non-Driving End side)
 Current
 Voltage at the bus (Line to Line voltage between R-Y, Y-B and R-B phases)
 Vibration
 No-Load current (after each stopping/shutdown)
 Insulation Resistance
The maintenance activities of the Coolant Pump Motor include:
 Greasing at Driving End and Non-Driving End side

 Bearing replacement
 Stud Replacement
 Cable Lead Re-tapping/Insulation
77
Monitoring of Temperature in Coolant Pump Motors
Temperature is monitored in Coolant Pump Motors using 8 RTDs. Two of the RTDs are
located at the Driving End and Non-Driving End of the motor. Two RTDs are located in each
phase of the stator winding. A typical scheme of RTD arrangement in the stator winding is
shown in figure 6.1 . Temperature is monitored in every 4 hours interval.
Figure 6.1: RTD position in stator winding of Coolant Pump Motor
78
Coolant Pump Motor –Winding Temperature Data
Winding Temperature data were collected for the period 1/9/2011 to 30/6/2015.
100
98
96
94
92
90
88
86
84
82 Motor-3 Winding Temeprature

80
2/26/2011 9/14/2011 4/1/2012 10/18/2012 5/6/2013 11/22/2013 6/10/2014 12/27/2014 7/15/2015 1/31/2016
Figure 6.2 : Coolant Pump Motor -3 Average Winding Temperature= 94.98 0C
100
98
96
94
92
90
88
86
84
Motor-1 Winding Temperature
82
80
2/26/2011 9/14/2011 4/1/2012 10/18/2012 5/6/2013 11/22/2013 6/10/2014 12/27/2014 7/15/2015 1/31/2016
79
90
88
86
84
82
80
78
76 Motor-0 Winding Temperature
74
72
70
2/26/2011 9/14/2011 4/1/2012 10/18/2012 5/6/2013 11/22/2013 6/10/2014 12/27/2014 7/15/2015 1/31/2016
6.3 DATA ANALYSIS
Coolant Pump motors provide heat removal from the reactor core by coolant circulation in
the closed circuit. Thus from the safety point of view, they are quite crucial and important.
The following tables contains a list of events/activities pertaining to the Coolant Pump Motor,
the information is obtained from the Service Card of the motors.
The data is further used for various analysis by using certain well-established models with
some assumptions.
The major events/activities in the lifetime of Coolant Pump Motor 1, 2 and 3 are given in the
Annexure A.
80
Causes of Coolant Pump Motor mal-operation:
Since Coolant Pump Motor are safety related equipment, the high reliability of the motors
shall be maintained. This is achieved by a robust preventive maintenance plan where various
parameters of the motors are recorded and if any of the parameters exceed the normal limits,
that motor is replaced with a spare motor.
An attempt has been made to analyze the various causes of motor mal-operation which is
explained below.
It is observed that Coolant Pump Motors were replaced with other/ Coolant Pump Motors in
the due course due to the following reasons:
1. High Vibration
2. Winding Failure
3. Abnormal Noise
4. High bearing temperature
5. High winding temperature
6. Overhauling
7. High Winding Temperature
8. Electrical Trips
9. Contactor Damage
This can be shown in the form of a pie-chart as shown:
81
Figure 6.5: Causes of Coolant Pump Motor mal-operation
Thus it can be seen that abnormal noise and high bearing temperature constitutes a major
cause of mal-operation of Coolant Pump Motors.
The various reasons of abnormal noise can be :
1. Mis-alignment between pump and motor
2. Unbalance
3. Frame resonance
4. Excessive bearing clearance
5. Defective rotor bars or end-rings
The various reasons of high bearing temperature are:
1. Over-greasing
2. Under-greasing
3. Bearing damage
82
6.4 MEAN TIME TO FAILURE (MTTF) FOR WINDING FAILURE [17]
Reliability is defined as the probability that a component or system will continue to perform
its intended function under stated operating conditions over a specified period of time.
The reliability level is derived by monitoring the functional stability of a number of
representative subjects operating under elevated stress conditions resulting in a statistical
prediction of reliability.
The most common method to establish a reliability level is to calculate the probability of
failure or Rate of Failure (λ).
Failure Rate (λ) in this model is calculated by dividing the total number of failures or rejects
by the cumulative time of operation.
/
= (6.1)
( )
= . ∗ ℎ ∗ (6.2)
To derive a more statistically accurate calculation for Failure Rate (λ), the number of rejects
(r) is replaced with the probability function using Chi-squared (X2).
( ′, )
. ~
2
X2/2 (Chi-squared/2) is the probability estimation for the number of failures or rejects
′ (alpha), confidence level (CL) or probability, is the applicable percent area under the X2
probability distribution curve; reliability calculations use α= 0.6 (or 60%). Note: The total
area under the X2 curve is always 1, so ≤ 1 (or 100%).
83
(nu), degrees of freedom (DF), determines the shape of the X2 curve; reliability
calculations use = 2r + 2 where r = number of failures or rejects
Figure 6.6: Chi-square distribution
Thus, failure rate can be calculated as using (6.1):
2
( ′, )
= (6.3)
2∗( )
MTTF (Mean Time to Failure) is standard industry value which provides the average time to
failure of Non-repairable Items. For items with long life expectancies, it is often a more
useful to report MTTF in years rather than hours.
1
= (6.4)
Motor S. No. Total duration of operation

