19

CHAPTER 2
TESTING OF STATOR WINDING

2.1

INTRODUCTION
The rotating machine consists of three main components in stator

such as the copper conductors, the stator core and the insulation. The copper
is a conduit for the stator winding current. In a generator, the stator output
current is induced to flow in the copper conductors as a reaction to the
rotating magnetic field from the rotor. In a motor, a current is introduced into
the stator, creating a rotating magnetic field that forces the rotor to move. The
copper conductors must have a cross section large enough to carry all the
current required without overheating (Say 1976).
Three basic types of stator winding structures are employed over
the range from 1kW to more than 1000 MW:

2.1.1

i.

Random -wound stators

ii.

Form- wound stators using multiturn coils

iii.

Form- wound stators using Roebel bars

Random-Wound Stators
Random-wound stators consist of round, insulated copper

conductors (magnet wire or winding wire) that are wound continuously (by
hand or by winding machine) through the slots in the stator core to form a

20

coil. Most of the turns in the coils can be easily seen. Each turn (loop) of the
magnet wire could, in principle, be placed randomly against any other turn of
magnet wire in the coil, independent of the voltage level of the turn, thus the
term random. Since a turn that is connected to the phase terminal can be
adjacent to a turn that is operating at low voltage (i.e., at the neutral point),
random -wound stators usually operate at voltages less than 1000 V. This
effectively limits random-wound stators to machines less than several
hundred kilo-watts (kW) or Horse Power (HP).
2.1.2

Form-Wound Stators-Coil Type

Form-wound stators are usually intended for machines operating at

1000 V and above. Such windings are made from insulated coils that have
been preformed prior to insertion in the slots in the stator core. The preformed
coil consists of a continuous loop of magnet wire shaped into a coil
(sometimes referred to as a diamond shape), with additional insulation applied
over the coil loops. Usually, each coil can have from 2 to 12 turns, and several
coils are connected in series to create the proper number of poles and turns
between the phase terminal and ground (or neutral). By minimizing the
voltage between adjacent turns, thinner insulation can be used to separate the
turns. For example, in a 4160 volt stator winding (2400 Volt line-to-ground),
the winding may have 10 coils connected in series, with each coil consisting
of 10 turns, yielding 100 turns between the phase terminal and neutral. The
maximum voltage between the adjacent turns is 24 V.
2.1.3

Form-Wound Stators - Roebel Bar Type

In large generators, the more the power output, the larger and

mechanically stiffer each coil usually is. In stators larger than about 50 MW,
the form-wound coil is large enough that there are difficulties in inserting
both legs of the coil in the narrow slots in the stator core without risking

21

mechanical damage to the coil during the insertion process. Thus, most large
generators today are not made from multi-turn coils, but rather from half-turn
coils, often referred to as Roebel bars. With a Roebel bar construction, only
one half of a coil is inserted into the slot at a time, which is considerably
easier than inserting two sides of a coil in two slots simultaneously.
2.2

STATOR WINDING INSULATION SYSTEM

The Stator winding insulation system contains several different

components and features, which together ensure that electrical shorts do not
occur, that the heat from the conductor I2R losses are transmitted to a heat
sink and that the conductors do not vibrate in spite of the magnetic forces.
The basic stator insulation system components are listed below (Stone et al
2004)
i.

Strand (or sub conductor) insulation

ii.

Turn insulation

iii.

Groundwall (or ground or earth) insulation

Figure 2.1 shows the cross section of form-wound coils in a stator

slot, and identifies the above components. In addition to the main insulation
components, the insulation system sometimes has high-voltage stress-relief
coatings and end-winding support components (Allison 2000). The StressGrading systems have to regulate the potential distribution on the surface of
the coil overhangs at the slot to prevent the surface discharge, which acting
over a longer period of time could destroy the insulation. They have to retain
this ability not only during the testing of windings when the voltage
considerably exceeds the rated voltage, but also during the whole operating
life of the machines. So the task is to suppress any discharges which may
occur at the maximum voltage applied to the winding, both within the slot and
at the ends.

22
Bottom packing
Strand insulation

Turn insulation
Groundwall
Insulation
Midstick packing

Iron

Semi conductive
Coating
Top packing

Slot wedge

Figure 2.1 Form-Wound Multiturn Coils

Normally inside the slot, semi-conductive, low resistance black

carbon/graphite corona shield materials are used; because conductive
materials have a uniform resistance which is not voltage dependent. They are
always black in appearance and are applied only on the straight portion of the
coil/bar. At the coil ends, semi-conductive high resistance gray siliconcarbide materials are used for stress grading system.
Stress grading materials are non-linear i.e. their resistivity is
changing depending on the applied electrical stress, even though it may seem
uniform over part of the range of applied voltage. They are normally gray in
color and are only applied at the coil ends. The characteristics of conductive
and stress-grading materials are shown in Figure 2.2.

23

0.12

Conductive tape in Slot

0.10
Current
(in mA)

Stress grading tape at ends

(non-linear)

0.08
0.06
0.04
0.02

500

1000

1500

2000

2500

3000

Voltage (V)

Figure 2.2 Material Characteristic Curves

2.2.1

Strand Insulation
Many form-wound machines employ separate strand and turn

insulation. The following mainly addresses the strand insulation in formwound coils and bars. There are both electrical and mechanical reasons for
stranding a conductor in a form-wound coil bar. From a mechanical point of
view, a conductor that is big enough to carry the current needed in the coil or
bar for a large machine will have a relatively large cross sectional area. That
is, a large conductor cross section is needed to achieve the desired capacity.
Such a large conductor is difficult to bend and form into the required coil/bar
shape. A conductor formed from smaller strands (also called sub-conductors)
is easier to bend into the required shape than one large conductor.
From an electrical point of view there are reasons to make strands
and insulate them from one another. A copper conductor has a large enough
cross-sectional area, where the current will tend to flow on the periphery of
the conductor. This is known as the skin effect. The skin effect gives rise to a

