International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882

Volume 4, Issue 10, October 2015

ONLINE CONDITION ASSESSMENT OF TRANSFORMER OIL FOR

INCIPIENT FAULTS DETECTION
Kamini Devi1, Mrs. Shimi S.L.2
1

ME Scholar, Electrical Engineering, NITTTR, Chandigarh.

Assistant Professor, Electrical Engineering Department, NITTTR, Chandigarh.

ABSTRACT
Transformer failure can occur as a result of different
causes and conditions. It is found that the main cause of
transformer failure is the damage occurring in insulating
oil and paper .Hence to achieve long life of transformer,
it is necessary to monitor the condition of insulating oil
and paper frequently at least on monthly basis.
Condition monitoring increases the transformer life and
reliability by making proactive decisions in case of any
failures. Also condition monitoring reduces the
maintenance cost of the system.
Oil of transformer serves as both coolant as well as
insulator. The degradation of oil under electrical and
thermal stresses influences electrical, chemical and
physical properties of the oil. Deterioration of insulating
oil and paper leads to the production of gases which
gradually get dissolved in oil. The failure in transformers
is due to deterioration of insulation over time. If these
failures can be predicted with some degree of
confidence, sudden failures and interruption in power
supply can be minimized. In this paper various
techniques for online condition monitoring of oil and
how incipient faults can be detected from the various
dissolved gases concentration has been discussed. Key
gas method, Rogers Ratio method, National Standard
method, and Duvels Triangle method implementation is
discussed for the fault detection
Keywords - Transformer, Key Gas, Rogers Ratio,

Duvels Triangle
1. INTRODUCTION
Transformer plays an important role in the transmission
and distribution of electric power to the consumers.
Moreover it is one of the critical and costly parts of the
whole Power grid, but like all other electrical machines,
it has too limited life .Different causes and conditions
results in transformer failure but after a lot of survey and
analysis it has been found that insulation failure damage
the transformer mostly. The insulation used in the case

of electrical conductors in a transformer is varnish or

enamel in dry type transformers. In large transformers to
improve the heat transfer characteristic, the conductors
are insulated using unimpregnated paper or cloth and the
whole core winding assembly is immersed in the oil
tank.
The transformer oil thus has dual role. It is an insulator
and also a coolant. The porous insulation around the
conductor helps the oil to reach the conductor surface
and extract the heat. Annular Bakelite and oil ducts are
used as part of insulation between the windings. The oil
used in the transformer tank should be free from
moisture or other contamination to be used as an
insulator. Under the influence of electrical and thermal
stresses oil gets decomposed and gases are formed which
gradually get dissolved in the oil. These gases are known
as key gases. The quantity of key gases concentration
can help in the identification of various faults.

2. PRODUCTION OF GASES IN OIL

High thermal and electrical stresses on transformer oil
leads to the breaking of chemical bonds between the
atoms that make up the hydrocarbons molecules in the
oil. The degradation of oil and paper leads the formation
of gases which will dissolve in the oil entirely or
partially [1]. The combustible gases are formed in large
quantities due to undesirable faults occurring inside the
transformer. These gases formed inside the transformer
termed as Key Gases. Several key gases generated
including Hydrogen (H2), Methane (CH4), Ethane
(C2H6), Ethylene (C2H4), Acetylene (C2H2), Carbon
Dioxide (CO2) and Carbon Monoxide (CO) There are
two kinds of molecular bonds exists between the gases
,one is between carbon and hydrogen atoms and other is
between carbon atoms. The energy required to break or
create the molecular bonds is different. The energy and
temperature required to create or break the bond between
same atoms (C-C) is greater than the energy and
temperature required for different atoms C-H).Also
energy required to make or break triple bonds greater
than double and single bonds. Concentration of gases in
the oil can detect many incipient faults in the
www.ijsret.org

991

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 4, Issue 10, October 2015

transformer. Normal values of dissolved gases indicates

no internal faults but when the concentration of
dissolved gases goes beyond the normal limits indicates
some incipient faults.
The table for the different gases limits is derived from
the information provided within ANSI/IEEE C57.104
[2]. Concentration of all the gases is given in ppm level.