Motor-3 20 Years 3 Months 20 Days
Motor-1 20 Years 1 Months 11 Days
Motor-0 10 Years 7Months 25 Days
Table 6.2: Coolant Pump Motor motor operating duration data
84
Total Duration of Operation = 51.083 years
Acceleration Factor= 1
Confidence Level( ′) = 0.9
No. of rejects= 1
Degrees of freedom ( )= 2r + 2 = 4
( , ) ( . , )
No. of rejects~ = = 3.89
.
= .
= 0.07615
MTTF= 1/ = 13.13 years
6.5 WEIBULL DISTRIBUTION OF WINDING FAILURE [18]
It has been found out by various studies that the most effective way of handling motor failure
data is by means of a Weibull Distribution Curve.
Weibull Cumulative Distribution Function is described as
( ) =1− ( / )
(6.5)
Where α is the characteristic life and β is the Weibull Slope.
It has been observed that winding failure below ambient temperature 125 0C have a mean
slope approaching 1 indicating random type failure with a constant failure rate.
Thus for motor winding failure,
β=1
α can be found using the data collected as explained below.
85
Motor Life in terms of Mean Time to Failure (MTTF) can be estimated at the 50 percentile
failure condition.
Thus, the motor life equation is
ln 1 − ( ) = − ( / ) (6.6)
Where F(t) is the Weibull Cumulative Distribution Function
is the characteristic life
is the Weibull Slope
No. of failures= 1
So, ( ) = = 0.5
ln 1 − ( ) = −0.693
Using, MTTF= 13.13 years and =1
= .
The cumulative distribution function (CDF), or just distribution function, describes the
probability that a real-valued random variable X with a given probability distribution will be
found to have a value less than or equal to x. In the case of a continuous distribution, it gives
the area under the probability density function from minus infinity to x.
The probability density function (PDF), or density of a continuous random variable, is a
function that describes the relative likelihood for this random variable to take on a given
value. The probability of the random variable falling within a particular range of values is
given by the integral of this variable’s density over that range—that is, it is given by the area
86
under the density function but above the horizontal axis and between the lowest and greatest
values of the range. The probability density function is nonnegative everywhere, and its
integral over the entire space is equal to one.
Figure 6.7 : Cumulative Distribution Function for Weibull Distribution for

and
Figure 6.8: Probability Density Function for Weibull distribution and
87
6.6 CALCULATION OF ACTIVATION ENERGY FOR HUMIDITY (Eb)
Based on the plant data obtained from the service cards of the Coolant Pump Motors, the
activation energy for humidity is calculated using the Eyring Equation as mentioned
explained below:
1 1 1 1
( )= ∗ − ∗ − (6.7)
Where, tr time at temperature Tr
RHr is relative humidity at temperature Tr
RHi is relative humidity at temperature Ti
(Activation Energy for temperature)= 1.05 eV for Class F insulation
(Activation Energy for humidity)= Unknown
K =0.8617* 10−4 eV/K
From the above calculations/results,
MTTF= 13.13 years
Thus, (Activation Energy for humidity) can be calculated as follows:
Parameter Symbol Value

MTTF tr 13.13 years
Design life of insulation at 155 0C and 90 % ti 2.3 years
RH
Average winding temperature of the Coolant Tr 91.83 0C
Pump Motor
Maximum permissible temperature for Class Ti 155 0C
F insulation
Average Relative Humidity RHr 95 %
Design value of Relative Humidity RHi 90 %
Activation Energy for temperature Ea 1.05 eV
Table 6.3: Input Data for Eb calculation
88
Using the Eyring Equation mentioned above , the calculated value is:
(Activation Energy for humidity)= 0.3853 eV
6.7 REMAINING LIFE OF INSULATION SYSTEM
It is assumed that the maximum temperature of the windings follow the same trend
throughout the entire service life of the motor.
Using the Eyring Relation and accounting for number of starts/stops, the remaining life for
each Coolant Pump Motor is calculated as using (6.8):
1 1 1 1
( )= ∗ − ∗ − (6.8)
Effect of starts/stops
Accelerating time for Coolant Pump Motor = 12-15 seconds
Average winding temperature after starting of Coolant Pump Motor (Measured from RTD) =
95 0C
. /
Life lost at each start= 2 ∗ (6.9)
/
Thus, life lost at each start= 2 ∗ 15= 10861 seconds= 3.01 hours
Thus, Average life lost during each starting = 3hrs
89
S.No. Total No. of Life lost during Avg. Relative
duration start/st starting/stopping Temp of Humidity
of ops (years) motor (C)
operation
(years)
Motor-3 20.301 730 0.25 94.98 95%

