A New Concept for Online Surge Testing for the Detection of Winding Insulation Deterioration in Low-Voltage Induction Machines

Abstract

A breakdown of the electrical insulation system causes catastrophic failure of the electrical machine and brings large process downtime losses. Preventive maintenance and online monitoring are some of the methods to improve the reliability and to reduce unscheduled downtime. One of the main reasons for the failure of the machine is the breakdown of the stator turn insulation. The offline surge test is the most commonly used offline test to assess the condition of the turn insulation. There is no online monitoring method that is applicable to low-voltage machines and has the same capabilities as the surge test. This paper introduces new concepts to implement an online surge test. The possibilities and limitations of the online surge test are presented, as well as the simulation and experimental results, to validate the concepts.

Full text

Page 1

IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 47, NO. 5, SEPTEMBER/OCTOBER 20112051
A New Concept for Online Surge Testing for the
Detection of Winding Insulation Deterioration
in Low-Voltage Induction Machines
Stefan Grubic, Student Member, IEEE, Jose Restrepo, Member, IEEE, Jose M. Aller,
Bin Lu, Senior Member, IEEE, and Thomas G. Habetler, Fellow, IEEE
Abstract—A breakdown of the electrical insulation system
causes catastrophic failure of the electrical machine and brings
large process downtime losses. Preventive maintenance and online
monitoring are some of the methods to improve the reliability
and to reduce unscheduled downtime. One of the main reasons
for the failure of the machine is the breakdown of the stator turn
insulation. The offline surge test is the most commonly used offline
test to assess the condition of the turn insulation. There is no online
monitoring method that is applicable to low-voltage machines and
has the same capabilities as the surge test. This paper introduces
new concepts to implement an online surge test. The possibilities
and limitations of the online surge test are presented, as well as the
simulation and experimental results, to validate the concepts.
Index Terms—Induction machine, motor diagnostics, online
testing, surge test, turn insulation failure.
I. INTRODUCTION
E
of the electrical energy in industry is consumed by electric
machines [1]. The unscheduled process downtime caused by
a failure of an electrical machine can cause enormous costs.
In some settings, an electrical machine failure can cause entire
assemblylinestostopandthusinterrupttheproduction process.
It is therefore desirable that a problem with the critical com-
ponents of the motor like the bearing or the stator insulation
system is identified at an early stage in order to perform
a scheduled machine service or replacement. The economic
LECTRICAL machines are an important component in
industrial applications. It is estimated that more than 60%
Manuscript received January 18, 2011; revised May 4, 2011; accepted
May 15, 2011. Date of publication July 14, 2011; date of current version
September 21, 2011. Paper 2011-EMC-028.R1, presented at the 2010 IEEE
Energy Conversion Congress and Exposition, Atlanta, GA, September 12–16,
and approved for publication in the IEEE TRANSACTIONS ON INDUSTRY
APPLICATIONS by the Electric Machines Committee of the IEEE Industry
Applications Society. This work was supported in part by the U.S. Department
of Energy under Grant DE-FC36-04GO14000.
S. Grubic and T. G. Habetler are with the School of Electrical and Computer
Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0250 USA
(e-mail: s.grubic@gatech.edu; thabetler@ece.gatech.edu).
J. Restrepo is with the Departamento de Electrónica y Circuitos, Universidad
Simón Bolívar, Caracas 1080A, Venezuela (e-mail: restrepo@ieee.org).
J. M. Aller is with the Departamento de Conversión y Transporte de Energía,
UniversidadSimónBolívar,Caracas1080A,Venezuela(e-mail:jaller@usb.ve).
B. Lu is with the China Innovation Center, Eaton Corporation, Shanghai
200335, China (e-mail: binlu@ieee.org).
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TIA.2011.2161972
losses of the process downtime caused by an unexpected outage
of the machine exceed the machine maintenance costs to a
considerable extent. On an offshore oil plant, for example, the
downtime losses caused by motor failures can be as high as
$25000/h [2].
The most common reasons for a breakdown of electrical
machines are bearing failures and failures related to the stator
insulation [3]–[5]. Stator faults account for a large percentage
of the machine failures. About 80% of all electrical failures in
the stator originate from a weak turn-to-turn insulation [6]. A
turnfaultcanpropagateandquicklydevelopintoagroundfault.
This may result in significant damage to the machine stator
core, which will require it to be repaired or replaced before
a stator rewind can be completed. Thus, the turn insulation is
one of the most critical components to be monitored. There are
several online methods that are capable of detecting solid turn
faultswhicharemostcommonlybasedonthesignatureanalysis
of a suitable motor quantity like the current or the analysis
of a sequence component of the motor impedance matrix or
the current [7], [8]. Despite all the progress made in the field
of stator monitoring, there is still no commonly used method
that is able to not only diagnose a solid turn fault but also
detect the deterioration of the turn insulation. The only test
that is capable of finding a weakness prior to the breakdown
of the stator insulation and applicable to low-voltage machines
is the surge test, which is conducted when the machine is not
operating [7], [9].
This paper proposes a method of how to apply the surge test
to an operating machine. The criteria used for the evaluation
of the online results are briefly introduced, and some of the
practicalchallenges aredescribed. Simulation andexperimental
results are presented to validate the proposed technique. A brief
synopsis of the offline surge test is given as well.
II. OFFLINE SURGE TEST
The surge test is a very commonly used offline test to assess
the integrity of the turn-to-turn insulation of the stator. The
benefits and dangers of the surge test have been discussed
elaborately in the literature, and the methods to conduct and
evaluate the measurements have continually been improved
over the years [6], [7], [9]–[17]. There is no other offline or
online test that is applicable to low-voltage machines with a
similar capability. For example, the partial discharge test is
0093-9994/$26.00 © 2011 IEEE