24

skin depth through which most of the current flows. The skin depth of copper
is 8.5 mm at 60 Hertz (Hz). If the conductor has a cross section such that the
thickness is greater than 8.5 mm, there is a tendency for the current not to
flow through the centre of the conductor, which implies that the current is not
making use of all the available cross section. This is reflected as an effective
AC resistance that is higher than the DC resistance.
The higher AC resistance gives rise to a larger I2R loss than if the
same cross-section had been made from the strands that are insulated from
one another to prevent the skin effect from occurring by making the required
cross section from the strands that are insulated from one another, all the
copper cross section is used for current flow, the skin effect is negated, and
the losses are reduced. The electrical reason for stranding requires the strands
to be insulated from one another. The voltage across the strands is less than a
few tens of volts; therefore, the strand insulation can be very thin. The strand
insulation is subject to damage during the coil manufacturing process, so it
must have good mechanical properties. Since the strand insulation is
immediately adjacent to the copper conductors that are carrying the main
stator current, which produces the I2R loss, the strand insulation is exposed to
highest temperatures in the stator.
2.2.2

Turn Insulation
The purpose of the turn insulation in form wound stators is to

prevent the shorts between the turns in a coil. A turn short occurs whereas the
shorted turn will appear as the secondary winding of an autotransformer. For
example, the winding has 100 turns between the phase terminal and neutral
and when the dead short appears across one turn then 100 times the normal
current will flow in the shorted turn. This follows from the transformer law as
given in equation (2.1).

25

n p I p ns I s

(2.1)

where n refers to the number of turns in the primary or secondary, and I refers
to the current in the primary or secondary. Consequently, a huge circulating
current will flow in the faulted turn, rapidly overheating it. This high current
will be followed quickly by a ground fault due to melted copper burning
through any groundwall insulation. Clearly, effective turn insulation is needed
for long stator winding life.
The power frequency voltage across the adjacent turns in a formwound multiturn coil is well defined. Essentially, one can take the number of
turns between the phase terminal and neutral and divide it into the phaseground voltage to get the voltage across each turn. For example, if a motor is
rated 4160Vrms (phase-phase) and the phase-ground voltage is 2400 V, this
will result in about a 24 Vrms across each turn, if there are 100 turns between
phase end and neutral. This occurs because coil manufacturers take
considerable trouble to ensure that the inductance of the coil is the same, and
that the inductance of each turn within a coil is the same. Since the inductive
reactance (XL) in ohms is given in equation (2.2).
XL

2 fL

(2.2)

where f is the frequency of the AC voltage and L is the coil or turn

inductance, the turns appear as impedances in a voltage divider, where the
coil series impedances are equal. In general, the voltage across each turn will
be between 10 Vac (small form-wound motors) to 250 Vac (for large generator
multiturn coils). The turn insulation in form-wound coils can be exposed to
very high transient voltages associated with motor starts, Inverter Fed Drives
(IFDs) operation, or lightning strikes (Gupta 1987). Such transient voltages
may age or puncture the turn insulation.

26

2.2.3

Groundwall Insulation
Groundwall insulation is the component that separates the copper

conductors from the grounded stator core. Groundwall insulation failure

usually triggers a ground fault relay, taking the motor or generator off-line.
Thus the stator groundwall insulation is critical to the proper operation of a
motor or generator. For a long service life, the groundwall must meet the
rigors of the electrical, thermal and mechanical stresses.
The groundwall insulation in form-wound multi-turn coils or bars
connected to the phase end of the winding will have the full rated phase to
ground voltage across it. A stator rated at 13.8 kV (phase-to-phase) will have
a maximum of 8 kV (13.8/ 3 ) between the copper conductors and the
grounded stator core. This high voltage requires a substantial groundwall
insulation thickness. The high groundwall voltage only occurs in the
coils/bars connected to the phase terminals. The coils/bars connected to the
neutral have essentially no voltage across the groundwall during the normal
operation.
The groundwall insulation in indirectly cooled form-wound
machines is the main path for transmitting the heat from the copper
conductors to the stator core. Thus the groundwall insulation should have as
low a thermal resistance as possible, to prevent high temperatures in the
copper. To achieve a low thermal resistance it requires the groundwall
materials to have a high thermal conductivity as far as possible and for the
groundwall to be free of voids. Such air voids block the flow of heat, in the
same way that two layers of glass separated by a small air space inhibits the
flow of heat through a window. Therefore the insulation must be able to
operate at high temperatures and be manufactured in such a way as to
minimize the formation of air pockets within the groundwall.

27

There are large magnetic forces acting on the copper conductors.

These magnetic forces are primarily the result of the two magnetic fields from
the current flowing in the top and bottom coils/bars in each slot. These fields
interact, exerting a force that makes the individual copper conductors as well
as the entire coil or bar vibrate up and down in the slot.
The groundwall insulation must also help to prevent the copper
conductors from vibrating in response to the magnetic forces. The groundwall
were full of air pockets, the copper conductors might be free to vibrate. This
would cause the conductors to bang against the remaining groundwall
insulations, as well as allowing the copper strands and turns to vibrate against
one another, leading to insulation abrasion (Stone and Maughan 2008). An
incompressible insulating mass exists between the copper and the coil surface,
and then the conductors cannot move.
2.2.3.1

Groundwall Partial Discharge Suppression

In form-wound bars and coils rated greater than about 4 kV, Partial

Discharges (PDs) can occur within the groundwall insulation or between the
surface of the coil or bar and the stator coil. These Partial Discharges, which
are sometimes called coronas, are created by the high voltage stress that
occurs in the groundwall. An air pocket exists in the groundwall; the high
electric stress will break down the air, causing a spark. This spark will
degrade the insulation and, repeated discharges will eventually erode a hole
through the groundwall to prevent stator winding failure. In addition, a partial
discharge suppression system is needed to prevent PD in any air gaps between
the surface of the coils and bars and the core.
Electric breakdown strength is also a property of an insulating
material. Electric breakdown is not governed by voltage alone. Also it

28

depends on the electric field. Electric stress, E in parallel plate geometry is

given by the equation (2.3).

