Practicing DGA 09072010

1
A BPL GLOBAL COMPANY
Agenda
Serveron Corporate Background
Transformer Reliability
Dissolved Gas Analysis
DGA Interpretation
Accuracy & Gas levels
On-Line and Lab DGA Case Studies
Serveron Corporation 2008. No use without permission.

2
2

3
3
Hillsboro, USA Headquarters
Branch offices in Beijing and Brussels
Incorporated in 2001
EPRI and proprietary GC based

technology
Siemens investment and rebranding agreement-2006
Fully integrated operation
Engineering
Manufacturing
Sales & Marketing
Field Service
Secure Data Center
Customer Base
Over 1000 On-line DGA monitors

Over 80 major utilities worldwide
Most major transformer
manufacturers
4
4
Michel Duval
Dr Michel Duval obtained a B.Sc. in chemical engineering in 1966

and a Ph.D. in polymer chemistry in 1970. He has joined IREQ
(Institut de recherche dHydro Qubec) in 1970. Since then, he has
made significant contributions in 3 main fields of R&D: dissolved gasin-oil analysis (DGA), electrical insulating materials and lithiumpolymer batteries.
In the field of DGA, M. Duval:
Is well-known for his Triangle method of DGA interpretation, used

worldwide.
Has developed and promoted the use of gas-in-oil standards in the IEC
and ASTM standards
Has established the typical levels of gas formation observed in various

types of electrical equipment in service, now used as a reference by the
industry
Has been the Convenor of several IEC working groups and CIGRE task
forces and is the principal author of several IEC international standards
on DGA (60567, 60599, 61181).

5
5
Michel Duval
In the field of electrical insulating oils, M.Duval has researched and

published in the areas of:
Metal passivators
Oil reclamation timing
Unstable oil detection and prevention
Paraffinic content of oils
The characterization of XLPE in HV cables and of HV outdoors insulators
M. Duval holds 16 patents and is the author of more than 75 scientific

and technical papers and books , 5 international standards (IEC,
ASTM), and numerous technical reports and presentations in
conferences. He is a Fellow of IEEE and of the Chemical Institute of
Canada. He may be contacted at duvalm@ireq.ca.

6
6

7
7
Transformer Asset Management
Transformers are a critical and costly element in the electrical grid
Unplanned failures at any point in the transformer lifecycle have

major consequences
Failure Rate
Bathtub Curve
New
Time
End of
Design Life
DGA condition assessment has been recognized for over 30 years for
improving reliability and lowering transformer asset maintenance
costs
8
8
Failures They Happen!

9
9
One Example
True Story: 520 MVA GSU Transformer
$3.5 million replacement cost
$0.5 million environmental cleanup
$1.5 million/day spot market buy
$17 million loss in

eight days!!

10
10
Number of Explosions
Increasing Failure Rates

250
200
150
100
50
0 79 80
82
84
86
88
90
92
94
96
98
00
02
Sergi Transformer Explosions in the United States
All catastrophic failures in any USA news media
Reported number of failures is increasing
Total % of explosive failures is increasing

11
11
Increasing Failure Rates

Transformer
Failure Rate (%)
5
4
3
2
1
0
1973
1978
1983
1988
1993
1998
2003
2008
2013
William H. Bartley P.E., Hartford Steam Boiler Inspection & Insurance Co.,
"Life Cycle Management of Utility Transformer Assets,"
Projected rates will reach unacceptable levels if nothing is done to

improve the system;
Replacing the fleet is not an alternative
One solution is increased monitoring
Published reports in the news validate an increasing failure rate
12
12
Factors Leading to Increased Failures

Stress with age: The average age of US power transformers is
>42 yrs, increasing 0.7 yrs./yr.
Age itself is not a cause of failure
Increased energy demand: Transformer peak and average
loading has increased

Stress experienced:
Mechanical
Thermal
Electrical
Chemical
Tight budgets:
Cutbacks
Deferred
in O&M
capital
replacement

13
13
Modes of Stress
MECHANICAL STRESS
ELECTRICAL STRESS
Failures due to internal stress, weather,

and accidents:
Failures initially categorized into areas

such as over-voltage or partial
discharge, typically accompanied by
thermal or chemical failure:
Tank failure
Radiator failure
Winding buckling
Core damage
Insulation damage
LTC failures
Lead failures
Bushing failures
THERMAL STRESS
Failures due to insulation destruction or
conductor burn-through as the result of:
Short or long-term overloading

Faults
Undersized leads
Contact coking
Bad joints
Loss of cooling
Design issues
Switching surges
Lightning
CHEMICAL STRESS
Failures due to:
Ingress of water
Ingress of oxygen
Loss of insulating oil
Paper degradation due to heat or aging
14
14
Types of Faults
Workmanship
Overloading
Loose
Connections
Moisture
Other
Lightning
Inadequate
Maintenance
Insulation
Deterioration
Through Faults
Source: Hartford Steam Boiler Inspection and Insurance Co.
15
15
What monitoring, other than DGA, do you perform on your
transformers?

16
16

17
17
Monitoring of Gases in Transformers

As insulating material breaks
down due to stress, gases are

formed which dissolve in the
transformer oil
ISSOLVED
Levels and combinations of the
gases formed are used to detect

incipient faults
AS
For over 50 years, DGA has been
the leading tool to assess

transformer condition
NALYSIS

18
18
Monitoring of Gases in Transformers

Gas originates from many places:
Mineral insulating oil

Conductor paper insulation
Pressboard barriers
Other materials
Gas generation is related to high
material temperatures (150C to

1,000C).
Gases are symptoms of:
Poor design or construction

Too much electrical stress
Too much thermal stress
Too many short circuits
Overall poor condition

19
19
Which Gases are Generated?