Partial Discharges can also occur along the surface of

solid insulating materials if the surface tangential electric
field is high enough to cause a breakdown along the
insulator surface. Partial Discharges in transformer oil
leads to hydrocarbon cracking. It generates large amount
of hydrogen and free radicals.

Table 1 Normal and Action limit of dissolved

gases [3]

3.2 Discharges of Low Energy (D1)

Partial Discharges of low energy which usually occur in
gas filled voids surrounded by oil. It carbonized the
paper and oil insulation in transformer .It produces small
amount of carbon particles in oil and creates voids in
paper insulation.

Normal
Limit
Action
Limit

H2
150

CH4
25

C2H4
20

C2H2
15

C2H6
10

CO2
10,000

CO
500

1000

80

150

70

55

15000

1000

Hence by identifying the gases concentration

transformer can be prevented from many internal faults.
Corrective action should be taken before the gases
reached the action limits. In case values goes beyond the
action limits transformer should be disconnected from
the service and oil should be replaced by a fresh oil.

3. TYPES OF FAULTS DETECTABLE BY

INCEPTION OF GASES
Identification of transformer oil gases concentration is
the best tool for indicating various types of faults in the
transformer. There are various types of internal faults
occurred in the transformer. Some of them are:
Partial Discharge
Discharges of low energy
Discharges of high energy
Thermal fault (T1)
Thermal fault (T2)
Thermal fault (T3)
Thermal and Electrical Fault (DT)
3.1 Partial Discharge
Partial discharge (PD) is a localized dielectric
breakdown of a small portion of a solid or fluid electrical
insulation system under high voltage stress, which does
not bridge the space between two conductors. While a
corona discharge is usually revealed by a relatively
steady glow or brush discharge in air, partial discharges
within solid insulation system are not visible. Partial
discharges within an insulating material are usually
initiated within gas-filled voids within the dielectric.

3.3 Discharges of High Energy (D2)

Partial Discharges of high energy in paper and oil results
in extensive damage to paper or large formation of
carbon particles in oil, metal fusion, tripping of
equipment and gas alarms.
3.4 Thermal Fault (T1)
A small amount of decomposition occurs at normal
temperature. As the fault temperature rises, the
formation of the degradation gases change from methane
(CH4) to ethane (C2H6) to ethylene (C2H4).Faults which
occur at temperature less than 300oCcategorized as
thermal fault, T1.These type of faults can be noticed
with change in color of paper. The paper generally turns
into brown color and also get carbonized.
3.5 Thermal Fault (T2)
Faults which occur at temperature greater than 300OC
but less than 700OC are categorized as thermal fault,
T2.High temperature results in generation of carbon
particles in both paper and oil.
A fault at higher temperatures (>300deg/C) produces
Ethylene. The higher the temperature becomes, greater
will be the production of ethylene.
3.6 Thermal Fault (T3)
Faults which occur at temperature greater than 700oC are
categorized as thermal faults T3 .These leads to the
generation of excess carbon particles in insulating paper
and oil.
3.7 Thermal and Electrical Fault (DT)
Sometimes both electrical and thermal faults occurs
inside the transformers. These faults accelerate the
degradation of oil and paper.
www.ijsret.org

992

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 4, Issue 10, October 2015

4. ONLINE MONITORING METHODS

The offline method of oil monitoring constitutes the oil
analysis in laboratory through gas chromatography
technique. Frequent monitoring cannot be done by
offline methods and chances of proactive decisions in
case of failures are very less. Hence todays demand is
for online monitoring methods of oil. Generally various
sensors are used for finding the concentration of
dissolved gases in the oil. These sensors can be simple
or complex. These sensors serve as dissolved gas
monitors for the transformer oil.
Examples are GE Hydran and the Morgan Schaffer
Calesto
4.1 Hydran Sensor
The simple Hydran sensor consist of permeable
membrane that allows the key gases to come in contact
with a miniature gas detection sensor operating as a fuel
cell [4] .It allows the detection of hydrogen gas
concentration. Carbon Monoxide is also detected with an
attenuation of 18%.A composite value of two gases with
additional traces of acetylene and ethylene is provided
and is known as Hydran reading.

oil. The calesto moisture measurement have the

advantage that transformer oil is always heated to a
constant and known temperature. This is necessitated by
thermal conductivity measurement of hydrogen but
allows a percentage of relative saturation of moisture to
be easily calculated.