V
d

kV
mm

(2.3)

where V is the voltage across the metal plates in kV and d is the distance
between the plates in mm. If the voltage is gradually increased across the
metal plates, there will be a voltage at which electric break down occurs, i.e.,
at which a spark will cross between the plates.
The presence of air pockets within the groundwall can lead to the
electric break down of the air pockets, a process called a Partial Discharge
(Kuffel et al 1998). To understand this process, consider the groundwall cross
section in Figure 2.3. For electric breakdown to occur in the air pocket there
must be a high electric stress across it.

Stator core
Copper

0.5 mm internal air-filled void

4 mm Groundwall insulation

Figure 2.3 Coils with Air Packets Next to the Turn Insulation

29

Using a simple capacitive voltage divider circuit as shown in

Figure 2.4, one can calculate the voltage across the air pocket. The
capacitance of the air pocket, to a first approximation, can be calculated,
assuming it is a parallel plate capacitor, i.e.,

Copper

Ca

Va

V0
Cin
Core

Figure 2.4

An Electrical Equivalent Circuit of a Coil with Air Packets

Next to the Turn Insulation

Ca

where

A
da

(2.4)

is the permittivity of the insulating material, A is the cross sectional

area of the void, and d a is the thickness of the void. The permittivity is often
represented as shown in equation (2.5).
(2.5)

r 0

where

is the relative dielectric constant and

is the permittivity of free

space, equal to 8.8510-12 F/m. the dielectric constant for air is 1.0. For most
stator winding insulation materials, the dielectric constant is about 4.

30

Thus, assuming unity cross-sectional area, the capacitance of the air

pocket in Figure 2.4 can be calculated. The air pocket is in series with another
capacitor, which represents the capacitance (Cin) of the solid insulating
material. Using equation (2.4), assuming a dielectric constant of 4 and the
thickness of the insulation is 4mm, the insulation capacitance can be
calculated to a first approximation. Using simple circuit theory, the voltage
across the air pocket can be calculated by the equation (2.6).

Va

CinV0
Ca Cin

(2.6)

where V0 is applied AC voltage (8 kV rms if the coil is at the phase terminal)

and Cin is the solid insulating material capacitance. Using the above
equations, the dimensions in Figure 2.3, recognizing that A and

will cancel

out, and assuming the dielectric constants are 1 and 4 for air and the
insulation, respectively, one can calculate that the voltage across the air
pocket is 33% of the applied voltage. For a V0 of 8 kV rms (rated phaseground voltage for a phase-end coil in a 13.8 kV stator), the voltage across the
air pocket is about 2.6 kV. From Equation 2.3, this implies that the electric
stress within the air pocket is 5.2 kV/mm. This far exceeds the 3 kV/mm
electric strength of the air, and thus electric breakdown will occur within the
air. The resulting spark is called a partial discharge. The discharge is referred
to as partial since the spark is only in the air pocket or void (Bartnikas 1980).
2.3

OFF-LINE TEST
The off-line tests describe the main tests that are commercially

available for assessing the condition of the insulation of stator windings as

shown in Figure 2.5. All the tests require the machine to be removed from
service, at least for a short time. Tremendous advancements in testing
technology have been made, due to availability of better electronic

31

equipments like 4 Digit Micro Ohm Meter, Insulation Tester, Surge

Comparison Tester, Tan Delta Test Kit, AC Hi - Pot Tester, DC Hi - Pot
Tester, Inter Turn Tester (High Voltage High Frequency test kit), Recurrent
Surge Oscilloscope (RSO), Digital Volt and Clamp Meter, Infrared
Temperature Scanner, Partial Discharge Detector, High Voltage Probe and
computers with sophisticated data analysis software like MATLAB 7.0.4.

Figure 2.5 Stator Winding of 11 kV Machine

In this chapter the purpose of each test is described, together with
the types of machines and/or windings it is used for. The procedure of the test
is also described where it is not obvious. In addition, each test is being
compared with other similar tests (Stefan Grubic et al 2008). Practical
information is given on how to apply the test, including the state the winding
must be in to do the test and the normal time required to conduct the test.
Finally, a practical guide is given on interpreting the results. This
interpretation will reflect the experience of test users. The condition of the
stator winding is critical for the overall motor healthiness. To ensure the

32

flawless operation of a motor system, various off-line tests can be performed.

These tests allow the user to assess the condition of the motor under test.
Off-line methods are normally more direct and accurate. However, most of
these tests can only be applied to motors and generators that are disconnected
from services. This is one of the main drawbacks compared to the onlinemonitoring methods. The off-line tests are summarized in Table 2.1.
Table 2.1

Different Methods of Tests in the Stator Insulation of

Electrical Drives

S.No.

Method

Standards

Insulation Tested and

Diagnostic Value

1.

Insulation Resistance IEEE 43.

Find contaminations and
(IR) / Megohm
NEMA MG 1 defects in phase-to-ground
insulation

2.

Polarization Index
(PI)

IEEE 43

3.

DC High Potential
Test (DC Hi Pot)

IEEE 95,

4.

AC High Potential
Test (AC Hi Pot)

IEC 60034

5.

Surge Comparison
Test

IEEE 522

6.