Page 2

2052IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 47, NO. 5, SEPTEMBER/OCTOBER 2011
Fig. 1.
for the surge test.
Offline surge test schematic. (a) Surge test setup. (b) Equivalent circuit
applicable to medium- and high-voltage machines only, and
other methods like the winding resistance test or the inductive
impedance test can detect a solid turn fault only due to the
imbalances in the resistance or inductance, respectively.
The principle of surge testing is to apply a short current pulse
with a fast rise time to the windings of the stator. By Lenz’s
law, there is a voltage induced between the adjacent loops of
the winding. If the voltage exceeds the maximum voltage that
the insulation can withstand, there will be an arc developing,
and the inductance of the coil will be modified for a short time.
This process can be detected by observing the impulse response
of the motor, which is also called the surge waveform. In a
practical application, a capacitor is charged up to a specified
voltage level and subsequently discharged in one of the motor’s
windings. The schematic is shown in Fig. 1(a) and (b).
In a first-order approximation, the capacitor and the motor
present an RLC series circuit. If there is a short circuit between
the turns of the insulation due to a deteriorated winding, a
change in the frequency and the magnitude of the impulse
response can be observed. The ringing frequency of the damped
sinusoidal waveform can be determined by solving the RLC
series circuit shown in Fig. 1(b) for the damped resonance
frequency [15]
??
where R is the overall resistance, L is the motor’s inductance,
and C is the surge capacitance.
By applying voltages that are significantly higher than the
rated voltage of the machine, a weakness in the insulation
that is not apparent under rated conditions can be found. The
recommended test voltages can be found in IEEE 522 [15]
and NEMA MG1 [16]. As a rule of thumb, the maximum test
voltage can be determined by [11]
f =
1
2π
1
LC−R2
4L2
?
(1)
Vmax= 1000 V + 2 ∗ Vrated.
(2)
Some of the topics investigated in the research related to the
surge test are the risk of performing a surge test [14], [18]–
[20], the analysis of the voltage distribution, and the effect of
the surge rise time [21]–[23].
In recent years, different methods to conduct and evaluate
the surge test have been developed, as well as reliable test
equipment that makes it easy to perform the surge test. A
common method to evaluate the surge test is the error area ratio
(EAR). It is very sensitive and detects a difference between
two waveforms that is difficult to observe visually. The EAR
is determined by the following equation [14]:
EAR =
N ?
i=1
???F(1)
j=1
i
− F(2)
???F(1)
i
???
N ?
j
???
∗ 100
(3)
where F(1)
waveform), F(2)
waveform (test waveform), and N is the number of data points
that are compared. Two identical waveforms will have an EAR
of 0%.
The EAR can be applied in different ways. A method called
thepulse-to-pulseEAR(P–PEAR)thatisimmunetotheeffects
of rotor position, rotor condition, winding configuration, motor
connection, ironcondition,andsaturationisthemostrobustand
reliable evaluation method for the surge test. During the test
using the P–P EAR, the voltage is increased in well-defined
steps. Successive waveforms are evaluated by using the EAR.
Sincethetestvoltagelevelisincreased,theEAR willbe> 0%.
If the EAR changes by more than the value expected, this is an
indicator forarcingand,thus,foraweak turn-to-turninsulation.
i
is the ith point of the first waveform (reference
is the corresponding point of the second
i
III. ONLINE SURGE TEST
A comprehensive literature survey has shown that there is no
equivalent to the offline surge test in online applications [8].
In order to overcome this limitation, an investigation on the
applicability of the surge test to an operating machine is inves-
tigated. A rather general schematic of the online test configura-
tionisshowninFig.2.Thesurgecapacitorsareconnectedatthe
phase terminals of the motor under test through insulated-gate
bipolar transistor (IGBT) switches. The device must be able to
test each phase of the motor. This can be accomplished by the
configuration depicted, for example, yet other configurations
are conceivable, too. Since only a single capacitor is needed at
a time, multiple switches to connect this capacitor to the phase
under investigation can be used to conduct the test. In an online
test, the capacitor discharges not only through the investigated
coils of the motor but rather through the line leads, i.e., the
energy-supplying circuit. The impedance of the line leads at the
frequencies of interest is well depicted by a serial combination
of a resistor and an inductor. Looking into the network from
the side of the surge capacitor, the mains output impedance is
placed in parallel to the motor impedance. For simplicity, the
mains output impedance will be called the supply impedance
in the following discussion. Similarly, the resistive part and
the inductive part of this impedance will be called the supply
resistance and supply inductance, respectively.
A simplified one-phase equivalent circuit is shown in Fig. 3,
where L1and R1are the supply inductance and resistance, L2
and R2are the equivalent motor inductance and resistance, C
is the capacitance of the surge tester, and S is the IGBT switch.

