Electrical condition monitoring part 2

1. Condition On Line MonitoringCondition On-Line Monitoring
October 19, 2011 – Night 2, g
Dominik Pieniazek, P.E. – VI Engineering, LLC

2. Outline
Day 2
• Power Transformer Monitoringg
– Partial Discharge
– Moisture in Oil
– Dissolved Gas
– Load Tap Changer
– Temperature (Oil and Winding)
– Load Trending
Bushing– Bushing
– Ancillary Equipment
• Communications
• Special Considerations• Special Considerations
• Advantages

3. Recap from Night 1
• PD is a very fast electrical spark (nsec)
• PD creates phenomena that can be measured:
• Electromagnetic pulse
• Light emission
• Ultrasound wave
• Electro-chemical reactions (Ozone)

4. Recap from Night 1
• Challenges in measuring PD
• Signal attenuation
• Noise
• Right frequency band

5. Recap from Night 1

6. Recap from Night 1
• Insulation Degradation
Insulation isolates HV from ground and other
dconductors.
• Electrical stresses
• Thermal stresses
• Environmental stresses

7. Transformer On-Line Monitoring
•The average age of the transformer fleet is increasing.
•Organizational changes reducing staffing levels resulting ing g g g g
fewer transformer and asset technical experts.
•Electrical systems are operated with lower margins.
•The transformer is the single most costly item in a substation•The transformer is the single most costly item in a substation.
•Replacement transformer lead times can exceed one-year.
•Financial pressure to reduce both capital and maintenance
di h l d i i d l d f dexpenditures have resulted in increased load factors and
reduced (if any)spares.
•New technologies and diagnostic techniques are making on-lineg g q g
monitoring practical and effective.

8. Power Transformer Monitoring
Why Consider On-Line Monitoring?
•Transformer faults detected in their infancy - Avoid costly
unplanned outages
f i i d f l i i i•Transformer output optimized safely, maximizing asset
•Transformer ageing can be calculated.
•Type of fault can be classified from results.yp
The utility industry does not have a good grasp on the
expected life of a power transformer Insurance companiesexpected life of a power transformer. Insurance companies
predict an expected life of 35 years,

9. Power Transformer Monitoring
Industry Experience (obtained from EPRI July 2006 Conference)
The utility industry has been relatively active in sharing information
about failures, but very in-active when it comes to sharing failure
statistics.statistics.
The industry knows about specific catastrophic failure events but
knows very little about major failures that do not cause widespreadknows very little about major failures that do not cause widespread
outages, minor failures, failure rates and retirements;
This information is critical to develop accurate statistical transformer
life models.

10. Power Transformer Monitoring
Hartford Steam Boiler Inspection and
Insurance Company data based on data
Cause
Percent of
Reported Failures
collected by the International
Association of Engineering Insurers
(IMIA) for the years 1997 through
Insulation failure 26%
Manufacturing failure 24%
Unknown 16%
(IMIA) for the years 1997 through
2001 involving 94 reported failures.
Loose connections 7%
Improper maintenance 5%
Overloading 5%
Oil contamination 4%
Line surges 4%
Fire/explosions 3%
Li ht i 3%Lightning 3%
Floods 2%
Moisture 1%
Total 100%Total 100%

11. Power Transformer Monitoring
Cause of Transformer Failure by Impacted System.
Failure Distribution byImpacted System
Unknown (45%)
Dielectric(38%to
45%)
Containment (3%-
6%) Current Carying
(7%)
Mechanical (0%-
4%)

12. Power Transformer Monitoring
•Partial Discharge
M i t i Oil•Moisture in Oil
•Dissolved Gas
•Temperature (Oil and Winding)
•Load Trending
•Bushing
•Load Tap ChangerLoad Tap Changer
•Ancillary Equipment