5. FAULT DETECTION METHODS

At initial stage generally low energy faults at low
temperature occur but failure of early detection of the
faults leads to the conversion of low energy faults into
high energy serious faults. It is necessary to detect the
faults at the early stage for preventing the transformer
shut down from the service. If we are able to find out the
trend of rate of increase in individual gases and the
faults, information about the health of transformer can be
collected.
Different methods can be implemented for the detection
of faults based on the concentration of gases such as
Key gas method
Ratios methods or Basic gas ratios method and
National Standards method
Duval Triangle method
5.1 Key Gas Method
The gases generated in the transformer oil are known as
key gases. Key gases can be used for qualitative
determination of fault types, based on which gases are
typical or predominant at various temperatures. The gas
present in high percentage will indicate specific faults.

Fig. 1 Hydran Sensor [5]

LIMITATION OF HYDRAN SENSOR
It is highly sensitive to the combustible gases. A high
content of acetylene and low hydrogen content can give
the same reading as a low acetylene and high hydrogen.
These are totally two different situations
4.2 Morgan Schaffer Calesto
The capillary tube probes of the calesto extracts only
hydrogen from the oil and insensitive to other
combustible gases. One extra feature is added in calesto
is the measurement of moisture [6]. Moisture is the
indicator of the health of the insulating paper in the
transformer. In calesto, moisture is measured by a
specialized integrated circuit which is in contact with the

Table 2 Fault Indicator Gases [7]

GAS
FAULT DETECTED
DETECTED
H2
Corona effect Arcing, overheated oil
CH4

Arcing, serious overheated oil

C2H6

Thermal fault like Corona &

overheated oil
Thermal fault, local or overheated oil,
Corona, arcing

C2H4
C2H2
CO
CO2
O2 & N2

www.ijsret.org

Electric fault like arcing and sparking,

Severely overheated oil
Overheated cellulose decomposition
Arcing if fault involves cellulose
Cellulose decomposition
Indicator of system leaks over
pressurization or changes in
temperature.

993

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 4, Issue 10, October 2015

Following graph shows the concentration of each gas

during various types of faults in Key Gas Method.

Fig. 5 various gases concentrations during high

energy arcing [8]
Fig. 2 various gases concentrations during thermal oil
fault [8]

5.2 Ratios Method or Basic Gas Ratio Method

The International Electro Technical Commission (IEC)
Standard recognized this basic gas ratio method. Three
ratios of gases are used for the detection of faults. In key
gas method single gas present in high percentage
determines the fault whereas in ratio method three ratios
of gases will be considered for the fault detection. The
three gas ratios are acetylene/ethylene, ethylene/ethane
and methane/hydrogen.
Table 3 IEC Gas Ratio Table [1]

Fig. 3 various gases concentrations during low energy

partial discharge fault [8]

Fig. 4 various gases concentrations during thermal oil

& cellulose fault [8]

CHARACTERISTICS
FAULTS

C2H2/
C2H4

C2H2/
C2H6

CH4/H

Partial Discharges (PD)

NS

<0.2

<0.1

Discharges of low
energy (D1)

>1

>1

0.1-0.5

Discharges of high
energy (D2)

0.6-2.5

>2

0.1-1.0

Thermal Fault (T1)

(T<300o C)
Thermal Fault (T2)
(300OC<T<700oC
Thermal Fault (T3)
(T>700Oc)

NS

<1

>1

<0.1

<1

>1

<0.2

>4

>1

NS No Significance
The ratio CO2/CO indicates thermal decomposition of
paper. The ratio CO2/CO should be greater than 7 for
normal operation as the rate of production of carbon
dioxide is 7 to 8 times greater than carbon monoxide. In
case the ratio goes below or equal to 5 there must be
some problem occurring in the cellulose. It indicates that
severe thermal deterioration of cellulose.
www.ijsret.org

994

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 4, Issue 10, October 2015

5.3 National Standards Method

In this method the condition of the generator step up
(GSU) transformers and grid transformers can be
determined based on the total concentration of all the
combustible gases (TDCG) for a specific time period.
Table 4 Condition vs. Operating Time of the
Transformer [10]
Condition
Concentration of Total Combustible
Gases
OPERATING TIME