Offline Partial
Discharge(PD) Test

IEEE 1434

Detects deterioration of the

phase-to-ground and turn-toturn insulation

7.

Dissipation-Factor

IEEE 286

Detects deterioration of the

phase-to-ground and phase-tophase insulation

Find contaminations and

defects in phase-to-ground
insulation

Find contaminations and

defects in phase-to-ground
IEC 34.1,
NEMA MG 1 insulation
Find contaminations and
NEMA MG 1 defects in phase-to-ground
insulation
Detects deterioration of the
NEMA MG 1 turn-to-turn insulation

IEC 60894

33

Common methods used to test phase-to-ground insulation are the

Insulation Resistance (IR) Test, the Polarization Index Test, the DC and AC
High Potential Test, and the Dissipation Factor Test. Since these tests can be
conducted on a frequent basis without using additional equipment, groundwall insulation problems can be diagnosed at an early stage.
2.4

IR, PI AND LEAKAGE CURRENT

The most widely used diagnostic test for windings of rotating

machines is IR/PI test and this test successfully locates pollution and
contamination problems in windings. In older insulation systems, the test can
also detect thermal deterioration. Insulation Resistance (IR) and Polarization
Index (PI) tests have been in use for more than 70 years. Both tests are
performed with the same instrument, and are usually done at the same time
(IEEE 43-2000).
The IR test measures the resistance of the electrical insulation
between the copper conductors and the core of the stator. Ideally, this
resistance should be infinite. In practice, the IR is not infinitely high. Usually,
lower the insulation resistance, it is more likely that there is a problem with
the insulation.
The PI test is a variation of the IR test. PI is the ratio of the IR
measured after voltage has been applied for 10 minutes (R10), to the IR
measured after just one minute (R1), i.e. PI is defined as given in
equation (2.7).
PI

R10
R1

(2.7)

A low PI indicates that a winding may be contaminated or wet. The

IEEE 43-2000 guide gives an extensive discussion of the theory of the IR/PI
tests as applied to rotating machines.

34

In the test, a relatively high DC voltage is applied between the

winding copper and the stator. The current flowing in the circuit is then measured. The insulation resistance at time t is given in equation (2.8).

Rt

V
It

(2.8)

where V is the applied DC voltage and It is the total current measured after
t minutes, as the current is not usually constant. There are at least four
currents that may flow, when a DC voltage is applied to the winding (Rux
1977). They include:
2.4.1

Capacitive Charging Current

This Current de

where C is the capacitance and R is the internal resistance of the voltage

source, typically a few hundred kilo ohm. The Capacitance of a form-wound,
stator coil may be about 10 nF between the copper and the core, and that of
large hydro generator may be about 1 F. This current may decay to zero in
less than 10 seconds.
2.4.2

Conduction Current
This current can flow if the insulation has cracks, cuts, or pinholes

and has absorbed moisture, and some contamination is present. This current is
constant with time. With modern insulation, like epoxy-mica or film
insulation, this current is usually zero. If this current is significant, it indicates
that the winding insulation has a problem.

35

2.4.3

Surface Leakage Current

This is a constant DC current that flows over the surface of the

insulation caused by oil or moisture mixed with dust, dirt, fly ash, chemicals
on the surface of the windings. If this current is large, it is likely that
contamination-induced deterioration (electrical tracking) may occur.
2.4.4

Absorption Current
The absorption current is due to reorientation of certain types of

polar molecules which many insulating materials contain. Polar molecules

have an internal electric field due to the distribution of electrons within the
molecule. Water molecules are polar. When an electric field is applied across
water, the H2O molecules all align. The energy required to align the
molecules comes from the current in the DC voltage supply.
Once the molecules are all aligned, the current stops. This current is
the polarization current. There are many polar molecules in asphalt, mica,
polyester and epoxy. Experience shows that after a DC electric field is applied
to such materials, the absorption current is first relatively high, and decays to
zero after about 10 minutes. In all practical respects, the absorption current
behaves like an RC circuit with a long time constant. It is merely a property of
the insulation materials.
The total current It is the sum of all these current components.
Unfortunately, we cannot directly measure any of these components of
currents. The Polarization Index (PI) was developed to make interpretation
less sensitive to temperature. If we assume that Rl0 and R1 were measured
with the winding at the same temperature, which is usually very reasonable to
assume, then the temperature correction factor will be the same for both R1
and R10, and will be ratioed out (Lamarre and David 2006).

36

The PI effectively allows us to use the absorption current as a

yardstick to see if the leakage and conduction currents are excessive. If the PI
value is about one, the leakage and conduction currents are high, so that the
electrical tracking occurs in the insulation. If the above currents are low when
compared with the absorption current after 1 minute and then the PI is greater
than 2, the insulation is not being affected.
The IR and PI can be measured with a high-voltage DC supply and
a sensitive ammeter of nano ampere range. The DC supply must have a wellregulated voltage; otherwise a capacitive charging current will flow
(= C dV/dt).
There

are

several

special-purpose

megohmeters

available

commercially. Modern instruments can apply voltages up to and exceeding

10 kV DC, and measure resistances higher than 100 G

.The IR/PI test will

depend strongly on humidity, if the results are poor it may be necessary to

heat the winding for hours or days to dry off the moisture that has condensed
on it. After each IR and PI test, the winding should be grounded for at least
four times as long as the voltage was applied, i.e., 40 minutes. Premature
removal of the ground will cause a high voltage to reappear, due to the time it
takes for the molecules to again become random in orientation, and for the
space charge to dissipate. Thus, a shock hazard exists.
2.4.5

Test Results and Inference

A 11 kV, 6 MW generator stator winding, having class F insulation

was tested. Results of the IR/PI tests and DC leakage current measurements
obtained on the three phases of the stator winding are presented in Table 2.2.

