Off-Line Insulation Testing and Diagnosis of HV Machines
Off-Line Insulation Testing and Diagnosis of HV Machines
Off-Line Insulation Testing and Diagnosis of HV Machines
Example 2
Motor data: 11500V 20000 HP. It has been periodically tested since 2000.
The picture below shows the trend of PD activity with time. It is clear to see
the increase in PD activity (Fig.1).
condition.
The leakage current depends on ambient moisture and surface 30000
contamination; therefore, this test detects moisture or dirt on the
windings. If a DC test is used, the test voltages should be those 25000
values recommended for AC testing multiplied by a factor of 1.7.
20000
2.7. AC high potential
Q,pC
The AC high potential test is a pass/fail test used to evaluate the L1 15000
condition of insulation and it should be noted that each time the L2
L3
AC high potential test is performed; the potential consequence is 10000
a winding failure.
New and rewound machines should have an AC high potential 5000
test of twice-rated terminal voltage plus 1 kV applied to them.
When the machine is repaired and the winding is not involved, 0
a reduced AC over voltage test may be performed prior to the 2012 2010 2008 2006 2004 2002 2000
commencement of works and again after completion of works. Years
In the case of failed windings, where insulation is replaced or a
partial rewind is performed, a reduced voltage final test may be
used. Fig.
Fig.11PD
PDactivity
activity trend
trendwith
with time
time
3. Test methods and test results The picture below also shows the trend with time of DDF f (U) (Fig.2).
The picture below also shows the trend with time of DDF f (U)
3.1. Insulation resistance and polarization index (Fig.2).
measurement.
For Insulation resistance and polarization index measurement,
CEL uses the Megger S1-5010 Insulation Tester. For rotating 2.4
L1, 2000
machines with a rated voltage of 3.3 kV, the applied test voltage
L2, 2000 2.2
is 2500 V. For rotating machines with rated voltages above 6.6 kV
the applied test voltage is 5000V. L3, 2000
2.0
Resistance and PI tests are performed simultaneously in an L1, 2006
automatic mode. L2, 2006
1.8
DDF, %
L3, 2006
Example 1 L1, 2006 1.6
Motor data: 6600V 1250 HP. The motor was put into operation L2, 2009
in 2000. 1.4
L2, 2009
During periodic testing, very low values of insulation resistance
and high value of DDF were obtained. It should be noted that L1, 2011 1.2
the motors are located near the sea and the heating elements of L2, 2011
the motors worked properly. After heating the motors from an L3, 2011 1.0
external source, the value of insulation resistance increased, but 7 5 3 1
U, kV
immediately after turning off the heating for a short time dropped
to a low value. Based on this data, impregnation of insulation was Fig. 2 DDF f (U) trend with time
Fig. 2 DDF f (U) trend with time
recommended. The one motor was removed for repair. During
inspection in a repair shop, it was discovered that there was in
fact weak impregnation of the winding insulation. ImpregnationThere is There is a direct correlation between PD activity and DDF.
a direct correlation between PD activity and DDF.
was carried out and additional heating elements were installedFor PDFor PD localization,
localization, CEL uses CELtheuses the ultrasonic
ultrasonic sensorsensor
UE 200 UE(UE200system)
(UE and
on the motor. There was a significant improvement in insulationthe UV detectorand
system) the UV
Daycor detector
(OFIL Daycor
Ltd). This (OFIL
test was Ltd). This
performed test was
on open stator.
test values and the problem was solved. A similar repair was performed on open stator.
performed on the remaining motors of this type. Example 3
Example 3
3.2. Generator
Dissipation factor and partial discharge measurement. Generator data: 22500
data: 22500 KVA, 11000
KVA, 11000 V V
For DDF and PD measurements, CEL uses the Mobile InsulationDuringDuring periodicperiodic
testing, testing, highoflevels
high levels partialof partial were
discharge discharge werein one
recorded
Diagnosis & Analyzing System MIDAS 2881 ( Haefely Testof the recorded phases. An in additional
one of the PDphases.
test was Anrecommended
additional PD withtest
an was
ultrasonic
AG ), and the partial discharge detector ICM system ( Powersensor.recommended witha an
During this test ultrasonic
strong sensor.
discharge Duringwinding
between this testand
a strong
stator core
Diagnostix System GmbH). A resonance reactor is used in parallelwas discovered. The reason
discharge between for and
winding the stator
problem corewas
wasthe destruction
discovered. Theof the
for compensation of high capacitance in large rotating machines. semiconductor
reason forlayer (coronawas
the problem protection).
the destructionRewinding of the stator was
of the semiconductor
The software allows the test engineer to simultaneously performrecommended. At the pictures
layer (corona below shows
protection). PD pattern
Rewinding of the(Fig.stator
3). was
both measurement and record all necessary parameters. recommended. At the pictures below shows PD pattern (Fig. 3).
Example 2
Motor data: 11500V 20000 HP. It has been periodically tested
since 2000.