Eight key gases in transformer oil are associated with fault
conditions.
DGA detects the level of gases indicative of incipient faults
which may lead to transformer failure.
Ethylene
(C2H4)
Carbon
Monoxide (CO)
Methane
(CH4)
Carbon
Dioxide (CO2)
Ethane
(C2H6)
Hydrogen
(H2)
Acetylene
(C2H2)
Insulating Oil
Oxygen (O2)

20
20
The Importance of DGA to Reliability
DGA enables the detection of the presence and severity of faults:
Air leaks
Hot metal faults
Arcing and partial discharges
DGA may help to indirectly detect root cause of faults in:
Windings (short circuits, insulation failure)
Cleats and leads (high contact resistance, loose contacts)
Tanks (ground problems, circulating currents)
Tap changer (resistive contacts, leaks into main tank)
Core (magnetic flux problems)
Oxidation of materials (mostly CO, CO2)

21
21

22
22
Traditional DGA Methods

DGA labs use Gas Chromatography (GC)
Samples of oil are manually taken

on a set schedule.
Samples are sent to Accredited Labs

23
23
Standards and Guideline Groups
IEEE is a worldwide organization (historically focused on North &

South America) that develops guides & standards for all types of
electrical & electronic equipment.
There has been increased effort in recent years to harmonize

with IEC but long-established equipment standards & practices
involve un-reconcilable differences.
ASTM is one of the largest voluntary standards development

organizations in the world; a source for technical standards for
materials, products, systems, and services.
CIGRE is a worldwide organization doing technical work in the field of

HV equipment and corresponds approximately to IEEE in the US.
IEC issues international standards and corresponds to ASTM in the

US.

24
24
Standards and Guidelines Governing

Laboratory DGA
ASTM D3612-2002 Standard Test Method for Analysis of Gases Dissolved in

Electrical Insulating Oil by Gas Chromatography
IEC 60567-2005 Oil-filled electrical equipment - Sampling of gases and of oil

for analysis of free and dissolved gases
Some other ASTM standards are available;

D3613-1998 - Standard Practice for Sampling Insulating Liquids for Gas
Analysis and Determination of Water Content
D3305-95 (1999) Standard Practice for Sampling Small Gas Volume in

a Transformer
D2759-2000 Standard Practice for Sampling Gas from a Transformer

under Positive Pressure

25
25
Oil Sampling
Manual Sampling
A small volume of oil (30 mL) is collected in a gas-tight syringe,
using a 3-way valve, then transported to the laboratory
ASTM method D3613 details procedures for oil sample handling
On-Line Sampling
A small volume of oil is continuously circulated through the
monitor and then returned to the transformer
The oil is sampled and analyzed for gas content by the monitor
On-Line monitors offer a closed-loop repeatable oil sampling
process, with no possibility for contamination

26
26
Oil Sampling Guidelines
References:
ASTM D923: Standard Practice for Sampling Electrical Insulating Oils for
Gas Analysis
IEC 60475: Method of Sampling Liquid Dielectrics
IEC 60567: Guide for the Sampling of Gases and Oil from Oil-filled
Electrical Equipment for the Analysis of Free Dissolved Gases
Sampling of Dielectric Liquids by Lance Lewand, Doble Client

Conference, 2003

27
27
Gas Extraction
Dissolved gases are present in transformer oil at
concentrations from <1 part-per-million (ppm) up to a few

percent of oil volume.
ASTM method D3612 specifies three methods to extract
dissolved gases from the Transformer Oil
Method A (Vacuum Extraction)
Partial De-Gassing method
Method B (Stripper Column Extraction)
Method C (Headspace method)

28
28
Gas Extraction
METHOD A (partial de-gassing method)
Introduce the oil sample into a pre-evacuated vessel
The extracted gases are compressed to atmospheric pressure

and the total volume measured
The gases are then analyzed by gas chromatograph
METHOD B
The oil is sparged with a carrier gas on a stripper column
The gases are then flushed from the stripper column into a gas
chromatograph for analysis

29
29
Gas Extraction
METHOD C (Headspace Method)
Introduce an oil sample in a glass vial with a headspace gas

phase of argon above it.
Some of the gases migrate from the oil to the headspace and
equilibrate according to Henrys law.
Henrys Law - At a constant temperature, the amount of a given gas

dissolved in a given type & volume of liquid is directly proportional to the
partial pressure of that gas in equilibrium with that liquid.
At equilibrium, the headspace is transferred to a sample loop then

introduced into a gas chromatograph

30
30
Gas Chromatography
Gas Detector
Stationary phase
Mobile
phase
flow
Mobile phase = gas
1.4
Stationary phase
= solid adsorbent or liquid
Mobile gas phase flows

through the column
under pressure
1.2
1.0
Intensity
0.8
0.6
0.4
0.2
0.0
0
50
100
150
Time (s)

31
31
200
250

32
32
Gas Solubility Coefficients
Some gases dissolve into the oil more than others

This is known as Solubility
In the following graph, Ethane, C2H6 dissolves the most and

Hydrogen, H2 the least.
Solubility coefficients change over temperature

Gases dissolve into oil in different amounts as temperature
changes
Solubility coefficients are required because all Laboratory gas

measurement systems measure gas-in-gas and these concentrations
are different than the gas-in-oil
Note:
Temperature of the Oil during a manual sample is required only for Lab
Moisture-in-Oil calculations
Temperature of the Oil during the gas extraction process is required to

determine the Solubility Coefficient

33
33
The triangles and squares in the graph represent ASTM solubility

coefficients at selected temperatures used in the Laboratory DGA process
34
34
Solubility coefficients are applied to the laboratory measured gas-in-gas

values to derive the true gas-in-oil concentration in transformer oil.
Example
Method C headspace sampling formula (from ASTM D3612)
CL = CG ( K + VG/VL)
CL = Gas-in-Oil value, ppm
CG = Gas-in-Gas value, ppm
K = Solubility Coefficient
VG = Volume of Oil space, 7 mL*
VL = Volume of Gas space, 15 mL*
(*)Typical Laboratory values
Example for Hydrogen, H2

CG = 40 ppm
K = 0.0558
mL
CL = 40 (0.0558 + 0.467)
CL = 40 (0.5225)
VG/VL = 7/15 = 0.467
CL = 20.9 ppm is the Gas-in-Oil value

35
35
Manual DGA Process

Gas Chromatograph
Gas Chromatograph
Transformer
Carrier Gas
Supply
(8.2 psig)
Manual Injection
Vent
Oil Sampled into

Syringe or Bottle
1
0
9
1.0
mL
1
2
Transportation
4
6
Thermal
Conductivity
Detector
Channel B
Porapack Column
(A)
Channel A
Molecular Sieves
Column (B)
1.0
mL
Oil Sampled From Syringe or

Bottle
Extraction Device
Vacuum (A)
"Stripper Column (B)
Headspace (C)
Transformer Oil
Transformer Gas
Each time a sample

is handled, there is a
chance that error
may be introduced
into the
measurement

36
36
Concerns with Laboratory DGA
Sampling:
If not done according to standard procedures by experienced personnel,
the samples may lead to erroneous results.
Transport:
If procedures are not followed properly, transportation may introduce
contaminations.
Exposure to external factors may also affect the oil sample:

If low-quality syringes are used, additional O2 & CO2 may be introduced
or key gases (e.g. H2) may be lost.
37
37
Good sample versus bad sample

It is sometimes very clear to the laboratory performing the
analysis if it was taken improperly.
The presence of free water or foreign objects such as

insects, pipe sealant, tape or putty are strong
indicators that the drain valve was not adequately
flushed out prior to sampling.
Materials used for collecting samples;
Galvanic fittings (zinc coated) used in the drain valve

assembly such as the drain plug, can create a galvanic
reaction with water, and cause very high levels of
hydrogen to be produced.

