Abstract: Monitoring and fault diagnosis of power transformers has always drawn the attention of all utilities worldwide. It has become important to detect the incipient fault of a power transformer at an early stage, diagnose properly and also to provide the remedies. Dissolved Gas Analyses (DGA) is widely used to detect incipient faults in oil filled power transformers. Various methods are used to interpret DGA results i.e., C.E.G.B. Rogers’s ratio, IEC and IEEE ratio codes: Rogers, Dernenburg and Duval triangle methods. In this study, a relationship between the formations of hydrocarbon gases and the specific temperature and their solubility has been presented. For this study, more than couple of 100s of faulty transformer DGA sample reports from the different utilities in India (>10 years) with fault interpretation by different methods are collected. Against the same reports the Duval triangle, a graphical manual and software (MATLAB-7.4) are crossed compared.

INTRODUCTION

Transformer is one of the most important but complex component of electricity transmission and distribution system. Much attention is needed on maintenance of transformers in order to have fault free electric supply and to maximize the lifetime and efficacy of a transformer. Thus, it is important to be aware of possible faults that may occur. It is equally important to know how to detect them early. Regular monitoring and maintenance can make it possible to detect new flaws before much damage occur.

Formation of gases in transformer oil: Mineral oils (transformer oil) are composed of saturated hydrocarbons called paraffins, whose general molecular formula is C_nH_2n+2 with n in the range of 20-40. The cellulosic insulation material is a polymeric substance whose general molecular formula is (C₁₂H₁₄(OH)₆)_n with n in the range of 300-750. Gases are formed inside an oil-filled power transformer, in that various gases begin forming at specific temperatures shown in Fig. 1.

Hydrogen and methane begin to form in small amounts around 150°C. Beyond maximum points, methane (CH₄), ethane and ethylene production goes down as temperature increases. At about 250°C, production of ethane (C₂H₆) starts. At about 350°C, production of ethylene (C₂H₄) begins. Acetylene (C₂H₂) starts between 500 and 700°C. In the past, the presence of only trace amounts of acetylene (C₂H₂) was considered to indicate a temperature of at least 700°C had occurred; however, recent discoveries have led to the conclusion that a thermal fault (hot spot) of 500°C can produce trace amounts (a few ppm).

Fig. 1:	Gas generation chart

Larger amounts of acetylene can only be produced above 700°C by internal arcing. Between 200 and 300°C, the production of methane exceeds hydrogen. Starting about 275°C and on up, the production of ethane exceeds methane. At about 450°C, hydrogen production exceeds all others until about 750-800°C then more acetylene is produced. It should be noted that small amounts of H₂, CH₄ and CO are produced by normal aging. Thermal decomposition of oil-impregnated cellulose produces CO, CO₂, H₂, CH₄ and O₂. Decomposition of cellulose insulation begins at only about 100°C or less. Therefore, operation of transformers at not >90°C is imperative.

Faults will produce internal hot spots of far higher temperatures than these and the resultant gases show up in the DGA.

Solubility of gases in transformer oil: The solubilities of the fault gases in transformer oil as well as their temperature dependence are also important factors for consideration in fault gas analyses.

It should be noted that there are almost two orders of magnitude difference between the least soluble (H₂) and the most soluble (C₂H₂) gas.

The majority of gases that are indicative of faults are also those that are in general the more soluble in the oil.

When the rates of gas generation are being followed, it is important to take into account the solubility of these gases as a function of temperature (Fig. 2). Over a temperature range of 0-80°C some gases increases in their solubility upto 79% while others decreases their solubility upto 66%.

Dissolved Gas Analysis (DGA): The DGA has become a popular technique and is successfully used for many years. The method is very sensitive and gives an early indication of incipient faults. The insulation oil used in sometimes transformer leads to degradation of insulating transformer is long chain of complex mixture of hydrocarbon compounds.

Fig. 2 :	Relative solubility (Y-axis) as function of temperature °C (X-axis)

Failures inside the produced remain in the insulating oil in dissolved state and hydrocarbon compounds. Due to smaller molecular size, many of these compounds are gasses. Types and quantity of hydrocarbon gases formed depends on the nature and intensity of the fault. The gases so if concentration goes beyond saturation level the gases come out and get collected in Bucholzs relay.