37

Table 2.2

Measurement of IR, PI and Leakage Current of 6 MW

Generator

Phases

IR (M

PI

DC leakage current ( A)

7860

2.63

1.0

9750

4.92

0.9

9470

4.59

1.0

Referring to the Table 2.2 the IR and PI values lie in the normal
range expected for a class F insulation machine. For clean and dry class F
insulation, the PI is higher than the minimum permissible value of 2. The DC
leakage current values are also quite low. These results indicate that the stator
winding is clean and dry.
2.5

CAPACITANCE TEST
Measurement of the winding capacitance can indicate thermal

deterioration or saturation of the insulation by moisture. This test is most

useful on form-wound motor stator, or very large direct-water-cooled
generator stators. Capacitance measurements are also made during
manufacturing.
If a form-wound stator deteriorates due to long-term overheating,
the groundwall insulation layers delaminate. The result is that the groundwall
now contains some gas, usually air. The dielectric constant of air is one,
whereas the dielectric constant of epoxy-mica is about 4. As the percentage of
gas within the ground-wall increases as a result of thermal deterioration, the
average dielectric constant decreases. The coil in a slot may be approximated
by a parallel plate capacitor. Its capacitance will decrease, as the dielectric
constant decreases. The dielectric constant of water is 80 and it increases the

38

capacitance. The end-winding of a stator is polluted with conductive

contaminant. This effectively increases the surface area A, of the capacitor
plate and winding contamination will increase the capacitance. If the
capacitance is unchanged over the years, then little deterioration will occur. If
the capacitance is found decreasing, then the winding is likely to have a
thermal deterioration, and when the capacitance is increasing, then the
winding has absorbed moisture.
The capacitance tests are generally performed with commercial
capacitance bridges having a precision better than 0.1 percent. Since the
amount of gas or moisture within the groundwall is usually a small percentage
of the normal insulation, the change in capacitance over the years is also very
small, even for very significant deterioration.
The capacitance of the entire phase or winding can be measured in
a global measurement. This version of the test determines the overall
insulation condition. In addition, for the specific problem of stator winding
water leaks, the local capacitance of a portion of a stator bar is to be
measured.
For the capacitance measurement the key for interpretation is the
trend. A significant amount of thermal deterioration will result in only one
percent drop in capacitance over the years. If the winding has been soaked
with water there will be a 5 percent increase in capacitance. If the entire
winding is affected, then the capacitance test is more likely to detect it.
Coil manufacturers often use the capacitance test to monitor the
impregnation and curing process. Impregnating resins have a very high
dielectric constant when in the liquid state. As they cure, the dielectric
constant asymptotically reaches it low value of about 4. When a coil is first
impregnated with the liquid resin, its capacitance increases as the resin

39

replaces the air between the mica-paper tape layers. It reaches a high steadystate value when complete impregnation is achieved. The capacitance starts to
decrease as it cures. With experience, manufacturers can define the optimum
cure time by monitoring the initial increase and then decrease of the
capacitance.
2.5.1

Capacitance Tip-Up Test

The capacitance tip-up test is a variation of the capacitance

measurement on complete windings (Emery 2002). Thermal deterioration,

load cycling, and poor impregnation methods can result in air pockets within
the ground wall insulation. The electrical stress within the voids exceeding
3 kV per mm, a partial discharge will occur within these voids. The ionized
gas has sufficiently high conductivity and increases the capacitance. There are
thousands of voids undergoing PD and then there will be a noticeable increase
in the capacitance.
By measuring the capacitance at high voltage and subtracting from
this the capacitance at low voltage, the result is the increase in capacitance
due to PD activity (Wolmarans and Geldenhuys 1991). By taking the
difference, one can estimate the capacitance of the voids. The larger the void
capacitance, the more deterioration within the groundwall and presumably,
the closer the winding is to failure. The HV Schering bridge with the accuracy
better than 0.1 percent is most commonly used. The low-voltage capacitance
(Clv) is first measured at about 20 percent of the rated voltage and the high
voltage capacitance is measured with 58 percent of rated voltage.
The capacitance tip-up is given in equation (2.9).
C

Chv Clv
Clv

(2.9)

It is usually expressed in percent rather than in Farads ( C < 1%).

40

2.5.2

Test Results and Inference

A 11 kV, 6 MW generator stator winding having class F insulation

was tested. The capacitance tip-up test results are shown in the Table 2.3.
Table 2.3 Measurement of Capacitance Tip-Up for 6MW Machine
Phases

C (%)

0.25

0.26

0.27

Referring to the Table 2.3, the capacitance tip-up value lies in the
normal acceptable range.
2.6

DISSIPATION FACTOR TEST

Dissipation factor (Tan ) provides an indication of the dielectric

losses within the insulation. Due to thermal deterioration and moisture

absorption, the dielectric loss increases. This test is relevant for stator
windings only and is usually applied only to form-wound stators. Dielectric
loss is a property of any insulating material and its chemical composition. An
increase in dielectric loss in a winding over the years may indicate insulation
aging due to overheating or radiation. Similarly, if a winding has been soaked
with water, the dielectric loss will increase. This occurs since the H2O
molecules are polar. The Dissipation Factor (DF) is measured with a balanced
bridge-type instrument, which can easily achieve 0.01 percent accuracy.
Typical DFs are about 0.5 percent for modern epoxy and polyester
impregnated insulations. The DF can be 3 to 5 percent for asphaltic mica
windings. If DF is measured every few years regularly, and if it remains

41

constant over time, then it indicates that there is no thermal aging or gross
contamination of the winding. If it increases over time, then it indicates that
insulation overheating is occurring or the winding is becoming more
contaminated by moisture or partly conductive contaminants (IEC std 608941987).
If the Dissipation Factor (DF) has increased by 1 percent or more
from the initial value, deterioration is significant. The loss angle
representation is shown in the Figure 2.6. If the capacitance and DF are
measured at the same time, and if the Capacitance (C) decreases with the
increase in DF, it strongly indicates general thermal deterioration. If both C
and DF increase over time, then it indicates that the winding is contaminated
or has absorbed moisture.