38
38
Good sample versus bad sample

Sample ports should have Brass Fittings or Stainless
Steel
Tubing - Tygon tubing or other compatible tubing is necessary for

sample collecting
Tubing should only be used once and then discarded as the walls
of the tubing have a memory (can hold gases, water and other
chemicals compounds in the wall of the tubing)
Incompatible tubing such as natural rubber or PVC will

contaminate a sample with unwanted materials.

39
39
Concerns with Laboratory DGA
Laboratory handling is a concern

the sample may be contaminated
in the process of being analyzed
Gas Chromatography analysis in

the laboratory is another source
of errors if not done by
experienced personnel and not
following QA/QC procedures.
The accuracy required for laboratory analysis is +/- 15 % on gas

concentrations, but several laboratories are much less accurate than
that.
40
40
Differences Among Lab DGA Results
Independent studies have shown lab-to-lab DGA differences can be

large, with some results necessarily not accurate
This study consisted of Sixteen (16) samples from the same

Transformer - All samples were collected by the same individual using
the same process.
Methane
All samples were sent to 3 different

laboratorys simultaneously.
2500
2000
1500
1000
500
Possible causes of differences are;
0
1
2 3
Even with the same sampler

and sample process, contamination
could have occurred
4 5 6 7 8 9 10 11 12 13 14 15 16
Lab A
Lab B
Lab C
Hydrogen
2000
Transportation may have affected

the samples
1500
1000
500
Laboratory accuracies are +15%

on average; some are much worse
0
1
2 3
4 5 6
Lab A
7 8
9 10 11 12 13 14 15 16
Lab B

41
41
Lab C
Questions
What is the current DGA process at your utility?
How are samples taken?
What is the frequency of samples?
What method (A, B or C) is used in your lab?
Do you have your own lab or rely on third party labs?
How often is the Lab Equipment Calibrated?
Do you send a known sample to the Lab?

42
42

43
43

44
44
Standards and Guidelines Governing

the Interpretation of DGA
IEEE Std. C57.104.1991 IEEE Guide for the Interpretation
of Gases Generated in Oil Immersed Transformers

IEEE PC57.104 Draft 11d IEEE Guide not approved
IEC 60599-1999 Mineral Oil Impregnated Electrical
Equipment in Service: Guide to the Interpretation of

Dissolved and Free Gas Analysis.
IEC 60599-1999, Amendment 1, 04/2007

45
45

46
46
Gas Sources
Gases in oil always result from the decomposition of
electrical insulation materials (oil or paper), as a result of

faults or chemical reactions in the equipment
For example:
Oil is a molecule of hydrocarbons, i.e., containing hydrogen and
carbon atoms, linked by chemical bonds (C-H, C-C)

47
47
Gas Formation
Some of these bonds may break and form H*,
CH3*, CH2* and CH* radicals.

48
48
Gas Formation
All these radicals then recombine to form the fault gases
observed in oil:

49
49
Gas Formation
In addition to these gases, the decomposition of paper
produces CO2, CO and H2O, because of the presence of

oxygen atoms in the molecule of cellulose:

50
50
The Main Gases Analyzed by DGA

Hydrogen
H2
Methane
CH4
Ethane
C2H6
Ethylene
C2H4
Acetylene
C2H2
Carbon monoxide CO
Carbon dioxide
CO2
Oxygen
O2
Nitrogen
N2

51
51
Fault Gas Formation

Most of the time, all gases are present in DGA results,
however, some are formed in larger or smaller quantities

depending on the energy content of the fault
Example; Low energy faults such as Corona Partial
Discharges in gas bubbles, or low temperature hot spots,

will form mainly Hydrogen, H2 and Methane, CH4

52
52
Fault Gas Formation
Faults of higher temperatures are necessary to form large quantities

of Ethylene, C2H4
Finally, it takes faults with a very high energy content, such as in

electrical arcs, to form large amounts of Acetylene, C2H2
By looking at the relative proportion of gases in the DGA results it is

possible to identify the type of fault occurring in a transformer in
service

53
53
Gas Formation Patterns

Are related only to the materials used and faults involved
Are the same in all equipment where these materials are
used
Sealed or air-breathing power transformers

Reactors
Instrument transformers
LTCs
Etc.

54
54

55

56

57

58

59

60

61

62
Fault Gas Formations
(Ref. S. D. Myers 1996)

63
63

64
64

65
65
Fault Types
Partial discharges of the corona-type (PD)
Typical examples:
Discharges in gas bubbles or voids trapped in paper

A result of poor drying or poor oil-impregnation
Discharges of low energy (D1)
Typical examples:
Partial discharges of the sparking-type

Inducing carbonized punctures in paper
Low-energy arcing, inducing surface tracking of paper and carbon
particles in oil
Ref. IEC 60599-1999

66
66
Fault Types
Discharges of high energy (D2)
Typical Examples
High Energy Arcing

Flashovers
Short Circuit with power follow through
These result in;
Extensive damage to paper

Large formation of carbon particles in oil
Metal Fusion
Tripping of the equipment or gas alarms
Ref. IEC 60599-1999

67
67
Fault Types
Thermal faults of temperatures <300C (T1)
Typical Examples:
Overloading
Blocked oil ducts
Evidenced by paper turning:
Brown (>200C)
Black or carbonized (>300C)
Ref. IEC 60599-1999

68
68
Fault Types
Thermal faults of temperatures between 300 and
700C (T2)
Typical Examples:
Defective contacts
Defective welds
Circulating currents
Evidenced by:
Carbonization of paper
Formation of carbon particles
Ref. IEC 60599-1999