NORMAL
DUBIOUS

Type

<8 Years

GRID
GSU
GRID

<350
<500
350.<
450
500..<6
00
450<
800
650<
1000

GSU
FAULTY

GRID
GSU

DANGER

GRID
GSU

800
1000

8-15
Years
<450
<650
450.
<800
650
<1000
800.
<1600
1000
<160
0
1600
1600

>15years
<800
<1000
800<16
00
1000<1
600
1600<3
000
1600.<
3000

of the suspected fault, it is advisable to subtract out the

background concentrations, provided that the differences
are large enough for interpretation. The diagnosis should
be based on recently-formed gas if possible, and
including pre-fault gas in the diagnostic calculations can
lead to misleading results.
The faults which are detected by Duval triangle are
expressed as follows:
Partial Discharge (PD), Low range thermal fault (T1)
(T1<300Oc), Medium range thermal fault (T2)
(300oC<T2<700oC,) High range thermal fault (T3)
(>700Oc), Low energy electrical discharge (D1,) High
energy electrical discharge and Indeterminate thermal
or electrical fault (DT).

6. PROCEDURE TO USE DUVEL TRIANGLE

Step 1 Consider the three sides of the triangle having
coordinates x, y and z .Each side represents percentage
for each gas which are CH4, C2H4 and C2H2.
Step 2 Find out the concentration of each gas.
Step 3 Calculate total concentration of gases.

3000
3000

5.4 The Duvel Triangle

The Duval Triangle diagnostic method for oil-insulated
high-voltage
equipment
(mainly
transformers),
developed by Michel Duval [11]of IREQ in 1974.In this
method Concentrations (ppm) of methane (CH4),
ethylene (C2H4), and acetylene (C2H2) are expressed as
percentages of the total (CH4 + C2H4 + C2H2) and plotted
as a point (%CH4, %C2H4, %C2H2) in a triangular
coordinate system on a triangular chart which has been
subdivided into fault zones. The fault zone in which the
point is located designates the likely fault type which
produced that combination of gas concentrations.
Duval Triangle method should be applied only when
there is some suspicion of a fault, based on an increase
in combustible gas or some other suspicious symptom.
The diagnostic method itself is not a means of fault
detection. Because of the relative inaccuracy of gas-inoil concentration measurements at low concentrations,
DGA diagnostic methods, including the Duval Triangle,
should not be applied unless the gas concentrations are
well above the detection limit. If reasonably stable
concentrations of the gases were present before the onset

Step 4 Calculate percentage of each gas to the total

concentration.
Step 5 Draw CH4 % quantity parallel to C2H2 line,
C2H4% quantity parallel to CH4 line and C2H2% quantity
parallel to CH4.,
Step 6 Intersection of all three lines will indicate the
fault for the gas

Fig. 6 The DuvelsTriangle

www.ijsret.org

995

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 4, Issue 10, October 2015

7. SOFTWARE IMPLEMENTTION OF
DUVELS TRIANGLE

Table 5. Triangular Coordinates for Duvels Triangle

Zone [13]
Area

Points

% CH4

%C2H4

%C2H2

PD1
PD2
PD3

98
100
98

2
0
0

0
0
2

D11
D12
D13
D14
D21
D22
D23
D24
D25

00
00
64
87
00
00
31
47
64

00
23
23
00
23
71
40
40
23

100
77
13
13
77
29
29
13
13

DT

DT1
DT2
DT3
DT4
DT5
DT6
DT7
DT8

00
00
35
46
46
87
47
31

71
85
50
50
00
00
40
40

29
15
15
04
04
13
13
29

T1

T11
T12
T13
T14
T15

76
80
98
98
96

20
20
02
00
00

04
00
00
02
04

T21
T22
T23
T24
T31
T31
T33
T34

46
50
80
76
00
00
50
35

50
50
20
20
85
100
50
50

04
00
00
04
15
00
00
15

PD

D1

D2

T2

T3

8. CONCLUSION

Fig. 7 A flow-chart for software development of

Duval triangle on MATLAB [12]

Monitoring of transformer oil is very effective tool to

increase transformer life because transformer is the most
expensive and important equipment of the power system.
It plays an important role in transmitting and distributing
electrical power at low voltage and high voltage to the
consumers. Frequent replacement of transformer because
of insulation failure is not the better choice. Provisions
should be make for the frequent condition assessment of
www.ijsret.org