Motor-1 20.11 723 0.247 93.27 95%
Motor-0 10.64 384 0.131 85.26 95%
Table 6.4: Input data for remaining life calculation of Coolant Pump Motors
Life (years) for Coolant Pump Motors
Motor-3 : 20.5 years
Motor-1: 23.56 years
Motor-0 (Replaced Winding): 26.71 years (For temperature = 95 0C)
: 50.6 years (For temperature = 85.26 0C)
Winding for Motor-0 was replaced and rewinding was done on 12/6/2000 after a Line-
Ground fault was observed and the Insulation Resistance (IR) of the winding was found to be
zero. Being re-wound, the winding is alike a new one and thus it has not aged as much as the
other windings.
It can also be observed that for Motor-0 the temperatures are relatively lower than that for
Motor-3 and Motor-1. This is mainly due to low flow of the coolant for that pump. Thus, to
estimate the remaining life for Motor-0 two cases can be considered, one with temperature
equivalent to that observed for Motor-3 and Motor-1 i.e. 95 0C and the other case being when
the temperature is equivalent to a low coolant flow condition i.e. 85.26 0C . The estimated
values of the remaining life for the two cases are mentioned below.
90
CHAPTER 7
CONCLUSION
Induction motors, being an important part of any industry are prone to failures despite being
robust. Various factors can lead to its mal-operation and efforts shall be made to minimize its
mal-operation/failures. Motor lifetime prediction with adequate maintenance and surveillance
is a function of localized environmental conditions and motor design and operating
conditions. Lifetime predictions fall in the ambit of Condition Based Maintenance in which
the health of the motor is assessed continuously monitoring various parameters pertaining to
an induction motor. Accurate results can result in optimizing the performance of motor while
minimizing the downtime. This forms a part of Ageing Management in which a conscious
decision can be taken for replacement/refurbishment of a motor.
It has been statistically seen by various surveys that winding failure accounts forms a major
portion of induction motor failures. Thus it becomes pivotal that the concept of winding
failure is understood fully and efforts shall be made to minimize any mal-operation due to
winding failure. This work has attempted to analyse the induction motor winding failure
through the lens of an integrated approach which combines the electrical, thermal and ageing
aspect of a motor winding and hence predict its remaining life/loss of life. This is a new
domain as the it involves an amalgamation of the electrical, thermal and ageing aspect of a
motor to determine its life.
This model was used to quantify the loss of life of a 2.3 kW, 415 V induction motor under
different operating anomalies e.g. overvoltage, unbalance, etc. In the present work , the
induction motor circuit derived using No load and Blocked Rotor test gives the winding
losses which goes as an input to the thermal model to give the winding temperature . This
winding temperature goes as an input to the ageing model, which basically uses temperature
91
and humidity as the stresses (Eyring Model) to estimate the remaining life. The effect of
various starts/stops on the remaining life is also seen.
It is seen that and increased voltage has a detrimental effect on the remaining life. One
important factor is the time for which the motor is under the aggravated stress. For shorter
durations, it is observed that there is negligible effect on the life loss but as the duration
increases, the effect is severe , i.e. even a 12 % sustained over-voltage can cause a 11%
reduction in the life of motor. 100% life loss is obtained at a much shorter duration for V=
112% when the life loss for V= 100% and V= 107% is 90% and 93% respectively. An initial
increase of voltage from 100% to 107% causes lesser deterioration than the increase from
107% to 112%. Thus , it can be concluded that rate of ageing or deterioration increases with
increase in the value of the applied stress.
It is seen that and unbalanced voltages also has a negative effect on the remaining life. In this
study, unbalance in one of the phases of the 3 phase induction motor was created by adding a
resistor in series with the phase winding. By changing the value of the resistor, reduced
voltage was developed in one of the phases and thus unbalance of 0.6%, 1.77% and 2.78%
were derived using the resistor. Then for all these cases, the stator winding temperature was
found and the curves for life loss with no. of operating hours were drawn. It was observed
that, one important factor in the estimation of life loss due to unbalance is the time for which
the motor is under the aggravated stress. For shorter durations, it is observed that there is
negligible effect on the life loss but as the duration increases, the effect is severe, i.e. even a
2.78 % sustained unbalance can cause a 12 % reduction in the life of motor. 100% life loss is
obtained at a much shorter duration for Unbalance= 2.78% when the life loss for Unbalance=
0% and Unbalance=0.6 % is 89% and 96% respectively. An initial increase of unbalance
from 0% to 0.6% causes lesser deterioration than the increase from 0.6% to 1.77% and from
92
1.77% to 2.78% Thus , it can be concluded that rate of ageing or deterioration increases with
increase in the value of the applied stress.
A Fuzzy Logic Approach to determine the remaining useful life (RUL) is also studied using
Simulation Software. Here the stator winding current are an input to the controller which
works on a set of 14 if-then rules and based on the values of the currents of the three phases,
the RUL is estimated.
Further, a Weibull model of the induction motor is developed using the data of Coolant Pump
Motors of a typical Research Reactor taken over a period of 30 years using Chi-Square
Analysis. Mean Time to Failure (MTTF) of the motor is calculated using the data such as
number of failures and operating duration and considering generally accepted values of