69
69
Fault Types
Thermal faults of temperatures >700C (T3)
Typical Examples:
Large circulating currents in tank and core

Short circuits in laminations
Evidenced by:
extensive formation of carbon particles in oil

metal coloration (800C) or metal fusion (>1000C)
Ref. IEC 60599-1999

70
70

71
71
Diagnostic Tools for DGA

Reference Standard
Tool
IEEE C57.
104-1991
Dornenburg Ratios
TDCG Procedure
Key Gas Procedure
TCG Procedure
Rogers Ratios
IEEE PC57.
104 D11d
IEC 605991999
IEC Gas Ratios
Duval Triangle
CO2/CO Ratio
O2/N2 Ratio
C2H2/H2 Ratio

72
72
Diagnostic Tools for DGA

Procedures vs. Methods
Procedures are guidelines on what steps are necessary to
continue to examine the potential for a fault or problem

condition.
Methods are results that provide condition assessments
of potential faults directly

73
73
Key Gas Procedure

Complements the TDCG procedure in order to identify
faults
Focuses on the main (or key) gas formed
Limitations:
High tendency to return inconclusive results

74
74
Key Gas Procedure

75
Key Gas Examples

Low Energy Partial Discharge
Thermal Oil Fault

100
100
80
80
60
60
40
40
20
20
CO
H2
CH4
C2H6
C2H4
C2H2
100
100
80
80
60
60
40
40
20
20
CO
H2
CH4
C2H6
C2H4
H2
CH4
C2H6
C2H2
CO
H2
CH4
C2H6

76
76
C2H4
C2H2
High Energy Arcing
Thermal Oil and Cellulose Fault
CO
C2H4
C2H2
Rogers Ratio Method

77
77
IEC Gas Ratio Method

78
78
Summary of the Ratio Methods

The Basic, Rogers and Dornenburg methods use the
same 3 basic gas ratios: (CH4/H2, C2H2/C2H4 and

C2H6/C2H4).
Depending on the values of these gas ratios, codes or
zones are defined for each type of fault

One drawback of these ratio methods is that no diagnosis
can be given in a significant number of cases, Dead

Zones - fall outside defined zones.

79
79
3-D IEC Gas & Rogers Ratio Methods

80
80

81
81
Triangle Method
The Triangle, developed empirically in the early 1970s,
and is used by the IEC.

Based upon the 3 gases (Methane, CH4, Ethylene, C2H4
and Acetylene, C2H2) corresponding to the increasing

energy levels of gas formation.
One advantage of this method is that it always provides a
diagnosis, with a low percentage of wrong diagnoses.

There are no indeterminate diagnostics using the Triangle
method.

82
82
Triangle Method
The triangle method plots the relative % of the 3 gases on
each side of the triangle, from 0% to 100%.

The 6 main zones of faults are indicated in the triangle,
plus a DT zone (mixture of thermal and electrical faults)

Approximately 200+ inspected cases in service were used
to develop the Triangle

83
83
Triangle Method

84
84
Triangle Method
How have fault zones been defined in the Triangle?
Answer:
They are based on a large number of cases of faulty
transformers in service which have been inspected visually.
The root cause of the failure was determined and matched to

the DGA data.
The Triangle was tested with all these cases and correctly
identifies the zone that matches the root cause of failure at a
very high percentage.

85
85
Triangle Method
Cases of faults PD and D1
Tracking; Sparking; Small Arcing

86
86
Triangle Method
Cases of faults D2

87
87
Triangle Method
Cases of thermal faults in oil only
Circulating Currents; Laminations; Bad Contacts

88
88
Triangle Method
Cases of thermal faults in paper
Brownish Paper; Carbonized Paper; Not Mentioned

89
89
Triangle Method FAQs
How are corona PDs, which form a lot of Hydrogen, H2, identified in
the Triangle without using this gas?
Answer: In such faults, Methane, CH4 is indeed formed in

smaller amounts than Hydrogen, H2 (typically 10 to 20 times
less), but which can still be measured easily by DGA
Low Energy Partial Discharge
100
80
60
40
20
0
CO
H2
CH4
C2H6
C2H4
C2H2

90
90
Triangle Method FAQs

In the Triangle method, why not use Hydrogen, H2 rather
than Methane, CH4 to represent low energy faults?
Answer: Because CH4 provides better overall diagnoses for all types of
faults
A possible explanation
Is that H2 diffuses much
more rapidly than
hydrocarbon gases from
transformer oil.
This will affect gas ratios

using H2 but not those
using hydrocarbon gases.

91
91
Using the Triangle Method
If, for example, the DGA lab results are:

Methane, CH4 = 100 ppm
Ethylene, C2H4 = 100 ppm
Acetylene, C2H2 = 100 ppm
First calculate: CH4 + C2H4 + C2H2 = 300ppm
Then calculate the relative % of each gas:

Relative % of CH4 = 100/300 = 33.3 %
Relative % of C2H4 = 100/300 = 33.3 %
Relative % of C2H2 = 100/300 = 33.3 %
These values are the triangular coordinates to be used on each side of the
triangle
To verify that the calculation was done correctly, the sum of these 3 values
should always give 100%, and should correspond to only one point in the
triangle
92
92
Triangle Method Example

93
93
The calculation of triangular coordinates can easily be done manually,

or with the help of a small algorithm or software
Errors are often made when developing such an algorithm, so check it

first with the free algorithm available. (duvalm@ireq.ca)
For those familiar with computer graphics, it is also possible to

develop a software displaying the point and the fault zones
graphically in the triangle
Software from vendors is available for that purpose
An example of the software follows, Courtesy of Serveron

94
94
Note: This is the same data as shown using Rogers Ratio example
95
95

The Triangle, being a graphical method, allows you to
easily follow the evolution of faults with time

The previous example shows evolution from a thermal
fault to a potentially much more severe fault such as D1

96
96
- Fault Severity The most severe faults:

Faults D2 in paper and in oil (high-energy arcing)
Faults T2-T3 in paper (>300C)
faults D1 in paper (tracking, arcing)
faults T3 in oil (>700C)
The less severe faults:
Faults PD/ D1 in oil (sparking)
Faults T1 in paper (<300C)
Faults T2 in oil (<700C)
Are difficult to find by inspection
A fault in paper is generally considered as more serious
than a fault in oil only, because paper is often placed in a