13. Partial Discharge
•Sources of Partial Discharge
•Corona discharges in uncontaminated oilg
•Corona discharges in moisture contaminated oil
•Corona discharges in metal contaminated oil
•Paper/pressboard dischargesPaper/pressboard discharges
•Surface Discharges
•Air-Oil Discharges
•PD is a Consequence
PD occurs as a result of defects within the insulation
•PD is a Cause
Insulation degradation increases as a result of PDInsulation degradation increases as a result of PD

14. Partial Discharge
Source Electrical
Detection
Acoustic
Detection
Remarks
PD on the Outside of Yes Yes Best use of acoustic detector locationPD on the Outside of
Winding
Yes Yes Best use of acoustic detector location
PD Within Winding Yes Unlikely Strong acoustic attenuation inside the
winding
PD between Winding
and Core
Yes Difficult Acoustic signal reflection at the core
required
Arcing/Tracking of the
Oil Surface
Yes Yes
Oil Surface
Arcing/Tracking on
Bushing Surface in the
Oil
Yes Yes
PD in the Bushing Yes Possible Safety concerns with sensor placement
PD in the NLTC Yes Yes
PD in the LTC Yes Yes

15. Partial Discharge – Acoustic Method
Acoustic emissions (AE) are transient elastic waves in theAcoustic emissions (AE) are transient elastic waves in the
range of ultrasound, usually between 20 kHz and 1 MHz,
generated by the rapid release of energy from a source.
Partial discharges are pulse-like and cause mechanical
stress waves (acoustic waves) to propagate within the
transformer. If the stress waves propagate to the
transformer tank wall, they may be detected with a
transducer that is tuned to the right frequencytransducer that is tuned to the right frequency.

16. Partial Discharge – Acoustic Method
Acoustic signals propagate from the PD source to theAcoustic signals propagate from the PD source to the
sensor.
Th i l ill t diff t t i l Th fThe signals will encounter different materials. Therefore,
acoustic signals can only be detected within a limited
distance from the source.
Consequently, the sensitivity for PD inside transformer
windings, for example, may be quite lowwindings, for example, may be quite low

17. Partial Discharge – Acoustic Method
Though not disturbed by signals from the electric network,
external and internal influences in the form of rain or windexternal and internal influences in the form of rain or wind
and non-PD vibration sources like loose parts, cooling fans
and oil flow from transformer oil circulating pumps will
t ti i l th t i t f ith th PDgenerate acoustic signals that interfere with the PD
detection.
These non-PD acoustic signals may extend up to the 50 to
100 kHz region. To diminish the effects of this disturbance,
acoustic sensors with sensitivity in the 150 kHz range areacoustic sensors with sensitivity in the 150 kHz range are
usually employed. Such sensors may,however, have less
sensitivity to PD signals than lower frequency sensors.

18. Partial Discharge – Acoustic Method
Most useful for events within theMost useful for events within the
line-of sight of the acoustic
transducers. This limits the
d t ti b t l thdetection range, but also the
amount of noise. Acoustic sensors
are typically installed
on the tank surface

19. Partial Discharge – Electrical Method
Covers a wider area, including e.g. bushing and tap changer.
External noise will also be detected and is difficult to remove.
Th l ti b t i t t di d t lThe correlation between instrument reading and actual
discharge magnitude is better than with the acoustic method.
Several international standards exist that define the
instrument response, which is the readout in pico-Coulomb or
micro-Volt, allowing a better comparison between manufacturer
and in-field measurementsand in field measurements

20. Partial Discharge – Electrical Method
PD creates electrical pulses in the form of a uni-polar pulsePD creates electrical pulses in the form of a uni polar pulse
with a rise time that is dependent upon the type of
discharge.
Oil gap discharge is very fast
Surface discharge may be up to 10x longer

21. Partial Discharge – Electrical Method
PD pulses have a wide frequency content at the origin ThePD pulses have a wide frequency content at the origin. The
high frequencies are attenuated when the signal propagates
through the equipment and the network and pulse shape is
l difi d d t lti l fl ti d itialso modified due to multiple reflections and exciting
resonant frequencies of elementary circuits.
The detected signal frequency is dependent on the original
signal, pulse propagation path to the sensing point and the
measurement methodmeasurement method.