Page 3

GRUBIC et al.: NEW CONCEPT FOR ONLINE SURGE TESTING FOR DETECTION OF INSULATION DETERIORATION2053
Fig. 2.Basic online surge schematic with additional supply impedance.
Fig. 3. Single-phase equivalent circuit of the online surge test.
If all resistance is neglected, the frequency of the surge
waveform can be determined as
fn=
1
2π
?
C
L1L2
L1+L2
(4)
where L1is the supply inductance, C is the surge capacitance,
and L2is the equivalent motor inductance, as shown in Fig. 3.
The equivalent motor inductance is approximately equal to the
combined leakage inductance of the motor in the vicinity of the
surge test frequency.
From (4), it can be shown that, for a supply inductance
much smaller than the equivalent motor inductance (L1? L2),
the resonance frequency is mainly determined by the supply
inductance
fn≈
1
2π√CL1
.
(5)
A comparison to the offline test can be made to show the impact
of the supply inductance on the test result. The equation for
the oscillatory frequency of the offline test can be obtained
from (4) by letting L1 grow to infinity. The frequency is
then fn= 1/(2π√CL2). Thus, the frequency is determined
only by the motor’s equivalent inductance L2. However, in
the online configuration, the frequency is determined mainly
by the smaller equivalent supply inductance L1as shown in
(5). Therefore, a modification of the motor’s inductance (L2)
in the online configuration due to a breakdown of the turn
insulation will have an impact on the surge waveform much
smaller than in the offline test. A decrease of the ratio between
the supply inductance and the motor leakage inductance results
in a decrease of the test sensitivity. If no provisions are taken, a
weakness in the turn insulation of the motor cannot be detected
due to the small supply impedance.
One option to realize the application of the online surge test
is to increase the supply impedance. Of course, the disruptive
effect of this impedance enhancement on the operation of the
machine has to be kept to a minimum. This requires a method
of turning the impedance “on and off” in such a way that it is
only present during the test and has almost no influence on the
steady-stateoperationofthemotor.Thetransientsduringthein-
sertion and removal of the impedance should not be disruptive.
The frequency of the surge test waveform can be adjusted to be
more than two orders of magnitude higher than the operating
frequency. An ideal supply impedance under these conditions
would be zero around the fundamental operating frequency and
infinitely large around the frequency of the surge transient.
Such an ideal supply impedance is physically not realizable,
although some implementation can exhibit features close to
those required. For the realizable impedances, there will always
be a tradeoff between how much the operation of the machine
will be affected and how high the sensitivity of the test will
be. In addition, there is a practical limit on the components of
this hypothetical impedance determined by the current and the
voltage ratings of the motor.
Fig. 4 shows the frequency characteristics of the ideal sup-
ply impedance and the three realizations that can be used to
separate the motor from the supply during the test. The basic
schematic of the online test configuration is shown in Fig. 2,
whereonlyoneofthedifferentimpedances showninthedashed
boxes is added to all three phases. The impedances investigated
are as follows:
1) a series LC resonant circuit with a resonance at 60 Hz;
2) the disconnection of the motor;
3) an additional inductive impedance.