I
I
Ic
- Loss angle
- Phase angle

Ir

Figure 2.6 Loss Angle Representation Graph

The dissipation factor test is used as a process monitor for the
impregnation process (Sang et al 2005). As the groundwall is impregnated,
the DF will increase, since liquid resin has a higher DF than the air. As the

42

coil cures, the DF will decrease to its steady final level, since the DF of liquid
resin is higher than that of the DF of cured epoxy or polyester.
2.6.1

Dissipation Factor Tip-Up Test

The dissipation factor tip-up is a complement of capacitance tip-up,

relevant to form-wound stator coils rated at 2300 volts and above. It is an

indirect way of determining partial discharges occurring in a high-voltage
stator winding. PD is a symptom of many high-voltage winding insulation
deterioration mechanisms. The tip-up test can indicate many failure processes.
The tip-up test is widely used by stator coil and bar manufacturers, as a
quality control test to ensure proper impregnation by epoxy and polyester
during coil manufacture (Emery 2002).
At low voltages, the DF is not dependent on voltage. As the AC
voltage is increased across the insulation in a form-wound coil, and if voids
are present within the groundwall, then at some voltage, partial discharges
will occur, producing heat, light, and sound, which consume energy from the
power supply. In a delaminated coil, as the voltage increases and PD starts to
occur, the DF will increase above the normal level due to dielectric loss, since
the PD constitutes an additional loss component in the insulation. The greater
the increase in DF, more energy is consumed by the partial discharge.
In the tip-up test, the DF is measured at about 20 percent of the
rated line-to-ground voltage of the stator and at the rated line-to-ground
voltage (Wolmarans and Geldenhuys1991). The tip-up is calculated as given
in equation (2.10).
Tip Up

DFhv DFlv

(2.10)

43

The higher the tip-up, the greater is the energy consumed by PD. The
DF may be recorded at several voltage levels and the tip-up as a function of
voltage may be plotted as shown in the Figure 2.7. The voltage at which PD
starts is sometimes measurable. The DF is measured in percentage, and hence
the tip-up is also represented in percentage. The tip-up on windings rated
greater than 6 kV is usually significant.

Deteriorated

Sound

Voltage in kV

Figure 2.7

In tests on machines, it is important to test a few coils as possible,

at a time, since this will increase sensitivity. Each phase should be tested
separately while the other two phases are grounded. The winding should be
partitioned into parallels or coil groups to gain maximum sensitivity. At low
voltage, the silicon carbide is essentially a very high resistance coating, and
no current flows through it. Thus, there is no power loss in the coating. At
rated voltage viz. 6 kV or above, the silicon carbide coating will have a
relatively low resistance. Capacitive charging currents flow through the
insulation and then through the coating. The charging currents flowing
through the resistance of the coating produce an I2R loss in the coating
(IEEE std 286-2000).

44

2.6.2

Test Results and Inference

Tables

normal acceptable range. These results indicate that the dielectric losses and
void contents in the stator insulation are quite low.
Table 2.4 Dissipation Factor Measurement on 11 kV Machine

Phase

Grounded
Terminals

Applied

Leakage

Voltage in
kV

Current in
mA

in nF

Factor

4.40

135.7

99.73

2.458

6.60

202.9

99.85

2.586

8.80

269.5

99.49

2.923

11.0

339.5

100.3

3.315

4.40

137.8

101.0

2.421

6.60

205.8

100.9

2.549

8.80

275.5

101.3

2.856

11.0

341.1

100.5

3.372

4.40

136.1

99.97

2.462

6.60

203.6

100.1

2.592

8.80

272.1

100.2

2.898

11.0

340.7

100.4

3.343

Y and B

B and R

R and Y

Table 2.5

Capacitance Dissipation

Tip-Up Test for 11 kV, 6MW Generator

Phase

-up (%)

0.273

0.291

0.282

45

2.7

PARTIAL DISCHARGE TEST

Form-wound stator winding, rated at 2300 V and above, at rated

line to ground voltage, the pulse currents resulting from PD, is measured.
Thus, any failure process that creates PD as a symptom can be detected with
this method. When a partial discharge pulse occurs, there is a very fast flow
of electrons constituting i = dq/dt from one side of the gas filled void to the
other side. The pulse has a very short duration, typically a few nanoseconds.
Also there will be a flow of the heavy positive ions in the opposite direction,
moving slowly with large transition time. The magnitude of the current pulse
due to ions is negligible.
Any device, sensitive to high frequencies can detect the PD pulse
currents. In the off-line PD test, the most common means of detecting the PD
currents is to use a high-voltage capacitor connected to the stator terminal.
Typical capacitances are 80 pF to 1000 pF. The capacitor offers very high
impedance to the high AC voltage with power frequency and very low
impedance to the high-frequency PD pulse currents (IEEE std 1434, 2000).
This can be displayed on an oscilloscope, frequency spectrum
analyzer, or other display devices having band width in MHz range. PD pulse
is proportional to the size of the void in which the PD occurred. Higher the
magnitude of the PD pulse, the larger is the defect. The advantage of the PD
test is that one concentrates on the larger pulses and ignores the smaller
pulses.
Unlike capacitance or power factor tip-up tests, which are a
measure of the total PD activity, the PD test enables the measurement of the
biggest defects. Since failure is likely to originate at the biggest defects and
not at the smaller defects, the PD test can indicate the condition of the

46

winding at its most deteriorated portion. The measurement of the PD pulses is

shown in Figure 2.8.