22. Partial Discharge – Electrical Method
Electrical PD detection methods are often hindered byElectrical PD detection methods are often hindered by
electrical interference signals from surrounding equipment
and the network.
Most common and most difficult noise sources are aerial
corona discharges and discharges to electrostatic shields
that are not properly connected to either the HV bus or
ground.
Any on-line PD sensing method must have methods to
minimize the influence of such signals.

23. Partial Discharge – Electrical Method
The most common method for PD detection is to decouple
the High Frequency partial discharge signals using sensorsthe High Frequency partial discharge signals using sensors
that are capacitively coupled to the HV bus (coupling
capacitor).
Most HV apparatus have a natural “capacitor” built into the
HV bushings or CTs have a convenient point for
connection of the PD instrument.
Bushing test tap or CT shield leads are frequently used forBushing test tap or CT shield leads are frequently used for
partial discharge measurements.

24. Partial Discharge – Electrical Method
The best method of noise rejection for in fieldThe best method of noise rejection for in field
measurements employs the use of multiple sensors.
U f i l d l i th fi ld i lik l tUse of a single sensor model in the field is unlikely to
produce satisfactory results. If several sensors of different
types or at different locations are employed, the
possibilities to reduce external influences are greatly
enhanced.

25. Partial Discharge – Electrical Method
The multi-sensor approach can be split into two processes:
separate detection of external signals and energy flowseparate detection of external signals and energy flow
measurements.
E fl t b th i d ti dEnergy-flow measurements use both an inductive and a
capacitive sensor to measure current and voltage in the PD
pulse.
By the tuning of the signals from the two sensors,
they may be reliably multiplied and the polarity of thethey may be reliably multiplied and the polarity of the
resulting energy pulse determines whether the signal
originated inside the apparatus or outside.

26. Partial Discharge – Electrical Method
A modern PD instrument should employ both processes ofA modern PD instrument should employ both processes of
the multi-sensor approach allowing the comparison of PD
pulse magnitude from different sensors and pulses polarity
f fl tfor energy flow measurements.

27. Partial Discharge – Electrical Method
Typical connection to HV
Bushing capacitance tap.

28. Moisture in Oil
There are two reasons for moisture in oil.
1 Mi ti f i t f th i di1. Migration of moisture from the winding
insulation.
2. Transformer breathing

29. Moisture in Oil
The quantity of dissolved and dispersed water in insulating fluids is
significant for two reasons:
1) The presence of polar water molecules in the fluid adversely affects) p p y
the dielectric properties of the fluid
2) The amount of moisture in the oil can be reflective of the amount2) The amount of moisture in the oil can be reflective of the amount
of moisture in the paper insulation.

30. Moisture in Oil
The solubility of water in a dielectric fluid is temperature dependent,
t t t di /k f t i th il ith t t tso a statement regarding mg/kg of water in the oil without temperature
information would not be adequate. The calculation of the percent
water saturation, (mg/kg water/mg/kg of water at saturation)×100,
has a greater significance as it indicates the possibility of free water
formation in the oil.

31. Moisture in Oil
Free water in the oil results in unacceptable dielectric strength values.
W t titi b t i l ti d il Thi titiWater partitions between paper insulation and oil. This partition
coefficient is temperature dependent so the water will move between
the paper insulation and the oil as the temperature changes.
For example, if the temperature increases, movement of water is from
the paper to the oil This increase in water content of the oil willthe paper to the oil. This increase in water content of the oil will
increase the percent saturation and could result in free water formation,
should the temperature of the oil decrease rapidly. Table 4 provides
l id li f th i t t ti f d t d i tgeneral guidelines for the interpretation of data expressed in percent
saturation.