confidence interval. Using the MTTF, the characteristic life for insulation failure is calculated
and thus the Weibull Plot for insulation failure is obtained. This plot is further used to plot the
Probability Distribution Function and Cumulative Distribution Function. Eyring Equation is
used to find the remaining life of the motor and an input parameter to this equation i.e. the
activation energy for humidity is calculated using the above data. After obtaining the
parameter, the Eyring Equation is used in the Insulation Ageing Model to estimate the
remaining life of motor. It is seen that the remaining life is in the range of the generally
observed/accepted values.
93
FUTURE SCOPE OF WORK
The purpose of this study was to estimate the remaining life of motor using an integrated
approach. The effect of various operating anamolies e.g. overvoltage and unbalance were
seen. This study can be extended to analyze the effect of over-loading, surges, etc. on the life.
Secondly, only temperature and humidity along with the no. of starts/stops are used as the
stressors. Other stressors like pressure, radiation, etc. can be added by using a multi-stressor
approach. Thirdly, the temperature model considered is a single R-C Circuit. Instead a more
detailed and elaborated thermal model can be used which gives a more accurate results.
Further, the activation energy for humidity as well as insulation class shall be found while
considering the geometry, environment, material, etc. which gives a more accurate result of
the remaining life. The activation energy for various classes of insulation can also be
updated/revised which will further refine the results.
A canary based approach is envisaged to be taken up to better comprehend the winding
failure in an induction motor. A canary is a product that provides early warning of a negative
condition, similar to a real canary bird in coal mines. In engineering applications, a canary
device fails by the same failure mechanism as the system that is being monitored; only
earlier, thus providing a prognostic indication of failure.
94
REFERENCES
[1] Xiao-Sheng Si, Wenbin Wang, Chang-Hua Hu, Dong-Hua Zhou” Remaining useful
life estimation – A review on the statistical data driven” European Journal of
Operational Research 213 (2011) ,1-14
[2] Vfflaran M. and Subudhi M “Aging Assessment of Large Electric Motors in Nuclear
Power Plants”, Brookhaven National Laboratory, March 1996.
[3] Gupta J.B., Electrical Machines-Induction Motor- No-Load Test and Blocked Rotor
Test, Page 399- 406, Edition 2007.
[4] Modak Amit and Inamdar H.P. “Computer-aided Simulation Model of Stator
Ground-wall Insulation of Induction Motor based on Current Park’s Vector
Approach” International Journal of Computer Applications, Number 8 - Article 5,
2010.
[5] Zeraoulia M., Mamoune A., Mangel H. and Benbouzid M.E.H. “A Simple Fuzzy
Logic Approach for Induction Motors Stator Condition Monitoring”. Journal of
Electrical Systems 1-1 (2005): 15-25.
[6] Vicente Pedro, Rodríguez Jover and Arkkio Antero “Detection of Stator Winding
Fault In Induction Motor Using Fuzzy Logic” Applied Soft Computing, Vol. 8, No.
2, pp. 1112-1120.
[7] Saravana Kumar R., Kumar K.Vinoth & Ray K.K. “Fuzzy Logic based fault
detection in induction machines using Lab view” IJCSNS International Journal of
Computer Science and Network Security, Vol.9, No.9, September 2009 .
[8] Uyar O. and Cunkas M. “Design of Fuzzy Logic Based Motor Protection System”
6th International Advanced Technologies Symposium (IATS’11), 16-18 May 2011,
Elazığ, Turkey.
[9] Pillay Pragasen and Manyage Marubini “Loss of Life in Induction Machines
Operating with Unbalanced Supplies” IEEE Transactions On Energy Conversion,
Vol. 21, No. 4, December 2006.
[10] Anbarasu E. and Karthikeyan M. “Modeling of Induction Motor and Fault Analysis”
International Journal of Engineering Science and Innovation Technology (IJESIT),
Volume 2, Issue 4, July 2013.
[11] Zhi Gao “Sensorless Stator Winding Temperature Estimation for Induction
Machines” PhD Dissertation, Georgia Institute of Technology, December 2006.
[12] Policarpo Jose, de Abreu G. and Eigeles Emanuel Alexander,” Induction Motor
Thermal Aging Caused by Voltage Distortion and Imbalance: Loss of Useful Life
and Its Estimated Cost”, IEEE Transactions on Industry Applications, Vol. 38, No.
1, January/February 2002.
95
[13] Brancato Emanuel L. “Estimation of Lifetime Expectancies of Motors” IEEE
Electrical Insulation Magazine May/June 1992.
[14] Siddique Arafat, Yadava G.S. and Singh Bhim “Effects of Voltage Unbalance on
Induction Motors” Conference Record of the 2004 IEEE International Symposium
on Elecnical Insulation, Indianapolis, IN USA, 19-22 September 2004.
[15] Mishra Sandesh Kumar, Tiwari Pratibha & Mishra Navneet Kumar “Comparative
Analysis Of Three-Phase Induction Motor Fed From Balanced And Unbalanced
Power Supply” International Journal of Electrical and Electronics Engineering
Research (IJEEER) ISSN(P): 2250-155X; ISSN(E): 2278 943XVol . 4, Issue 2, Apr
2014, 347-354.
[16] IEEE Guide 1205 “IEEE Guide for Assessing, Monitoring, and Mitigating Aging
Effects on Class 1E Equipment Used in Nuclear Power Generating Stations” 2014.
[17] Ellerman Paul “Calculating Reliability using FIT and MTTF: Arrhenius HTOL
Model” MircoNote 1002, September, 1992.
[18] NP-3887 “Life Expectancy of Motors in Mild Nuclear Plants Environments”, EPRI
Report, February 1985.
[19] NEMA MG-1-25 Motors and Generators –Classification of Insulation System.
[20] Chafai Mahfoud, Refouf Larbi , Bentarzi Hamid “Reliability Assessment and
Improvement of Large Power Induction Motor Winding Insulation Protection
System Using Predictive Analysis” WSEAS Transactions On Circuits And Systems,
Issue 4, Volume 7, April 2008.
96
Annexure A
Major maintenance activities/events in the lifetime of Coolant Pump Motors
(As discussed in Chapter 6)
Motor-0 DATA
Position of motor Maintenance Activity Total time in that position