Figure 2.8 Measurement of PD Pulse Current

The off-line PD test requires a power supply to energize the

winding to at least rated phase-to-ground voltage. For large generator stators,
a conventional or resonant transformer rated at 20 to 40 kVA may be needed.
The PD test is performed at the machine terminals, energizing one phase at a
time, grounding the other two phases. In the off-line PD test, the applied
voltage is raised gradually while monitoring the PD pulses on an oscilloscope
screen. The voltage at which the PD is first detected is called the Discharge
Inception Voltage (DIV). The voltage then is raised to normal line-to-ground
operating voltage. The winding should remain energized for the soak time of
10 to 15 minutes at this voltage, and then the PD is recorded. Now the voltage
is then gradually lowered. The voltage at which the PD is not detectable is
called the Discharge Extinction Voltage (DEV), usually lower than the DIV is
measured. It is desirable to have the DIV and DEV as high as possible
(Campbell and Stone 2000).

47

For machines rated at 6 kV or more, the maximum test voltage is

normally the rated line-to-ground voltage. A test at this voltage will usually
detect deterioration, years before an in-service failure is likely. For machines
rated at 2300-4100 V, a test at rated voltage may not produce significant PD,
even in severely deteriorated stator insulation. This is because there may be
insufficient electric stress within the defects to achieve the 3 kV/mm needed
in atmospheric air to cause PD.
2.7.1

Test Results and Inference

Unlike the tip-up test, which produces a single number representing

the total PD activity, direct measurement of the PD produces several results.

The key measure is the peak PD magnitude Qm, i.e., the magnitude of the
highest PD pulse. This can be measured in several units like pico Coulombs
(pC), millivolts (mV), milliamps (mA) and decibel (dB). The detected PD
magnitude of a PD pulse within the winding but measured at the stator
terminals depends on the size of the defect, the capacitance of the winding,
the inductance between the PD site and the PD detector (Stone 1998).
The off-line PD test is a comparison test (Zhu et al 2005). One can
determine which phase has the highest Qm i.e. the greatest deterioration. One
can also compare several similar machines to see which has the highest PD or
the lowest DIV or DEV. Finally, one can compare the PD from the same
stator over time. If the PD doubles every 6 months, then the rate of
deterioration is increasing (IEEE 1434-2000). Direct measurement of the PD
indicates how widespread the PD is. As many as 10,000 PD pulses may occur
per second in a stator winding. It seems that a single defect only produces at
most one or two PD pulses per half AC cycle.
Thus, if only a few hundred PD pulses are occurring per second,
then there are only a few PD sites in the winding and the deterioration is

48

localized. If there are 10,000 PD pulses per second, then there are thousands
of PD sites and the deterioration is widespread. The pulse count rate can be
easily measured with a pulse magnitude analyzer, which is incorporated into
most modern commercial PD analyzers (Kurtz et al 1984). If there is
dominant deterioration in a winding, the PD test can sometimes give the
approximate location of the deterioration within the groundwall. Both positive
and negative PD pulses are created. If the positive PD pulses are larger than
negative PD pulses, then the PD is occurring on the surface of the coil. If the
negative PD is predominant, then the PD is occurring at the copper. If there is
no polarity predominance, then the PD may be between the groundwall
insulation layers.
Table 2.6 PD Test Results of 6 MW Generator
PD inception
voltage (kV)

PD magnitude at
5 kV (pC)

3.78

600

3.83

600

3.83

600

Phases

Referring to the Table 2.6 the PD magnitude lies in the normal

acceptable range.
2.8

SURGE COMPARISON TEST

None of the tests discussed above directly measure the integrity of

the turn insulation in form-wound or random-wound stator windings. The

stator voltage surge test directly measures the integrity of the turn insulation
by applying a relatively high voltage surge between the turns. This test is a

49

hipot test for the turn insulation. The insulation, may fail requiring a repair,
coil replacement, or rewind.
Voltage surges occur from Inverter Fed Drive Motors (IFDs) and
faults in the power system. These fast risetime surges result in a non uniform
voltage distribution across the turns in the stator winding. If the rise time is
short and, the surge voltage is high, the turn insulation becomes weak and
punctures. This test is analogous to the AC and DC hipot tests. The surge test
is a destructive test. If the turn insulation fails, then the assumption is that the
stator would fail in service due to transients (Stefan et al 2008).
If the winding does not puncture, then the turn insulation will
survive any likely surge over the next few years. The main difficulty with the
surge test is determining when the turn insulation puncture has occurred.
A turn-to-turn puncture in a winding does not cause a huge increase in current
from the power supply. In fact, if there are 50 turns between the phase
terminal and neutral, the failure of one turn will only slightly reduce the
inductive impedance of the winding, since the impedance of only one turn has
been eliminated. Thus the other 49 turns can continue to impede current flow,
and the circuit breaker does not trip.
In the surge test, turn failure is detected by means of the change in
resonant frequency caused by shorting out one turn. The inductor is the
inductance of one phase of a stator winding, or two phases in series. A highvoltage capacitor within the surge tester is charged from a high-voltage DC
supply via the winding inductance. Once the capacitor is charged to the
desired voltage, the switch is closed. The energy stored in the capacitor then
oscillates back and forth with the winding inductance. If there is no turn fault,
there will be a fixed frequency of oscillation. If a turn fault occurs together
with weak turn insulation, the inductance of the winding will decrease and the
resonant frequency will increase. The increase in frequency being small, it is