32. Moisture in Oil

33. Moisture in Oil

34. Dissolved Gas in Oil
Dissolved Gas Analysis (DGA) and moisture measurement of
th i l ti il i d th t i t t t tthe insulation oil are recognized as the most important tests
for condition assessment of transformers.
Normal degradation of the oil occurs primarily as a result of
oxidation, which is a relatively slow process. However, under
the influence of electrical and thermal faults, oil can be degradedthe influence of electrical and thermal faults, oil can be degraded
to form a variety of low-molecular weight gaseous products that
are soluble in oil.

35. Dissolved Gas in Oil
The interpretation of the data can be a complex process because
f th l b f i t t d tiof the large number of equipment parameters and operating
conditions that affect gas formation.
Monitoring the rate of gas generation is typically more important
than the absolute levels at any given time.

36. Dissolved Gas in Oil

37. Dissolved Gas in Oil
Determining the concentration of the fault gases in hydrocarbon
Oils is best accomplished in a two-step process
(1) extraction and
(2) chromatographic analysis.

38. Dissolved Gas in Oil
Extraction - Separating the gases from the oilExtraction Separating the gases from the oil
This is accomplished by subjecting the oil to high vacuum. The
volume of the extracted gases is measured and a portion of the
i t f d t h t hgas is transferred to a gas chromatograph.
Chromatographic analysis allows for both the qualitative and
quantitative composition of the dissolved gases.

39. Dissolved Gas in Oil
Fault IdentificationFault Identification
•Corona (low level PD): Low energy electrical fault resulting from
th i i ti f th i l ti fl id di th f ltthe ionization of the insulating fluid surrounding the fault.
Typically, increasing levels of hydrogen without concurrent
increases in hydrocarbon gases identifies corona.
•Sparking (elevated PD): Intermittent high-voltage discharge that
occurs without high current Sparking is characterized byoccurs without high current. Sparking is characterized by
increasing levels of hydrogen, methane, and ethane without a
concurrent increase in acetylene.

40. Dissolved Gas in Oil
Fault Identification
•Overheating: Can be caused by overloading, circulating currents,
improper grounding and/or poor electrical connectionsimproper grounding, and/or poor electrical connections.
Overheating is characterized by the presence of hydrogen, together
with methane, ethane, and ethylene. A concurrent increase in
b id b kd f ll lcarbon monoxide suggests breakdown of cellulose.
•Arcing: Most severe of all fault processes, involves high current
and high temperatures. Arcing is characterized by the presence
of acetylene.

41. Dissolved Gas in Oil
Load Tap Changer
•Load Tap Changers (LTCs) are typically housed in a separate tankp g ( ) yp y p
•Depending on type of tap changer, different gases will be present
•On Line DGA can be applied to the LTC tank

42. Dissolved Gas in Oil

43. Dissolved Gas in Oil
Common Fallacy:y
“Dehydrating and/or degassing the oil will fix your problem.”
The truth is:
Cleaning the oil without finding the source will just mask the potential
problem and provide you a false sense of security.

44. Dissolved Gas in Oil
Proper Sampling Portp p g
It is very important that the sample of oil is representative of the
entire tank Therefore it is critical that the sampling port doesentire tank. Therefore, it is critical that the sampling port does
not provide oil which is stagnant.

45. Temperature and Loading
Transformers are typicallyyp y
equipped with two (2)
temperature indicators:
•Top Oil Temperature
•Winding Temperature
Most newer gauges have
provisions for analog
outputs

46. Temperature and Loading
Top Oil Temperature
Thi li i i f i l i h f d A b d iThis application is fairly straight forward. A submerged in
oil temperature probe is installed in the top section of the
transformer tank.

47. Temperature and Loading
Winding Hot Spot Temperature
Th diff h d h hi l i d i d ThThere are different methods how this value is derived. The
most common method simulates/estimates the winding
temperature by using a proportional current transformerp y g p p
output to power a heating element contained in the well
(pocket). Winding temperature is simulated by adding the
rise due to the heater output to oil (liquid) temperaturerise due to the heater output to oil (liquid) temperature.
This is an estimate based on the transformer design.