996

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 4, Issue 10, October 2015

insulating material. Offline methods including gas

chromatography test of insulating oil sample in the
laboratories is an inconvenient method and does not
provide accurate results. Moreover it leads to the
wastage of time .Online gas monitors such as Hydran
sensor and Morgan Schaffer Calesto gives accurate and
fast results. The inconveniency of sending the oil
samples to the laboratories and waiting for the results is
totally eliminated by using online methods. The
limitation of hydran sensor being sensitive only to the
combustible gases can be overcome by using multi gas
monitors which are nowadays available in the market.
Software implementation of key gas method, Rogers
Ratio method and Duvels triangle increases the
accuracy as compare to the manual method.

REFERENCES
[1] Er. Lee WaiMeng, Dissolved Gas Analysis (DGA)
of mineral oil used in transformers, The Singapore
Engineer, pp. 04-05, May 2009.
[2] IEEE Std C57.104-1991, Guide for the
Interpretation of Gases Generated in Oil-Immersed
Transformers, 1991.

and DGA Feature Extraction , Proceedings of the 7th

World Congress on Intelligent Control and Automation,
Chongqing, China, pp. 1800-1805, June 25-27, 2008.
[10] Blint Nmeth, SzilviaLaboncz and Istvn Kiss,
Condition Monitoring of Power Transformers using
DGA and Fuzzy Logic, IEEE Electrical Insulation
Conference, pp. 373- 376, 2009.
[11] M. Duval, A Review of Faults Detectable by Gasin-Oil Analysis in Transformers, IEEE Electrical
Insulation Magazine, Vol.18, No.3, pp.8, 2002.
[12] Sukhbir Singh, Dheeraj Joshi and M.N.
Bandyopadhyay, Software Implementation of Duval
Triangle Technique for DGA in Power Transformers,
International Journal of Electrical Engineering, Vol. 4,
No. 5, pp. 529-540, 2011.
[13] Sukhbir Singh and M.N. Bandyopadhyay, Duval
Triangle: A Noble Technique for DGA in Power
Transformers, International Journal of Electrical and
Power Engineering, Vol. 4, Issue-3, pp.193-197, 2010.

[3] Lynn Hamrick, Dissolved Gas Analysis for

Transformers, Neta World, pp. 01 -04, 2009- 2010.

[14] Lars E.Lundgaard, Walter Hansen, Dag Linhjell and

Terence J.Painter, Aging of oil Impregnated Paper in
Power Transformers , IEEE Transactions on Power
Delivery ,Vol. 19,No. 1,pp. 230-239,January 2004.

[4] Doncuk, VaclavMentlik and Jiri Velek, Condition

assessment of power Transformer using gas detection
methods, University of west Bohemia, pp.164-168,
2013.

[15]W.L.Chan, T.M.Chan, S.L. Pang and A.T.P. So, A

Distributed on Line HV transmission Condition
Monitoring Information System, IEEE Transactions on
Power Delivery, Vol. 12,No. 2, pp. 707-713, April 1997.

[5] Belanger , Apparatus for Detecting and Measuring

the Hydrogen content in a liquid substance ,United
States Patent 4,271,474,United States Patent and
Trademark office, Washington DC,1981.

[16] IEEE Guide for Acceptance and Maintenance of

Insulating Oil in Equipment, IEEE Std C57.106-2006
(Revision of IEEE standard C57.106-2002), pp.1 -36,
2006

[6] Morgan, Transformer Fault Detection United

States Patent 4, 112, 737, United States Patent and
Trademark Office, Washington DC, 1978.
[7] Imad-U-Khan, Zhongdong Wang, Ian Cotton and
Susan Northcote, Dissolved Gas Analysis of Alternative
Fluids for Power Transformers, IEEE, Vol. 23, Issue.
5, pp. 05-14, September/October 2007.
[8] Serveron White Paper, DGA Diagnostic Methods,
2007.
[9] HongshengHu, SuxiangQian and Jian Cao ,
Monitoring and Fault Diagnosing System Design for
Power Transformer based on Temperature Field Model
www.ijsret.org

997