Page 4

2054IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 47, NO. 5, SEPTEMBER/OCTOBER 2011
Fig. 4.
the motor from the supply.
Frequency characteristics of supply impedances suggested to separate
Clearly, the series LC resonant circuit comes closest to the
ideal impedance around the frequencies of interest. However,
the analysis shows that this impedance can introduce motor
instability and is therefore unsuitable. The phenomenon leading
to instability is called subsynchronous resonance and has been
investigated in [24].
Disconnecting the motor for a short time offers the most
auspicious test conditions due to the infinite impedance in
the supply line but also poses the biggest disturbance to the
operation of the machine. To avoid voltage spikes, the motor
can be disconnected only when the current is zero. The du-
ration of the disconnection has to be kept short in order to
avoid a large drop-off in speed and large transients when the
motor is reconnected to the supply after the application of the
surge.
More details, as well as simulation results for the insertion
of a series LC resonant circuit and the disconnection of the
machine, are shown in [25].
The simplest choice of performing an online test is the inser-
tion of an additional inductance, which is also the most suitable
method for a practical application. The sensitivity of the test
will be lower than that for a machine disconnected or for the
offline test, respectively, but it will still be high enough to detect
a turn insulation problem for an appropriate impedance. The
reduction in frequency sensitivity can be determined as follows.
The frequency of the surge waveform will approximately be
determined by (4). The frequency sensitivity is given by the
following equation:
Sω=∆ω/ω
∆L/L= −1
2
L2
L1+ L2
(6)
where L1 is the equivalent inductance of the motor and L2
is the supply inductance. The thorough derivation of the fre-
quency and EAR sensitivity is beyond the scope of this paper
and will be discussed in more detail in future publications.
If, for example, the supply inductance chosen is equal to
the equivalent motor inductance (L1= L2), the frequency
of the surge waveform will increase by a factor of√2, and the
frequency sensitivity will decrease by a factor of two, compared
to the offline test (Sω= −1/4). A bigger supply inductance
results in a higher sensitivity of the test. The disadvantage
of a bigger inductance is the increase in the voltage drop
across the inductor. For high currents and voltages, the geo-
metrical size of the proper inductor might be an additional
drawback.
There are different options for inserting the additional supply
inductor, i.e., for how to switch it on and off. Two different
approaches are to be distinguished here: an abrupt inductance
transitionandasmoothinductancetransition.Theabruptinduc-
tance transition refers to a supply inductance that is changed in
a stepwise fashion. This can be achieved by installing a switch
in parallel to the supply inductance that is closed during normal
operation (bypass mode) and open during the test (test mode).
Another solution is to use a transformer with a switch on the
secondary that is closed during operation and open during the
test. If the switch is closed, the leakage inductance, which is
negligibly small, is seen on the primary. When the switch is
open, the sum of primary leakage and magnetizing inductances
is seen on the primary. The switching of the inductor has to
be synchronized with the zero crossings of the line current to
mitigate voltage spikes.
A smooth inductance transition refers to a supply inductance
with a gradual transition between a low value, that it assumes
during operation, and a high value, which is suitable for the
test. The advantage of this method is that the transients in
current, torque, and speed during the modification of the in-
ductance are significantly smaller than for an abrupt change of
the inductance. Even if the inductance is increased when the
current is nonzero, there will be no voltage spikes. In order to
illustrate the difference between the transients in torque and
speed for the abrupt and the smooth inductance transitions, a
simulation has been implemented in MATLAB. The inductance
simulated changes linearly between its minimum (10−5p.u.)
and maximum values (0.2 p.u.). The transition time between
the lowest and highest inductance values for the smooth change
is 0.1 s. The machine is operated at rated load and has a
combined leakage inductance of 0.2 p.u. The results are shown
in Fig. 5.
The oscillations in torque and speed for the smooth inductor
changearenegligiblecomparedtotheoscillationscausedbythe
abrupt inductor change. The disadvantage of the smooth induc-
tor change is a significant increase in hardware requirements
and control complexity. An example for an inductor that can
perform a smooth change in inductance is given in [26].
A simulation of the online surge test has been implemented
in MATLAB/Simulink to show the influence of the supply
inductance on the surge waveform. Fig. 6 shows the surge
waveform for an initial capacitor voltage of 4.34 p.u. (equiv-
alent to 2000 V for a 460-V line–line voltage). The supply
inductance is chosen equal to the combined leakage inductance
of the motor. For comparison, the offline surge waveform
is depicted, as well as the test results without an additional
supply inductance (i.e., the supply impedance switch is closed
during normal operation and during the test). The addition of
the supply inductance only makes it possible to conduct the
surge test online. Without the additional supply inductance,
the frequency of the waveform is significantly higher while the
influence of the motor inductance on the waveform becomes
negligible.