HV area (windings, barriers)
97
97
Fault Method Comparisons

% Correct
Diagnoses
% Unresolved
Diagnoses
% Wrong
Diagnoses
IEEE Key Gas

Method
42
58
IEEE
Rogers
Rogers
Ratio
Ratios
62
33
71
26
77
15
96
Doernenburg
Ratios
IECBasic
Gas Gas
IEC
Ratio
Ratios
IEC Duval
Triangle

98
98
CO2/CO Ratio
CO2/CO ratio used to detect paper related faults
If <3, it is a strong indication of a fault in paper, either a hot spot or

electrical arcing of T>200C
If >10, it indicates a fault of temperature T<150C
CO2 & CO tend to move in and out of the paper insulation as the
temperature of the oil varies. CO2 is highly soluble and accumulates;
CO is less soluble and tends to escape.
The CO2/CO ratio, however, is not very accurate, because it is also

affected by the background of CO2 and CO coming from oil
oxidation and insulation aging.
The amounts of furans in oil may also be used in some cases to

confirm paper involvement, however, the interpretation of results
is often difficult

99
99
Other Useful Gas Ratios

O2/N2: a decrease of this ratio indicates
excessive heating
Acetylene/ Hydrogen, C2H2/H2;
A ratio >3 in the main tank indicates contamination by the LTC

compartment

100
100
Gassing not related to faults in service

Catalytic reactions on metal surfaces:
Formation of H2 only
Stray gassing of oil:

The unexpected gassing of oil at relatively low
temperatures (80 to 200C):
Gassing of the T1 or T2 type

101
101
Any questions about what was just presented?

102
102

103
103

104
104
Accuracy and Gas Levels
IEC & CIGRE recommendation
DGA diagnosis should be attempted only if gas concentrations or

gassing rates are high enough to be considered significant
Low gas levels may be due to contamination or aging of

insulation, not necessarily to an actual fault.
Limitations to Accurate Diagnostics
First limit - Lab Accuracy & Gas Concentrations
Second limit - Typical Values & Diagnosis

105
105
Accuracy and Gas Levels

First limit: LAB ACCURACY & Gas Concentrations
Accuracy is defined here as the deviation from a prepared known

value
The accuracy of the average CIGRE /IEC lab (including in the

US) is ~ 15% at medium (routine) gas concentrations (> 10
ppm for hydrocarbons)

106
106
Lab Accuracy & Gas Concentrations

(medium gas concentrations of >10 ppm)

107
107
Lab Accuracy & Gas Concentrations

(low gas concentrations of <10 ppm)
The accuracy of the average laboratory decreases to ~ 30% at 6
ppm, and 100% near the lab detection limit (2 ppm)

108
108
Lab Accuracy & Diagnostics
109
109
Effect of lab accuracies of 15% (in RED) and 30% (in BLUE) on DGA
diagnosis uncertainty
When an area of uncertainty crosses several fault zones, a reliable diagnosis

cannot be given
This
is particularly true for lab accuracies > 30%
Diagnosis uncertainties corresponding to lab inaccuracies of 15, 30, 50

and 75 % (at 10, 6, 4 and 3 ppm for the average lab):
This applies not only to the triangle but to all diagnosis methods
110
110

The actual accuracy of your laboratory can be obtained by
using gas-in-oil standards, which are available

commercially for instance from Morgan Schaffer
(Truenorth)
Below 15% of lab accuracy, a calculation of diagnosis
uncertainty should be done, and commercial software are

available for that purpose, for instance from Delta-X
Research (TOA4)

111
111
Typical Values and Diagnosis

Second limit: TYPICAL VALUES
Typical Values are not gas Limit Values
Typical Values are not the same for all Transformer Systems (may
change slightly from system to another).
Typical Values correspond to a given percentile of the population

of DGA results
Typical Values should be calculated for a specific Transformer

System (Utility specific)
Typical Values observed nationwide or worldwide may be used,

but with some caution.

112
112

Second limit: TYPICAL VALUES
Diagnosis only attempted when gas concentrations levels and

rates of gas increase is above Typical Values.
When gas levels are below Typical Values, gas concentrations

may be too close to detection limits.
Below Typical Values, the IEC recommends to use normal

sampling frequency (monthly, semi-annual, etc.,..) and not to
attempt a diagnosis.
Above Typical Values, the IEC recommends to use increased

sampling frequency (e.g., monthly or weekly) and a DGA
diagnosis may be attempted.

113
113

- 90% Typical Values
90% Typical Values are represented;

Typical Gas Concentration (TGC) levels
Typical Rate of Gas Increase (TRGI) levels.
TGC and TRGI values are the calculated values from a specific
population of transformers in service.
Reference - CIGRE Brochure # 296, Recent developments in DGA

Interpretation, Joint Task Force D1.01/A2.11, June 2006 (available
from publications@cigre.org)

114
114

- 90% TGC Values
Typical Gas Concentration (TGC) levels observed by CIGRE in

power transformers, in ppm:
C2H2
All transformers
No OLTC
Communicating
OLTC
2-20
60-280
H2
CH4
C2H4
C2H6
CO
CO2
50150
30130
60280
2090
400600
380014000
No OLTC refers to a transformer not equipped with

an OLTC or has a LTC not communicating with or
leaking to the Main Tank
Communicating OLTC means that some oil and/or gas communication
is possible between the OLTC compartment and the Main Tank of a
Transformer.
Reference; Table 3, pg 5; CIGRE Brochure #296
The data shown here represents 25 Electrical Networks worldwide,

including more than 20,000 Transformers
115
115

- 90% TGC Values
Typical Rate of Gas Increase(TRGI) levels observed by CIGRE in

power transformers, in ppm/yr
C2H2
All transformers
No OLTC
Communicating
OLTC
0-4
H2
CH4
C2H4
C2H6
CO
CO2
35132
10120
32146
590
2601060
170010,000
No OLTC refers to a transformer not equipped with

an OLTC or has a LTC not communicating with or
leaking to the Main Tank
21 - 37
Communicating OLTC means that some oil and/or gas communication

is possible between the OLTC compartment and the Main Tank of a
Transformer.
Reference; Table 3, pg 5; CIGRE Brochure #296
The data shown here represents 25 Electrical Networks worldwide, including

more than 20,000 Transformers
116
116

- 90% TGC & TRGI Values
The large ranges of TGC & TRGI values are an indication of the
differences in operation practices of individual Utilities.
Values in the preceding tables are from Core-Type Xfmr Shell type
values are likely to be higher.
Each individual Utility should preferably calculate its own specific