at pump (Y: Years M: Months D:
days)
Pump 1 Motor was test run on No-load for 1 hr. 3 Y 11 M 7 D

Direction of rotation was found reverse. No-
load current found 46.5A, 46A,46.5A which
got reduced to 44,40,42 (at 3.45 kV) after 1 hr
running. Some abnormal noise and current
hunting noticed at no load which disappeared
when motor ran on load
Pump 1 Motor tripped on fault while starting the pump.
Motor cable and box was opened and found
fully carbonized. Cable lug of phase B found
cut. Dismantled the cable and motor terminal
box. Through cleaning done. Newly fabricated
stud of higher size brazed with motor leads. All
the three insulators replaced with
reconditioned-varnished insulator
Pump 1 Motor started from control room at 17:44 and
pump tripped immediately. No flag registered.
Again pump started at 17:53. Same tripped on
I0 and I2. Observed some smoke from star side
terminal. IR value found to be ZERO of Motor-
0
Pump 1 Spare Motor installed in COOLANT PUMP
MOTOR#1 position (18002901) as earlier
COOLANT PUMP MOTOR# 1 IR value was
found ZERO (18002900)
Pump 2 Spare motor installed in COOLANT PUMP 6 Y 8 M 6 D
MOTOR 2 position after rewinding and bearing
replaced. No load current=43 A and rated
current= 113 A. Slight abnormal noise
observed after some interval. Motor put off
from sub-station. IR value=20 M ohms
Pump 2 IR value=20 M ohms

97
Pump 2 IR value more than 100 M ohms
Pump 2 Current found 111 A at 9:30. Current came
down to 104 A. Frequency = 50.5 Hz and
voltage=20.6 kV. During this load current of all
3.3kV motors increased
Pump 2 IR=20 M ohms

Pump 2 IR=200M ohms
Pump 2 IR=200M ohms
Pump 2 Found B phase stud sheared off. Replaced the
stud. Motor lead wire and cable connection
done. IR value before and after cable
connection 150 M ohms. No load test=38 A
Pump 2 NDE side grease tray insulation provided and

put back in position as insulation part found
broken
Pump 2 Motor tripped twice due to Ith as problem was
with the CTM relay.
Pump 2 NDE side grease tray insulation provided and
put back in position as insulation part found
broken
Pump 2 IR=500 M ohms
Pump 2 IR> 20 M ohms
Pump 2 NDE side bearing temp rise to 77. Blackish
grease had come out. Further instrument
maintenance tightened the RTD terminal and
temp came down to 61 C
Pump 2 DE side 'OFF NORMAL ALARM' registered

on control room at 12:15 hrs. Temp 78.1 and
cleared at 13:30 hrs Temp 74.1
Pump 2 IR=150 M ohms

Pump 2 IR=20 M ohms
Pump 2 IR=20 M ohms
Pump 2 d-2 contactor replaced
Pump 2 At 23:15 hrs, DE side bearing temp suddenly
raised to 113 C. Cooling provided and temp
came down to 78 C at 00:30 hrs
98
Pump 2 Shift Engg. Informed that some abnormal noise
observed from NDE side. 1 gram of grease
injected at NDE side as required by shift engg.
NDE temp=56 C. Motor disconnected due to
high noise.
Pump 3
Pump 3 Sttep rise of temp was noted for NDE bearing. 9Y 5 M 19 D
Injection of 5 grams of grease was causing 5 C
temp rise.
Pump 3 Bearing Temp at D.E.=67 C
Bearing Temp at N. D.E.=57 C
Pump 3
Pump 3 Bearing Temp at D.E.=70 C
Pump 3 Bearing Temp at D.E.=65
Pump 3 I.R.= 200 M ohms