Due to dissolved gases in the transformer oil, the insulation property of this oil goes weak and lead to transformer failures. The composition and quantity of the gases generated depend on types and severity of the faults. Both these kinds of information together provide the necessary bases for the evaluation of any fault and the necessary remedial actions. Advantages that dissolved gas analyses can provide:

The regular monitoring of these dissolved gases interpret useful information about the condition of the transformer and prior information of the faults by observing the trend of the various gas contents. The relative distribution of the gases is used to evaluate the origin of the production of these gases and the rate at which the gases are formed to assess the intensity and propagation of the gases. A diagnostic code; warnings of any gas concentrations, increments, rates of change or ratios that exceed standard limits and short interpretive remarks and recommendations become proper fault diagnosis. Supported diagnostic methods for transformer DGA include the Duval triangle method and the Rogers, Doernenburg and CIGRE gas ratio diagnostics. The main Interpretation methods in fault diagnosis of power transformers used are:

Table 1:	Key faults in power transformer

Key faults in power transformers (PC57.104/D3.3, 2008; IEC 60599 {Ed.2.1}, 1999) shown in Table 1.

Duval triangle in dissolved gas analyses: Duval triangle method (Duval, 2003, 2008; Singh, 2008a, b; Akbari et al., 2008; Duval and Dukarm, 2005) shown in Fig. 3 developed empirically in early 1970s and is used by IEC (IEC 60599 {Ed.2.1}, 1999). It is based on the use of three gases methane (CH₄), ethane (C₂H₄) and acetylene (C₂H₂), corresponding to the increasing energy levels of gas formation. About 1000 sample test reports were collected from different utilities in India. Out of those reports, with abnormal gas formations and interpreted faults (PC57.104/D3.3, 2008; IEC 60599 {Ed.2.1}, 1999) are separated with suggested remedial actions. The dissolved gas analysis by Duval triangle involves percentage of gas (CH₄, C₂H₄ and C₂H₂) ratios in graphical presentation:

(1)

(2)

(3)

Fig. 3	: Duval triangle

Key faults presented in Duval triangle (Table 2).

How to use Duval triangle?: There are two different procedures to use this Novel method:

By the use of this method both the procedures indicate the same fault.

A procedure to use the Duval triangle: Graphical use of Duval triangle is very simple. Consider the three side of triangle in triangular coordinates (x, y and z) representing the relative proportion of CH₄, C₂H₄ and C₂H₂, from 0-100% for each gas.

To find the faults graphically (manual), first calculate the percentage of each gas as per Eq. 1-3. Then the draw the lines CH₄% quantity parallel to C₂H₂ line, C₂H₄% quantity parallel to CH₄ line and C₂H₂% quantity parallel to CH₄. Thus, drawn intersection of all three lines would indicate the fault for the GA results in the transformer.

Table 2:	Key faults in Duval triangle

Table 3 :	Triangular coordinates for Duval triangle zones.

For example, Fig. 4 indicates a fault and such verification of faults by Duval triangle (manual) DGA has been done (for >100 fault reported transformers).

These results were verified with DGA interpretation for total dissolved combustible gases by other procedures used by different utilities in India. Example: CH₄ = 56 ppm, C₂H₄ = 55 ppm and C₂H₂ = 43 ppm, manually calculated result D2 shown in Fig. 4.

Software implementation of Duval triangle DGA carried out for the same samples on MATLAB 7.4 and cross verified. Used software can be obtained on request. This software is developed with the knowledge of computer graphics and by fixing the points for fault zones coordinates (polygons) (Table 3). To define each polygon, the defined points are converted to Cartesian coordinates (Duval, 2002) for percentage of gases for type of fault.

Fig. 4:	Graphical analysis on Duval triangle responsible

Fig. 5:	Software analysis on MATLAB

Same example analysed by software on MATLAB-7.4 and provides the same result D2 in Fig. 5.

RESULTS AND DISCUSSION

In this study, manual and software implementations of Duval triangle for DGA provides the following results:

Fig. 6:	Comparative fault analysis for power transformers (India)

A comparative chart from DGA fault reports is given in Fig. 6.

CONCLUSION

The results show that Duval triangle interpretation is a robust technique and does not require much expertise.

Software implementation for Duval triangle can be done on the computer with many high level languages. Also, it is found from existing fault diagnostics tools for Indian conditions, the maximum fault occurring is T3 (44%).