TGC & TRGI values.
To calculate the TGC & TRGI values, separate out the different types
or sizes of transformers to be evaluated into specific groups or
classifications.
Each transformer group and classification will have its own TGC &
TRGI ranking or 90% values
117
117

- 90% Typical Values -
Cumulative curves of DGA results (concentrations or gassing rates)

% 100
90
50
10
100
90% typical concentration value
1000 ppm

118
118

- 90% Typical Values Calculate TGC Concentration Values - Example
0
0
0
1.1
1.7
2.3
7.4
7.6
16.3
18.8
22.8
23.6
24
30.6
51.0
53.8
0
0
0
0
0
0
0
1.5
6.2
9.4
10.2
19.9
30.5
49.5
52.6
63
0
0
0
0
0
0
0
0
0
2.9
3.3
5.4
7
9.9
11.7
18.4
0
0
0
0
0
0
0
0
1.8
2.5
15.9
39.5
49
76.2
76.4
76.7
Carbon
Monoxide, CO
0
0
21.5
26.3
29.1
52.9
56
121.5
135.1
143
189.4
222.1
239.4
282.3
296.8
306.9
87.4
76.4
22.8
118.3
342.1
8426.1
2.1
136.5
138.6
186.8
1867.8
55.8
5930.9
541.2
1078.2
582.3
1760.1
9457.3
19661.7
3.2
5.7
50 - 150
30 - 130
60 - 280
20 - 90
400 - 600
3800 - 14000
2 - 20
Hydrogen, H2
90% TGC
90% TGC
(CIGRE)
Methane, CH4 Ethylene, C2H4 Ethane, C2H6
Carbon
Dioxide, CO2
177.3
258.6
487.4
494.9
662.3
1206
1263.6
1289.2
1748.4
2812.4
3369.8
4604.5
4639
6820.6
7244.1
8381.3
119
119
Acetylene
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

- 90% Typical Values Calculate TRGI Gassing Rates Values - Example
Hydrogen, H2
Carbon Monoxide, CO
Carbon Dioxide, CO2
Acetylene, C2H2
1st
2nd
1st - 2nd
1st
2nd
1st - 2nd
1st
2nd
1st - 2nd
1st
2nd
1st - 2nd
24.8
26
1.2
22.3
22.6
0.3
47.5
51.5
6.7
5.8
0.9
262.2
248.9
13.3
81.7
82.8
1.1
4358.5
4319
39.5
133.9
135.1
1.2
30.1
28.5
1.6
123.7
168
44.3
30.3
32
1.7
28.1
30.1
2476.5
2563.5
87
22.8
20.9
1.9
2.1
2.1
134.1
223.8
89.7
145.5
142.9
2.6
2.4
2.4
138.1
229.3
91.2
291.8
294.9
3.1
3.7
3.7
31.1
161.3
130.2
303.4
328
24.6
4.8
4.8
2.9
177.6
174.7
1578.9
1615
36.1
11.5
16.3
4.8
316
494.8
178.8
517
573.8
56.8
24
32.4
8.4
9751.9
10095
343.1
304
367
63
2.3
0.1
15.4
2.8
12.6
607
1011
404
1160.2
1254.7
94.5
2.2
2.4
3.1
0.7
4.5
23.2
18.7
872.7
1281.4
408.7
152.8
298.8
146
7.3
6.6
0.7
28.6
50.2
21.6
17666.6
17120.6
546
6379.4
6627.5
248.1
2.2
1.2
144.3
172
27.7
4382
5069
687
326.2
668.2
342
1.5
1.5
48.3
11.7
37
6338.5
7110.4
772
489.2
1524.9
10367
2.2
2.2
140.2
91.8
48.4
6202.9
7296.9
1094
2324.6
5435.9
3111.3
6.4
1.4
186.2
134.1
52.1
6642.1
8622.1
1980
34
3426
3392
1.8
10.4
8.6

120
120

- 90% Typical Values Influences on Typical Values:
Transformer Age
Values are significantly higher in young equipment (suggesting some

unstable chemical bonds in new oil and paper)
Values are a bit higher in very old equipment
Transformer Type
Values are higher in shell-type and shunt reactors (operating at
higher temperatures), lower in instrument transformers
Typical values are very similar in air-breathing and in sealed or

nitrogen blanketed equipment, contrary to a common belief in the
US
Transformer Oil Volume

Values are not affected by oil volume (suggesting that larger faults
are formed in larger transformers)
121
121
90% Typical Concentration values

observed in the US
Condition 1
122
H2
CH4
C2H4
C2H6
C2H2
CO
CO2
TDCG
IEEE-2008
100
120
50
65
350
2500
720
Weidmann
96
88
57
79
613
5990
1058
APS
80
45
70
30
950
1250
GE
80
50
73
28
950
1183
IEC/CIGRE
100
80
170
55
500
8900
908

- 90% Typical Values When DGA results in service reach typical values:
A diagnosis may be attempted to identify the fault (if

lab accuracy is good enough)
The equipment should not be considered at risk
However, it should be monitored more frequently by

DGA

123
123
What are your current typical (normal) values for the
different gases?

124
124

125
125
Gas Levels above Typical Values

Risk of Failure in Service
Probability of Failure in Service (PFS);
Probability of having a failure related event in service, such as any
tripping, fault gas alarm, fire, etc. event
PFS has been defined as;
PFS = Number of DGA analyses followed by a failure

related event/Total number of Analyses@ each gas
concentration value

126
126

Risk of Failure in Service
Probability of having a failure-related event (PFS, %) vs. the
concentration of C2H2 in ppm at HQ
90
98
99 Norm, in %
Pre -Failure
Value
PFS, in %
100
200
300
400
ppm

127
127

Pre-Failure Values
The PFS remains almost constant below and above the
90% typical value, until it reaches an inflexion point on the

curve (Pre-Failure Value)
DGA monitoring should be done more and more
frequently as gas concentrations increase from typical to

Pre-Failure Value
Time to reach the Pre-Failure Value is unknown, could be
Hours, Days, Weeks or Months.