Pump 3 I.R.= 200 M ohms
Pump 3 Motor cable end box got burst. Found studs
damaged. Replaced all three studs.
Pump 3 Replaced R& B phase studs. Also replace all

three insulators as they were found carbonized
and damaged.
Pump 3 I.R=50 M ohms

Pump 3 COOLANT PUMP MOTORM # 3 motor side
cable end box inspected. Found fully
carbonized. R phase cable lead found damaged.
I.R. value wrt earth was zero. Cable
disconnected and cable end box dismantled and
fully cleaned. Thereafter I.R. improved to 40
Mohms. Also bushing found damaged.
Replaced all the three bushing. After that IR
improved to 100 M ohms. Replaced Y phase
motor stud.
Pump 3 Motor tripped on fault. I.R wrt found zero
99
Pump 3 IR=2000 M ohms with 2.5 kV megger. No-load 4 Y 11 M 21 D
current=44/43/40 A
Pump 3 Noload current=43/42/41 A. Motor started

again. Noload current=43/42/41 A
Pump 3 IR=200 Mohms

Pump 3 IR= 100 Mohms. No-load current=43/39/41 A.
Coast down time noticed 12 minutes
Pump 3 Noload current=43/41/41 A. Motor started

again. Noload current=43/42/42 A
Pump 3 IR=70 Mohms

Pump 3 No load current=43.1,41.26,39.46
Pump 3 IR=1000 Mohms
Pump 3 DE side temp increase upto 106 C and was
reduced to 62 C
Table A.1: Maintenance Data card for Motor-0
Motor- 1 DATA
Position of motor Maintenance Activity Total time in that

at pump position (Y: Years M:
Months D: days)
Pump 1 No load current= 44 A IR=500 M ohms 11 M 13 D
Pump 1 Disconnected the motor cable/space
heater/earthing as abnormal noise was observed
from NDE side of motor Motor-1. Spare motor
18802902 was installed in COOLANT PUMP
MOTOR# 1 position
Pump 1 Bearing replacement No load current= 40 A IR 8 Y 10 M 8 D
value= Infinity
Pump 1 During checking of terminal tightness 'Y' phase
stud get sheaved off due to damaged thread.
Replaced Y phase stud with new one. No-load
current = 42/43/42 A
Pump 1 NDE side grease pipe was found damaged
Pump 1 T.O.T. replaced
Pump 1 D.E. side temperature raised from 70 to 95
Pump 1 N.D.E. side temperature gone upto 86 from 59
Pump 1 Y phase stud bolt found loose and same was
tightened. Insulation provided for R phase
cable lead
100
Pump 1 Checked radial bearing alarm as it was
registering in control room and local control
panel. Corrected wiring in PILP.
Pump 1 DE side temperature up to 93 and come down
to 81 at 15:00
Pump 1 DE side temperature up to 111. Alarm
registered in LCP
Pump 1 DE side temp shot from 68 to 112 after
injecting 2 grams of grease. Alarm registered
and cleared at 78
Pump 1 After bearing replacement, No-load test was
done. During no-load test, smoke was noticed
Pump 1 Y phase cable lead re-tapping done.
Pump 1 d-2 contactor replaced
Pump 1 DE side temp upto 112 from 70 at 15:00
Pump 1 DE side temp upto 98 from 80
Pump 1 Radial bearing alarm contact jumper removed
and termination done. Radial bearing module
was malfunctioning
Pump 1 Replaced NDE side grease pipe
Pump 1 While stopping the pump from C/Room
COOLANT PUMP MOTOR # 1 electrical fault
alarm registered and E-45 relay actuated and
the relay flag also found down
Pump 2 This motor was installed on COOLANT PUMP 4 Y 9 M 24 D
MOTOR 3 position on 7/7/10 . But RTD
reading of position 6 was increasing from 67 to
200 within 20 min. So motor cable
disconnected. Y phase motor stud thread found
damaged and replaced with new one. IR= 200
M ohms. No load current=44 A. From 15:40 to
18:10 Position 1: 50.9 to 76.1, Pos 2: 50 to
74.6, Pos 3: 49.4 to 72.7, Pos 4: 50.5 to 73.2,
Pos 5 51.3 to 75.4, Pos 6: 2.3, Pos 7: 61.1 to
69.7, Pos 8: 41.5 to 44.1
Pump 2 IR= 200 M ohms
Pump 2 Motor stopped from Contro Room to replace

thrust bearing/radial bearing and mechanical oil
seal. No load test: 43.10, 42.01 and 43.09
Pump 2 Current found 107.14 A and 103 A and motor
stopped on 14/2/14 to inspect mechanical seal
Pump 2 Motor started/stopped twice. Motor started on
load and current=101 A. Again motor stopped.
Motor started again and current found to be 102
A
Pump 2 Greasing
Pump 2 DE side temp rise upto 80
Pump 2 Greasing done at DE side-one gram at 15:46
hrs. Tem gone upto 83.4 C . At 19:30 hrs temp
gone upto 103.2
101
Pump 2 DE side temp 80
Pump 2 Disconnected motor for overhauling
Pump 2 Durung greasing, DE side temp raised upto 108
C
Pump 3 A2 and C2: 0.11 Ohms 3 Y 4 M 26 D
A2 and B2: 0.11 Ohms
B2 and C2: 0.11 Ohms
IR with 500 V Megger: 175 Ohms
Pump 3 No-load current=45 A. IR value: 80 M ohms