The picture below shows the trend of PD activity with time. It is
clear to see the increase in PD activity (Fig.1).
L1 L2 L3
Fig. 3 PD pattern
L1 L2 L3
Fig. 3 PD pattern
Fig. 3 PD pattern
Fig. 5 Corona detection by UV camera
The picture below, taken by a burrowscope, shows destruction of 3.3. Applied high voltage testing
The picture
the corona protection layer ofbelow,
the statortaken
coil (Fig.by
4). a burrowscope, shows destruction of the corona
protection layer of the stator coil (Fig. 4). Mainly in CEL,high
3.3. Applied allvoltage
HV testing
testingis performed at AC up to 50 KV,
by using an HV compensation trans-reactor, which allows the use
Mainly in CEL, all HV testing is performed at AC up to 50 KV, by using an HV
of a low consumption power supply. The test circuit includes:
compensation trans-reactor, which allows the use of a low consumption power
regulating
supply. The testtransformer (1), regulating
circuit includes: trans-reactor (2), (1),
transformer testtrans-reactor
object (3),(2),
voltmeter
test (4),
object (3), ampere-meter
voltmeter (5) (Fig.6).
(4), ampere-meter (5) (Fig.6).
InInthe
thecase
caseofofspecial
specialrequirements fromfrom
the client, it is possible to perform
Generator data: 5 Fig. KVA,
133750
Example 4 Destruction
11500 Vof corona protection requirements the client,
DC high voltage testing for leakage current measurement. For this propose
it is possible to
Generator
In the course data: 133750
of periodic KVA,Corona
testing, 11500 V tests were carried out toCEL perform
locate DC high voltage testing for leakage
uses the DC Dielectric Test Set 70 kV (Biddle).
current measurement.
In the course of periodicwindings
testing, Corona tests were carried During
out For this propose CEL uses the DC Dielectric Test Set 70 kV
corona discharge on generator at nominal voltage. Corona
Example 5to locate corona discharge on generator windings atwas
nominal (Biddle).
3.4. Special test
testing, points of discharge were found. The generator repaired by the
voltage. During Corona testing, points of discharge were found.
manufacturer.
The
Isolation
generator
was
was
strengthened
repaired
with resin impregnated glass
byVthe manufacturer. Isolation was
The tape.
3.4. Partial Special
Dischargetesttest, in a traditional sense, performed at the phase
Generator
Corona data:
discharge 133750 KVA,
disappeared 11500
after repairing. The Corona
pictures voltage, does not allow testing of the insulation between phases in full, as the
below shows PD
In theup strengthened
course of of with
periodic resin impregnated
testing,(Fig.
Corona glass tape. discharge The
were carried out toapplied Partial
voltageDischarge
is 1.7 timestest,less
in a than
traditional
the truesense,
voltage.performed at theline
By applying
picked with aid
disappeared aafter
UV repairing.
camera 5). tests
The pictures below shows PD picked
locate
phase voltage,
voltage, does not
phase to ground allowistesting
insulation of to
subjected theadditional
insulation between
stress, because
corona discharge on generator windings at nominal voltage. During Corona
the applied
up with aid of a UV camera (Fig. 5). phases in voltage
full, asisthe
1.7applied
times larger than is
voltage the1.7
truetimes
voltagelessand thatthe
than increases
true
testing, points of discharge were found. The generator was repaired the by the
risk
voltage.
of insulation breakdown.
By applying line voltage, phase to ground insulation is
manufacturer. Isolation was strengthened with resin impregnated glass Therefore,
tape.for to our laboratory has developed and implemented a three-phase
subjected
scheme additional
measuring stress,
partial becausewhich
discharge, the applied
approximatesvoltagetheistesting
1.7
Corona discharge disappeared after repairing. The pictures below shows times PDlarger
conditions to bethan the true
as close voltagetoand
as possible realthat increases
working the risk
conditions. This ofKIT
picked up with aid of a UV camera (Fig. 5). includes
insulationa three phase regulating transformer and 3 HV trans-reactors with an
breakdown.
adjustable core gap, to regulate the inductance. This scheme allows the
Therefore,
application of our laboratory
three has developed
phase voltage andmachine
to the rotating implemented a three-
with compensation
phase
for scheme of
the capacity forthe
measuring partial
test object. discharge,
This scheme alsowhich
allowsapproximates
the testing of 3-
Phases simultaneously,
the testing conditions thus
toshortening
be as close the overall testingto
as possible time.
real working
The test circuit includes: regulating transformer (1), trans-reactor (2), test
conditions. This
object (3) (Fig.7).
KIT includes a three phase regulating transformer
and 3 HV trans-reactors with an adjustable core gap, to regulate
the inductance. This scheme allows the application of three phase
voltage to the rotating machine with compensation for the capacity
of the test object. This scheme also allows the testing of 3-Phases
simultaneously, thus shortening the overall testing time.
The test circuit includes: regulating transformer (1), trans-reactor
(2), test object (3) (Fig.7).