128
128
The IEC/CIGRE Approach

(Gas Concentrations)
129
Probability of having a failure-related event in service (PFS)

in %, vs. the concentration of all gases in ppm. T = 90%
typical value; P = pre-failure value
The IEC/CIGRE Approach

(rates of Gas Increase)
Probability of having a failure-related event in service

(PFS) in %, vs. the rate of increase of all gases in ppm/ yr.
130

Pre-Failure Values
Pre-Failure Value concentrations were found by CIGRE to
be surprisingly close on different networks:
H2
CH4
C2H4
C2H6
C2H2
CO
2401320
270460
700990
7501800
310600
9843000
(in ppm)
This suggests that failure occurs when a critical amount of
insulation is destroyed
131
131

Pre-Failure Gassing Rates
Pre-failure rates of gas increase in power transformers at
CIGRE, in ppm/day
C2H2
H2
CH4
C2H4
C2H6
CO
CO2
0.5
11
NS
NS
Note: This is an unpublished chart

132
132

Caution and Warning Values
In-between typical and pre-failure values, specific caution and

warning values can be defined, depending on the tolerance to risk of
the maintenance personnel, and on the maintenance budget available
For example, higher values may be used when the maintenance

budget is low, and lower values in the case of strategic equipment
CIGRE caution and warning values are significantly different from

those proposed by IEEE and are available upon request(from
duvalm@ireq.ca)

133
133
Sample Intervals & Concentration Limits
Sampling intervals and gas concentration limits in ppm

calculated for an average US power transformer
134
Numerical values of sampling intervals and

gas concentration limits in ppm calculated
for an average US power transformer
Concentration
CH4
C2H4
C2H6
C2H2
CO
CO2
TDCG
Sampling intervals
Nuclear
Transmission
Yearly
Typical
88
69
65
53
2.5
780
5990
1058
6 months
Level 2
143
104
117
102
983
9829
1546
Monthly
164
116
137
123
11
1048
11274
1717
214
145
188
175
22
1186
14695
2104
237
158
212
201
29
1245
16294
2277
366
226
355
360
84
1524
25161
3178
388
237
380
388
97
1566
26661
3322
725
400
800
900
450
2100
50000
5380
Level 3
Level 4
Pre-failure
135
H2
Monthly
Weekly
Weekly
Daily
Daily
Hourly
Hourly
Numerical values of sampling intervals and

rate of gas increase limits in ppm/ year
calculated for an average CIGRE/ IEC power transformer
Gassing rate
CH4
C2H4
C2H6
C2H2
CO
CO2
TDCG
Sampling intervals
Nuclear
Transmission
Yearly
Typical
83
65
89
47
660
5850
946
6 months
Level 2
151
141
180
132
1408
12472
2049
Monthly
179
175
218
176
1737
15382
2539
247
266
319
308
13
2607
23067
3841
280
313
369
382
17
3054
27012
4513
475
620
686
951
42
5940
52492
8896
509
679
745
1074
47
6491
57351
9738
1095
1825
1825
4015
182
17000
150000
26000
Level 3
Level 4
Pre-failure
136
H2
Monthly
Weekly
Weekly
Daily
Daily
Hourly
Hourly
Sampling Intervals based on Combined Gas

Rate and Gas Concentration Levels of
Individual Gases (IEC/CIGRE approach)
Rate
Level #
4
4
4
4
3
3
3
3
2
2
2
2
1
1
1
1
137
Conc.
Level #
4
3
2
1
4
3
2
1
4
3
2
1
4
3
2
1
Daily
Sampling Intervals based on Combined

Gas Rate and Concentration Levels
Weekly
Monthly
Quarterly
Yearly
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
TDCG Procedure
Focuses on Total Dissolved Combustible Gas (TDCG) content
TDCG, is the sum of all combustible gases, (H2+CH4+C2H2+

C2H4+C2H6+CO) dissolved in the oil
The table provides guidelines concerning gas levels in
transformers and suggests actions in terms of sampling

intervals and operating procedures

138
138
IEEE Guide for the Interpretation of Gases

IEEE Std C57.104-1991
139
IEEE Guide for the Interpretation of Gases

IEEE Std C57.104-1991
140
What are your current caution/ warning/ alarm values for
the different gases?

141
141
Triangle 2 for LTCs of the Conventional Oil-type

(UZB, UBB, URS, UTT, URT, 550, 394, TLH, TC, LR, LRT, etc)
: Normal operation; : Severe coking; : Light coking; :

Heating; : Strong arcing D2; : Arcing D1
142
Triangle 2 for LTCs of the Conventional

Vacuum Type (LRT, UVT)
: no
143
inspection made
Triangle 3 for Non-Mineral oils
FR3
144
Mineral oil
Triangle 4 for Low-Temperature Faults in

Transformers
: Corona partial discharges; : Stray gassing at

120C; ; Stray gassing at 200C; ; : Hot spots with
carbonization of paper; ; : Overheating (T < 250C).
145
Triangle 5 for Low-Temperature Faults in

Transformers
: Corona partial discharges; : Stray gassing at

120C; ; Stray gassing at 200C; ; : Hot spots with
carbonization of paper; ; : Overheating (T < 250C).
146
Triangle 6 for Low-Temperature Faults in FR3

Oils
stray gassing at T < 150 C;;

stray gassing at T > 200C;;
147
gas formation in Alliant transformers
Triangle 7 for Low-Temperature Faults

in FR3 Oils
stray gassing at T < 150 C;;

stray gassing at T > 200C;;
148
gas formation in Alliant transformers

149
149
How to proceed with DGA results coming from the

laboratory ?
1. Examine DGA laboratory values for errors or
inconsistencies.
2. Attempt a fault diagnosis when possible
3. Evaluate the severity of the fault
4. Decide on appropriate actions on the equipment

150
150
#1 - Examine DGA laboratory values

Eliminate zero values (replace them by the detection
limits of the laboratory, e.g., 2 or 1 ppm)
Compare with previous values in ppm on the same
transformer (DGA history)
Check for inconsistencies which might indicate
contamination during sampling or a laboratory analytical

error. Consider suspect values with caution or discard
them
151
151
#2 - Attempt a fault diagnosis

Compare with in-house typical or normal values of
concentration and gassing rate, or compare with published

values (e.g., from IEEE and CIGRE)
If measured DGA values are above typical/ normal values, a
fault diagnosis may be attempted
If measured values are below routine concentration values (10
ppm for hydrocarbons), a fault diagnosis should be attempted

only after calculating the uncertainty on the diagnosis, based
on the accuracy of the laboratory at these low concentration
levels
152
152
#3a - Evaluate the severity of the fault