Pump 3 No load current found 38 A. Motor stopped
after 2 hrs
Pump 3 Motor started . No load current=40 A
Pump 3 Motor was tripped by shift operation as smoke

was noticed from DE bearing. IR value afer
disconnecting the power cable was 500 M
ohms
Table A.2: Maintenance Data card for Motor-1
Motor-2 DATA
Position of Maintenance Activity Total time in that

motor at pump position (Y: Years M:
Months D: days)
Pump 1 IR value found to be 400 M ohms and no-load 2 M 19 D
current as 45 amp
Pump 1 As required by mechanical maintenance, the
power cable/earthing (18802902). Some
abnormal sound observed from NDE side
Pump 1 Motor Installed after bearing replacement. No- 1 Y 9 M
load current= 42/42/41
Greasing done
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Greasing
102
Pump 1 High Vibration
Pump 2 11 Y 1 M 11 D
Pump 2 IR value= 200 M ohms
Pump 2 R phase stub damaged
Pump 2 Abnormal noise from motor NDE bearing.
Current and temp found normal . Motor
disconnected
Pump 2 Cyclic noise observed. Noise was heard 2-3 2 Y 11 M 17 D
min after starting the motor on no load. Noise
was heard coming and going after irregular
timing. Vibration reading were slightly more
than normal values. On load condition noise
was found high

Pump 2 B phase cable lead insulation was burnt. Re-
tapping done for all three cable leads. IR
value=300 M ohms
Pump 2 Power cable connected after all the three studs
replaced.
Pump 2 Flexible Copper conductor earthing provided

Pump 2 Replaced cover plate bolts as its thread had
damaged
Motor disconnected for bearing replacement .
Pump 3 IR value=20 Mohms. R phase cable lead 4 M

insulation provided. No-load current=38 A.
Some clipping/metallic noise noticed during
no-load operation.
Pump 3 R phase cable lead insulation provided.
Pump 3 Motor disconnected as abnormal noise was

noticed.
Table A.3 Maintenance Data card for Motor-2
103
Motor-3 DATA
Position of motor Maintenance Activity Total time in that

at pump position (Y: Years M:
Months D: days)
Pump 1 Motor replaced due to High Vibration 8Y 3 M 16 D
Pump 1 No load test 41/41/38 A. Greasing done at no 3 Y 9 M 22 D
load. Motor stopped at 22:45. Max Temp: 95
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Greasing
Pump 1 COOLANT PUMP MOTOR 1 motor not
starting as starting circuit ladder is not getting
completed due to K-2 key contact
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Greasing
Pump 1 Removal of earthing as required by mech.
maintenance
Pump 2 COOLANT PUMP MOTOR motor shifted 10 M
after overhaul and bearing replacement.
Abnormal noise noticed during initial start. IR
value=1000 M ohms

Vibration didn't improve even after injecting
small amount of grease . Motor disconnected
for bearing replacement and overhaul.
Pump 2 Installed after Bearing replacement .No load 3 Y 6 M 18 D

current=43 A. DE side temp rose from 34 to 83
and NDE side 35 to 68. COOLANT PUMP
MOTOR 2 OFF NORMAL alarm registered.
Motor stopped at 21:40 hrs. ISI section took
vibration reading on no load and after the
clearance motor was coupled
Pump 2 16/10/08 3 core 150 sq. mm

FRLS PVC cable laid
from motor to C.B. No-
104
load current=44 A
Pump 2 IR=100 M ohms. Motor side B phase cable lug
got carbonized. Motor lead box opened and no
abnormal thing noticed. However, it was
insulated with empire tape and PVC tape. B
phase cable lead retapping done with fibre
glass tape/empire tape/PVC tape. All the
cables were cleaned and new washers provided
for all studs.
Pump 2 R/Y/B motor side lug cleaned with CRC as it

was found carbonized.
Pump 3 No load current found 37/37/35 A. IR= 200 M 3 Y 6 M 9 D
ohms
Pump 3 Fexible copper conductor earthing provided
Pump 3 100 M ohms

Pump 3 200 M ohms
Pump 3 Disconnected power cable/space
heater/earthing for flywheel bearing
replacement.
Pump 3 As required by mech maintenance, connected
back the power cable/space heater/earthing.
Checked IR value of motor abd found 500 M
ohms. Replaced flywheel bearing. No load
current=38 A
Pump 3 IR=100 Mohms. Replaced cover plate bolts as

its thread had damaged.
Pump 3 IR=20 Mohms

Pump 3 IR=20 Mohms
Pump 3 IR=20 Mohms
Pump 3 IR=20 Mohms
Pump 3 Motor tripped on I0 twice. But worked for the
thrid time. Load current=100 A
105
Pump 3 Motor tripped on I0 twice. But worked for the
thrid time. Load current=100 A COOLANT
PUMP MOTOR 3 CTM relay put on
COOLANT PUMP MOTOR 2
Pump 3 Motor tripped on Ithand coil wire also burnt
Pump 3 IR=20 Mohms

Pump 3 Spare serviced motor Motor-3 installed. Motor 3 M 15 D
cable terminal block assembly removed from
spare COOLANT PUMP MOTOR motor
11804600. R phase cable lead insulation done
with empired and PVC tape. IR=50 Mohms.
No-load current= 40/40/38. Load current=103
A.
Pump 3 Replaced DE side grease pipe with a new one
Pump 3 No-load current= 35 A.