50

difficult to detect. Modern surge testers digitally capture the resonant

waveform at low voltage, where the turn insulation is still intact. The surge
voltage is gradually increased by raising the voltage and triggering the switch
after the capacitor has charged up (once per second or once per 50 or 60 Hz
cycle). If a change in the waveform is noted above a certain voltage, then turn
insulation puncture has occurred.
It is easy to detect turn insulation failure on individual coils, since
the shorting of one turn will have a much larger impact on the total inductance
of a coil, thus drastically changing the waveform. Machine manufacturers and
rewind companies use individual coil surge testing to check the quality of the
turn insulation (IEEE std 522- 992).
Such testing is best done after the coils are wound, wedged, and
braced, since by then they have been exposed to all the mechanical handling
and stresses associated with the winding process. As a quality check, the
surge test is done prior to inserting coils into the slot. Ground faults are easily
detected by a surge test, since the waveform collapses. Surge testing is also
useful to identify wrong connections in the winding. As an acceptance test,
the surge is recommended to have a risetime of 100 ns and a maximum
magnitude of 3.5 per unit, where l per unit is the peak line-to-ground rated
voltage. For a maintenance test performed after the winding has been service,
the surge should have the same risetime, but reach only 2.6 per unit. Voltages
higher than these maximum should not be applied to the stator winding,
otherwise there is a significant risk that good turn insulation will fail
unnecessarily.
National Electrical Manufacturing Association (NEMA) requires
that Inverter Fed Drive Motors should withstand 100 ns risetime surges at
3.7 per unit. The longer the surge risetime, the lower is the voltage applied
across the turn insulation. Thus, longer rise-time surges are not as severe as

51

short risetime surges. The surge voltage is gradually increased to the

maximum recommended test voltage. If the waveform changes on the
oscilloscope, then the turn insulation has likely been punctured. If the winding
is form wound, the failed coil will have to be located and isolated. The coils
must be separated from one another and tested separately until the faulted coil
is found. If a turn puncture has occurred, the punctured turn insulation will
break down again, allowing power-frequency currents to flow, rapidly leading
to groundwall failure.
2.8.1

Interpretation
The surge test and the partial discharge test are combined, and then

it may be possible to detect significant voids between the turns, before actual
puncture occurs. This requires a special PD detector, since conventional PD
detectors will be damaged by the high-voltage surges. Although the surge test
stated above to be the only test that directly determines the condition of the
turn insulation, (Gupta et al 1987) the IR/PI, capacitance, and dissipation
factor tests discussed above will also indicate the condition of the turn
insulation.
The Capacitance, Tan delta and Partial Discharge measurements are
adequate for testing winding insulation to ground but not the insulation
between turns. Surge testing is accurate methods of identifying inter turn
faults. Surge voltage is applied on a winding consisting of a number of coils
which in turn consist of many turns. A ringing pattern is seen on the Cathode
Ray Tube (CRT). The fast raising pulse spreads along the coil and creates a
voltage gradient along the turns. Since the three phases are wound identically,
comparison of all the phases will show the same single pattern as shown in
Figure 2.9(a). In case of faults in any one of the phases, the wave pattern gets
separated indicating a fault as shown in Figure 2.9(b).

52

(a)

(b)

Figure 2.9 Surge Comparison Test

2.9

AC HIGH POTENTIAL TEST

In AC High Potential (Hipot) test, power-frequency (50 or 60 Hz)

voltage is used. The AC voltage dropped across each component in the

groundwall or in the end-winding depends on the capacitance (dielectric
constant) of each component (Gupta et al 1995, 2001).
For modern windings, the AC hipot test yields an electric stress
distribution that is the same as that which occurs during normal operation.
Consequently, the AC hipot test is more likely to find defects that could result
in an in-service stator failure if a phase-to-ground fault occurs in the power
system, causing an overvoltage in the unfaulted phases. Figure 2.10 shows
AC Hipot test set up. NEMA MG1 and IEC 60034 define the AC acceptance
hipot level as (2E + 1) kV, where E is the rated rms phase-to-phase voltage of
the stator. IEEE 56 recommends the AC maintenance hipot be 1.25 to 1.5 E.
The AC hipot for a 4.0 kV motor that has been in service, is about 6 kV rms,
applied between the copper conductor and the stator core. A winding usually
either passes the test or it fails because of a puncture. A one minute AC hipot
test at 1.5 E is equivalent to about 235 hours or 10 days of operation at normal

53

operating voltage. Therefore, the life is not significantly reduced by a hipot

test if the expected life is about 30 years.

Figure 2.10 AC High Voltage Test on Form Wound Coil

2.9.1

Test Results and Inference

An 11 kV generator stator winding having class F insulation was

tested. The Hipot test results are shown in the Table 2.7.
Table 2.7 Measurements of Hipot Test Parameters

Phase

Grounded
Phase

Applied Voltage Leakage current

in kV
in mA

Y and B

23

1181

B and R

23

1185

R and Y

23

1177

54

Referring to the Table 2.7 the stator winding passes the AC hipot
test. If the winding fails, as determined by the power supply circuit breaker
tripping or an observed insulation puncture, then repairing of coil or
replacement of winding is required.
2.10

CONCLUSION
Offline testing is the main approach used for insulation testing of

the stator winding and the stator insulation problems prove to be drastic as far
as failure mechanisms are concerned.
One of the restrictions in using online PD monitors for the
insulation monitoring is the capital cost in the deployment of the dedicated
sensors and data processing hardware. In addition, as there are symptoms
which are the causes of insulation failure other than corona, relying on PD
monitoring alone may not be fruitful for reliable assessments of the overall
insulation condition.
So tests like, Insulation resistance test, Polarization Index test,
Leakage current test, Capacitance and Capacitance tip-up test, Dissipation
factor and Dissipation factor tip-up test, Hipot test and Surge comparison test
are conducted for the precise determination of the insulation performance.