48. Temperature and Loading
Outputs from the Top Oil and Winding Hot Spot
Temperature Gauges are used to active stages of
f li d ltransformer cooling and alarms.
•Fans
•Oil Pumps

49. Temperature and Loading
Winding Hot Spot Temperature
Newer technology provides fiber optic temperature probes
strategically placed in the winding to obtain direct measurements.
The trick is identifying the hottest spot in the windings

50. Temperature and Loading
Other Temperature Probes
Other locations for temperature probes may include:
•Ambient
•LTC•LTC
•Top and Bottom of Radiators

51. Temperature and Loading
Additional data that provides additional data for analysis of the
transformer is load trending.transformer is load trending.
The ability of tying all the data points together can help optimizing
the transformer performance:the transformer performance:
•Top Oil
•Bottom Oil
•Winding Hot Spot
•Ambient
•Load

52. Bushing Monitoring
According to Doble Engineering client questionnaires, the leading
cause (35 percent) of transformer failures (above 100 MVA) iscause (35 percent) of transformer failures (above 100 MVA) is
bushing problems. While most utilities periodically test bushings,
there is a solid argument that periodic testing is not enough.
Many companies rely on regularly scheduled maintenance testing.
For those still relying only on periodic testing, there are some
compelling reasons to consider continuous monitoring.

53. Bushing Monitoring
While it obviously costs money to purchase and install an online
monitoring system it's important to note that periodic bushing testingmonitoring system, it s important to note that periodic bushing testing
is not without its costs as well. Regardless of the frequency, testing
takes time, ties up workers and requires an outage. And, periodic
testing is performed at lower voltages and under temperatures
conditions that differ very much from the operating ones.
Most importantly, periodic testing misses problems that can arise
between tests.

54. Bushing Monitoring
If you lose a bushing, it is not necessarily true that the transformer
will be damaged beyond repair But following a bushing failurewill be damaged beyond repair. But, following a bushing failure,
there is typically some amount of damage. How much depends on
how the bushing failed. A typical bushing failure occurs because the
dielectric degrades. If no corrective action is taken, there will
eventually be internal arcing, which often leads to a violent failure.
If the bushing fails at the air end, substation crews have to contendg ,
with flying porcelain, which is a real danger to anyone in the
substation. Further, those flying pieces of porcelain can damage
nearby equipment most notably other porcelain bushingsnearby equipment—most notably, other porcelain bushings.

55. Bushing Monitoring
If the bushing fails at the oil end, the oil in the main tank may be
contaminated the transformer coils may be damaged and/or the tankcontaminated, the transformer coils may be damaged and/or the tank
might rupture. And, there is always the concern about fire. Oil,
oxygen and arcing are not a good mix. With luck, the fire will be
limited to the oil contained in the bushing. In a worst-case scenario,
oil from the main tank will also ignite—leading to a major fire and
the complete loss of the transformerp

56. Bushing Monitoring

57. Bushing Monitoring
Where are these
bushings used?bushings used?
23kV to 46kV
Sometimes
69kV to 800kV
AlwaysC1
C1 C2
HV
Tap
C2

58. Bushing Monitoring

59. Bushing Monitoring
Failed 500 kV Bushing

60. Bushing Monitoring
Failed 230 kV Bushing – Oil End

61. Bushing Monitoring
•Consider the clearances associated
with bushings vs main tank.
Drawing of a 500-18 kV GSU.

62. Bushing Monitoring
•Consider transportation of bushings vs main tank (impact recorder)•Consider transportation of bushings vs main tank (impact recorder)
•Consider handling of bushings vs main tank
•Consider testing done to bushings vs main tank

63. Bushing Monitoring
Tap Cap (removed)
BushingGard Adapter/Sensor
(side-view)
BushingGard Adapter/SensorBushingGard Adapter/Sensor
(connector-view)