Page 5

GRUBIC et al.: NEW CONCEPT FOR ONLINE SURGE TESTING FOR DETECTION OF INSULATION DETERIORATION2055
Fig. 5.
the additional supply inductor. (a) Torque and speed transients for the abrupt
and smooth inductance changes. (b) Close-up of (a).
Simulation results for the abrupt and smooth inductance changes of
IV. EXECUTION AND EVALUATION
OF THE ONLINE SURGE TEST
The execution of the surge test requires the application of
multiple impulses to the motor at increasing voltage levels. The
voltage is increased in well-defined increments of ∆V (e.g.,
50 V) up to a maximum test voltage or to the level where
insulation breakdown is detected. In the offline test, there is
no restriction to the point of time when the capacitor can be
discharged since no other voltage is applied to the machine.
In the online test, the discharge of the capacitor has to be
synchronized with the operating voltage. In order to have a
well-defined test voltage, the moment that is the most suitable
for the discharge of the capacitor is the zero crossing of the
line–line voltage. If a different point of time is chosen for the
capacitor discharge, the line–line voltage at that point has to be
determined and taken into account, which increases the control
complexity.
The surge test requires the capacitor to be discharged multi-
ple times. If the supply impedance is increased by disconnect-
Fig. 6.
inductance in all three phases. (a) Online surge waveform for the additional
supply inductance in all three phases compared to the offline waveform and the
online waveform without additional supply impedance. (b) Online surge wave-
form for the additional supply inductance in all three phases with fundamental
removed.
Simulation results for the online surge test with an additional supply
ing the motor, transients as shown in [25] occur each time the
capacitor is discharged. However, if an additional impedance is
used, the operation of the motor is hampered by the transients
as shown in Fig. 5 and [25] during the test only once. The
impedance can be turned on before the first capacitor discharge
and turned off after the last capacitor discharge and does not
have to be switched on and off continually, i.e., the supply
impedance will be in the test mode until the final test voltage
is reached. This feature makes the use of an additional supply
inductance more practical than the temporary disconnection of
the motor.
A major drawback of the online test compared to the offline
test is the change of the rotor position. In [27], it is shown that
a change in inductance due to a change in the rotor position
is reflected in the surge waveform. During the offline test, the
rotor does not change its position, and thus, the test result is not
affectedbyrotoreccentricityorrotorslotting.Themostsuitable
moment to conduct the online test is at the zero crossing of
the line–line voltage. If the motor is operated at no load, the
slip is close to zero, and the rotor position does not change
significantly with respect to the zero crossings of the line–line