Pump 3 Pump stopped to check vibration reading.
After some time all running pump got tripped
on EMFF
Pump 3 DE side temp gone upto 77.4 C
Pump 3 DE side temp gone upto95 C after injecting 2
grams of grease
Pump 3 No-load current=38 A
Pump 3 R phase cable lead insulation burnt. Re-tapping
done. IR value=20 Mohms
Pump 3 R phase stud found hot and insulation tape

burnt. Insulation provided and tightened with
lock washer.
Pump 3 Motor disconnected due to abnormal noise.
Table A.4 Maintenance Data card for Motor-3
106

M.tech Thesis 22122015

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

M.tech Thesis 22122015

Diunggah oleh

Hak Cipta:

Format Tersedia

An Integrated Approach to Remaining Life Prediction of

[Enrolment No. ENGG01201301084]

Bhabha Atomic Research Centre

A thesis submitted to the

Board of Studies in Engineering Sciences

In partial fulfilment requirements

for the Degree of

Designation Name Signature

I hereby certify that no modifications was suggested by the M.Tech Committee

Date: Prof. P.V.Varde

diploma at this or any other Institution / University.

My Parents & Brother

attention to detail provide me with tremendous motivation.

and invaluable input into my research.

Sh. Ashish Khandelwal (SO/G, NPCIL).

make this project a success.

[Enrolment No: ENGG01201301084]

partial fulfilment of M.Tech programme.

Prof. P.V. Varde

3.3 STRESSORS, FAILURE MECHANISMS, AND CAUSES .................................. 18

5.2 DETERMINATION OF EQUIVALENT CIRCUIT OF MOTOR...........................60

5.3 TEMPERATURE MEASUREMENT.......................................................................64

5.4 EFFECT OF OVER-VOLTAGE...............................................................................65

5.5 EFFECT OF VOLTAGE UNBALANCE.................................................................69

6.2 COOLAT PUMP MOTOR.........................................................................................77

6.3 DATA ANALYSIS.....................................................................................................80

6.4 MTTF FOR WINDING FAILURE.............................................................................83

6.5 WEIBULL DISTRIBUTION FOR WINDING FAILURE........................................85

6.6 CALCULATION OF ACTIVATION ENERGY FOR HUMIDITY..........................88

6.7 REMAINING LIFE OF INSULATION SYSTEM.....................................................89

IN THE LIFETIME OF COOLANT PUMP MOTORS....................97

Table 3.1: NEMA Insulation Class of Motor (MG-1)........................................................ 15

Table 3.2.: Origin of electric motor stresses ………………………………………………21

Table 3.3: Percentage of failure by component Failed Component ……………………… 25

Table 3.4: FMEA of an Induction Motor ………………………………………………… 28

Table 4.1 Equivalent circuit parameters of induction motor ...............................................36

Table 4.2 No-load Equivalent circuit parameters of induction motor .................................37

Table 4.4 Parameters of induction motor ............................................................................40

Table 5.1: Motor Name Plate Data ......................................................................................57

Table 5.2: Auto-Transformer Name Plate Data ..................................................................57

Table 5.3 : No-load Test Results ........................................................................................60

Table 5.4 : Blocked-Rotor Test Results ..............................................................................62

Table 5.5: Equivalent circuit parameters of induction motor ..............................................63

Table 5.6: Parameters of induction motor ...........................................................................63

Table 5.7: Induction Motor- Temperature measured at various intervals ...........................64

Table 5.8: Thermal Model parameters of induction motor ................................................64

Table 5.9: Thermal Model parameters of induction motor-2 ..............................................65

Table 5.10 : Input Data for over-voltage test ......................................................................65

Table 5.11: Observed temperatures for various voltages ....................................................66

Table 5.13: Sequence Voltages for different unbalance conditions ....................................71

Table 5.15 : Input Data for unbalance test ..........................................................................72

Table 5.16: Observed temperatures for various % voltage unbalance ................................72

iii) 1.77% iv) 2.78 % .........................................................................................................74

Table 6.1: Coolant Pump Motor Data ................................................................................77

Table 6.2: Coolant Pump Motor operating duration data ..................................................84

Table 6.3: Input Data for Eb calculation ...........................................................................88

Table A.1: Maintenance data for Motor-0 ........................................................................97

Table A.2: Maintenance data for Motor-1 ......................................................................100

Table A.3: Maintenance data for Motor-2 ......................................................................103

Table A.4: Maintenance data for Motor-3 ......................................................................104

Figure 1.1: Types of Maintenance .............................................................................................4

Figure 3.1: Block Diagram of a Motor ....................................................................................12

Figure 3.2: Diagram of a stator of an Induction Motor ..........................................................13

Figure 3.3: Factors causing insulation failure in a Motor ……………………………….......24