64. Bushing Monitoring
In addition to the advantages associated with bushings, equipment
can be connected to the bushing capacitance tap for monitoringcan be connected to the bushing capacitance tap for monitoring
of the main windings:
•Partial discharge
•Transient overvoltage
•FRA

65. Infrared Monitoring
Another relatively inexpensive method of on-line monitoring
is infrared scanningis infrared scanning.
Identifying a temperature delta of just several degrees can identify
a problem.
On a transformer, infrared can be used to inspectOn a transformer, infrared can be used to inspect
•Main tank
•Bushings
•LTC•LTC
•Radiators
•Arresters

66. Infrared Monitoring
NETA Guidelines for Electrical Equipment

67. Infrared Monitoring
120.7°F
110
120
84 1°F
90
100
84.1°F

68. Infrared Monitoring

69. Power Transformer Monitoring
Other points to monitor include:
• LTC Position indicators• LTC Position indicators
• LTC motor currents
• Status of cooling
• Loss of AC / DC

70. Communications / Integration
Many IEDs now have communication options
If communications options are not available,
h (di i l l ) f h IEDthe outputs (digital or analog) of the IEDs can
be connected to a master processor.

71. Communications / Integration
Examples of a master processor include:
Eaton - BushingGard
Schweitzer – SEL2414, Transformer Monitor
GE – Kelman TapTransp
Alstom – MS 3000, Online Condition
Monitoring and Expert SystemMonitoring and Expert System

72. Failure Statistics
30
Australia & New Zealand
20
25
10 Years of Data
457 Failures
10
15
rcent-%
5
10
Per
0
Windings Winding
Accessories
Core Tap Changer Motor Drive
Parts
Bushings Cooling Other

73. Failure Statistics
35
Eskom (South African Public Utility)
25
30
136 Severe Failures
1996 - 2006
15
20
Percent-%
5
10
0
Windings Core Tap Changer Main Tank & Oil
System
Bushings Auxiliaries Others

74. Failure Statistics
35
Canadian Electricity Association
25
30
Forced Outages
1997 - 2006
15
20
Percent-%
5
10
P
0
5
Bushings Windings OLTC Core Leads Cooling Auxiliaries Other

75. Failure Statistics
45
50
Doble Engineering
35
40
1993 - 1998
25
30
ercent-%
10
15
20
P
0
5
Bushings Tap Changer Tank and Oil Core Winding Other

76. Failure Statistics
ΔT
Partial Discharge
DGA
Moisture in Oil
Oil Dielectric
Cooling System
Monitoring
Fan Current
Bushing Monitoring
Operations Counters
Contact Wear
Control Monitoring
Oil Dielectric
Partial Discharge
Leakage
Reactance
Online FRA
Control Failure
Power Factor
Capacitance
Partial Discharge

77. Case Study #1
JEA avoids costly transformer failures
As the largest community-owned utility in Florida and the eighth largest in the United
States, JEA serves 360,000 customers in Jacksonville and parts of three adjacent
counties. JEA has net generating capability of over 2,300 MW and owns and operates
a combination of generation, transmission and distribution assets. Water and sewer
systems are also part of its utility service offeringssystems are also part of its utility service offerings.
As a part of efforts to maximize reliability across this extensive service area, JEA
installed an automated energy analysis system. The system helps engineering
personnel isolate and correct power quality issues and equipment risks throughout
their network. The solution has repeatedly proven its value, identifying degradation in
a number of load tap changer (LTC) units. The early discovery of these problems has
avoided failures that could have cost JEA millions of dollars in associated transformeravoided failures that could have cost JEA millions of dollars in associated transformer
replacements, as well as avoiding service interruptions for commercial, industrial and
residential customers.