The severity of the fault will depend on:
The rate of gas formation
The concentration of gases
How far measured values are from typical values, and how close
they are to pre-failure values
The severity of the fault will also depend on:
The nature of the fault (electrical or thermal)
The location of the fault (paper or oil only)

153
153
#3b - Evaluate the severity of the fault
Gas formation not related to faults (e.g., catalytic reactions of oil with
metals or stray gassing of oil) is of no particular concern
The CO2/CO ratio (incremented) can be used to determine paper

involvement in the fault
The C2H2/ H2 ratio can be used to detect contamination of the main

tank by an LTC leak
To fully assess the severity of a fault, it may be useful at this stage to

seek the advice of experienced DGA experts

154
154
#4 - Actions on the equipment
Increase the frequency of oil sampling
(as values move from typical values to pre-failure values)
Determine the dependence of load on gas formation, and adapt load

accordingly
Consider complementary tests
(acoustic detection of partial discharges, detection of hot spots by infra-red

cameras, etc)
For critically located or severely affected equipment, install on-line

gas monitors
For the most severe cases, plan the replacement of the transformer
and/or its removal from service and inspection
Engineering judgment should be exercised before taking actions on

the equipment (or deciding not to take actions)

155
155
Any questions about what was just presented?

156
156

157
157
Benefits of On-Line DGA
Detects both gradual and sudden trends in all gases
Correlates gassing events with external events such as transformer

load, oil temperature, LTC changes, etc.
Provides historic trail for delayed analysis of gassing events
May provide diagnosis on-line

158
158
Advantages of On-Line DGA

Oil is sampled automatically via closed-
loop GC:
No risk of human intervention
Repeatable sampling technique

No atmospheric exposure
Data is collected up to hourly:
Resulting in faster, more accurate

determination of trends
All 8 fault gases + moisture are monitored
and correlated with oil temperature and

load

159
159
On-Line DGA
Serveron Transformer Monitor

160
160
On-line DGA
On-line gas monitors
Are best suited for measuring rates of gas increase (trends)
Will detect faults between regular oil samplings
May now also provide on-line diagnosis

161
161

162
162
Serveron Story #11
Three-phase 700 MVA 400 kV GSU transformer

Hints: - Increasing Methane (CH4), Ethane (C2H6), Ethylene (C2H4)
- Sudden
increase in
Acetylene
(C2H2permission.
),
Serveron Corporation
2007.
No use without
163
163
- Transformer refurbished and re-commissioned 10/27/03
Serveron Story #11 Analysis:
The Duval Triangle is a DGA tool included in the IEC 60599 Gas Guide.
1. The Duval Triangle shows problem evolving from a

T3 Thermal Fault to D1 Discharge of low energy
HV lead corona ring

shield connection
(note burn mark on corona ring
material & eroded bolt)
2. Intermittent grounding was provided by the fastening bolt causing a transient potential
rise and subsequent discharges occurring between the corona ring and the main tank
ground point.
3. An on-site
repair
was2007.
performed
andpermission.
the transformer
Serveron
Corporation
No use without
was returned to service.
164
164
Serveron Story #12
3-phase, 1100 MVA, 345 kV GSU transformer

165
165
Serveron Story #12 Analysis:
The Duval Triangle is a DGA tool included in the IEC

60599 Gas Guide.
Rogers Ratios are included in IEEE 57.104 Gas

Guide (similar to Basic Gas Ratios in IEC-60599)
1. Both the Duval Triangle and Rogers Ratio analysis shows the fault condition is in T2
indicating a thermal problem getting worse in the range of 300C to 700C
2. Combustible gas levels were rising very quickly, exceeding preset rate of change limits.
Transformer load reduction began approximately 32 hours after levels began to change
and was fully de-energized within approximately 52 hours
3. Root cause of the problem was insulation design issues around HV and LV leads
166
166
Serveron Story #12

167
Serveron Story #12

168
How to proceed with DGA results coming from the online gas monitor
No need to check for errors or inconsistencies in DGA
values
Diagnosis may be available on-line using the main
diagnosis methods
What should still be evaluated is the reliability of the
diagnosis, depending on gas level, and the severity of the

fault as in the case of manual DGA
Also, decide on appropriate actions on the equipment as
in the case of manual DGA, once the diagnosis is

confirmed

169
169
Your Laboratory DGA

Examination of your DGA data from selected transformers
to get a diagnosis

170
170
The TM series monitors
Transformer Mounted

Practicing DGA 09072010

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Practicing DGA 09072010

Transféré par

Droits d'auteur :

Formats disponibles

1

A BPL GLOBAL COMPANY

On-Line and Lab DGA Case Studies

Serveron Corporation 2008. No use without permission.

A BPL GLOBAL COMPANY

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

Hillsboro, USA Headquarters

Branch offices in Beijing and Brussels

EPRI and proprietary GC based

Over 1000 On-line DGA monitors

A BPL GLOBAL COMPANY

Dr Michel Duval obtained a B.Sc. in chemical engineering in 1966

In the field of DGA, M. Duval:

Is well-known for his Triangle method of DGA interpretation, used

Has established the typical levels of gas formation observed in various

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

In the field of electrical insulating oils, M.Duval has researched and

M. Duval holds 16 patents and is the author of more than 75 scientific

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

Transformer Asset Management

Transformers are a critical and costly element in the electrical grid

Unplanned failures at any point in the transformer lifecycle have

A BPL GLOBAL COMPANY

Failures They Happen!

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

$17 million loss in

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

Increasing Failure Rates

Sergi Transformer Explosions in the United States

All catastrophic failures in any USA news media

Reported number of failures is increasing

Total % of explosive failures is increasing

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

Increasing Failure Rates

Projected rates will reach unacceptable levels if nothing is done to

A BPL GLOBAL COMPANY

Factors Leading to Increased Failures

>42 yrs, increasing 0.7 yrs./yr.

Age itself is not a cause of failure

Increased energy demand: Transformer peak and average

loading has increased

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

Failures due to internal stress, weather,

Failures initially categorized into areas

Short or long-term overloading

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

A BPL GLOBAL COMPANY

What monitoring, other than DGA, do you perform on your

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

Serveron Corporation 2007. No use without permission.

A BPL GLOBAL COMPANY

Monitoring of Gases in Transformers

down due to stress, gases are