78. Case Study #1
The system
The energy analysis system currently includes 17 advanced power quality meters at
generation sites, 73 at transmission and distribution substations, and 92 at important
industrial customer sites. The meters track power quality conditions and monitor the
performance of equipment including relays and transformers. All real-time and
historical data is uploaded automatically over hard line modems Ethernet orhistorical data is uploaded automatically over hard line modems, Ethernet, or
wireless links to a set of central servers that run the energy analysis software. Meters
also connect directly with the RTU/SCADA system. JEA personnel receive local and
web-enabled remote access to data, including standardized SARFI 80 reports that
regularly summarize all events in the system. Monthly PQ reports are also offered to
customers as a value-added service. Extensive data analysis tools are used by the
Systems Analysis Group to help reveal and address any power quality issues
throughout the T&D network Information also benefits managers and engineers inthroughout the T&D network. Information also benefits managers and engineers in
other departments, including electric delivery, customer sales and service, systems
protection and controls, substation maintenance and generation plant management.

79. Case Study #2
Prevention of a 25 MVA Transformer Failure
A west coast investor owned utility is monitoring a large population of transmission
and distribution transformers with the Dynamic Ratings Comprehensive Monitoring
System. Monitoring of cooling, thermal performance, dynamic ratings calculations,
DGA as well as bushings are all part of this system.
Recently the bushing health monitoring system went into alarm indicating a critical
problem. Before jumping to conclusions all data was reviewed and a site visit ensued.
The polar chart indicated that the B phase capacitance and power factor were changing
with the most significant changes occurring with the power
factor. Trending also indicated a high level of temperature dependency.
For further confirmation a Dynamic Ratings Continuous PD monitoring Module wasFor further confirmation a Dynamic Ratings Continuous PD monitoring Module was
added to the system. Significant PD levels exceeding 1 volt were detected on B Phase.
This was done without any outage. There is no indication of any gassing from the
DGA monitor and this coupled with pulse shape analysis of the PD signal it was
d i d h PD i l h b i f h b hidetermined the PD signals have to be coming from the bushing.

80. Case Study #2
Finally an outage was scheduled and off-line power factor tests were performed under
ambient conditions. The off-line tests shows little change in capacitance, but a
significant change in power factor.
Lessons Learned
Bushings are one of the top failure modes of power transformers. Many times the
failure modes of bushings are voltage or temperature dependent and may fail in afailure modes of bushings are voltage or temperature dependent and may fail in a
matter for weeks. Offline provide misleading results since the full affect of operating
are not present. There have been many cases where off-line show “good” readings, but
in fact the bushing is defective.

81. Case Study #2

82. Special Considerations
The purpose of on-line monitoring is to
identify incipient faults (faults in the infancyidentify incipient faults (faults in the infancy
stage).
Do not confuse on-line monitoring with
t tiprotection.

83. Special Considerations
When on-line monitoring is utilized,
consideration must be made about fastconsideration must be made about fast
developing faults.
Each alarm should be considered as a “real”
d f f ilmode of failure.
Do not discount alarms / alerts

84. Off-Line Testing
• Good Bill of Health• Good Bill of Health
Only means there is no clue as to the unit will fail
• Easier to Predict when Failure is imminentas e to ed ct w e a u e s e t
Advanced stages of deterioration
• The only true way to capture impending failure is to
monitor continuously

85. Why On-Line Monitoring?
•Time between outages extended
•Many failure modes happen quickly (days, weeks, months)
•Off-line tests can not simulate actual operating conditions
(temperature, voltage, load, mechanical)
•Historical data not sufficient to make a good decision
L h h i•Less thorough maintenance

86. Advantages at a Glance
• Integration of the system into the existing control• Integration of the system into the existing control
technology
• Condition monitoringg
• Precise fault analysis
• Trend recognition
• Condition oriented maintenance

87. High-Voltage Technical Portal

88. High-Voltage Technical Portal

89. References
•IEEE C57.12.00
•IEEE C57.19.00
•IEEE C57.91
•IEEE C57 100•IEEE C57.100
•EPRI Diagnostic Conference – July 2006
Business Case for Transformer On-Line Monitoring

90. The End
Thank youThank you.
Questions?Questions?