The dissolved gas analysis of a transformer also called transformer gas analysis is an essential process or technique through which the condition of the transformer and its life expectancy can be calculated. Periodic dissolved gas analysis or DGA test can forecast the faults present in the oil-insulated electrical equipment well ahead of its failure, by studying various gases and their proportions present in the test sample.
Table of Contents
The generation of gases in the oil-filled equipment is because of the disruptive arcing and severe overheating which induces chemical reactions as a result of faults. The high thermal and disruptive discharge because of faults, causes the breakdown of solid and liquid insulation. However, the gaseous product is of more interest and concern because these gaseous products can give out whether they are the result of air contamination, oxidation, partial discharge, or severe thermal conditions resulting from heavy electrical faults which can cause damage to the insulation and the equipment.
Methods of Dissolved Gas Analysis
The dissolved gas analysis is performed by processes such as gas chromatography and infrared spectroscopy after the sample is collected from the transformer.
How to collect the sample?
The sample is collected in a syringe by opening the oil inspection valve or a drain valve of the transformer. It may be noted that after opening the drain valve, the surrounding of the valve must be purged out of moisture and dirt otherwise it may contaminate the sample. We have to attach a plastic hose which will be the intermediate connection between the valve and the syringe. By the pulling action of the syringe, the sample is drawn from the valve up to 50cc or as per requirement. After the withdrawal of the sample, the air has to be evacuated from the syringe. Finally, closure of the valve has to be ensured after the process.
Gas chromatography:
In this process of dissolved gas analysis, the gas is detected with the use of a flame ionization detector(FID) and thermal conductivity detector (TCD).
The oil is passed through a very high vacuum where most of the gasses are extracted from the oil. The gas is then collected and measured in a graduated tube with a mercury piston. The measured gas is then passed or injected into the gas chromatograph where it is taken by an inert gas toward the column where a different component of the gas interacts with a stationary phase leading to separation based on volatility and affinity.
Various separated gases exit the column passing through the detectors where the amount of gas is detected.
Theory of Gas
The formation of the gas in the insulating oil depends on many factors other than the fault. Also, the test sample is influenced by factors not directly related to the fault. The factors may be:
- Previous history of the transformer
- Loading history
- Insulation condition
- Location of fault.
However, it is possible to access the evolution of gas based on the thermal behavior of the fault, and from the previous knowledge base and careful assessment of all the relevant factors, the cause and seriousness of the fault can be ascertained.
The enthalpy of the fault results in the immediate breakdown of the hydrocarbon molecules of the insulating oil into free radicals as cited in the figure below. These radicals subsequently recombine to form low molecular weight hydrocarbon gases. This recombination is determined by the temperature.
For lowest temperature faults, hydrogen and methane are generated, methane being the dominant. As the temperature increases with the intensity of fault increasing, ethane starts evolving, reducing the methane so that the ethane to methane ratio becomes predominant. At higher temperatures, ethane reduces and production of ethylene starts and soon outruns the proportion of ethane. And finally, at the most elevated temperatures, acetylene evolves and becomes the gas which is most predominant.
In the above figure, the operating temperature extends up to 150 degrees Celsius, 65 degrees being the normal operating range, 200-300 degrees Celsius for hotspots, and 1000 degrees for high temperature and thermal faults. The peak ethylene production occurs at around 700 degrees Celsius.
Interpretation of the Dissolved Gas Analysis
The main ratios are Methane/Hydrogen, Ethane/Methane, Ethylene/Ethane, Acetylene/Ethylene. For each ratio having a value less than one is coded as 0 and if greater than one, is coded as 1. After these ratio codes, the compilation of these data in a table is done along with the temperature scale and associated with certain faults. These ratios with some modifications have become a popular mark of dissolved gas analysis and came to be known as the Rogers ratio.
Diagnostic interpretation of gas ratio by CEGB in 1972
| CH₄/H2 | C₂H₆/ CH₄ | C₂H₄/C₂H₆ | C₂H₂/C₂H₄ | Diagnosis |
| 0 | 0 | 0 | 0 | If CH₄/H₂ = or < 0.1 → Partial discharge; otherwise normal deterioration |
| 1 | 0 | 0 | 0 | Slight overheating below 150°C |
| 1 | 1 | 0 | 0 | Slight overheating 150-200°C |
| 0 | 1 | 0 | 0 | Slight overheating 200-300°C |
| 0 | 0 | 1 | 0 | General conductor overheating |
| 1 | 0 | 1 | 0 | Circulating currents and/or overheated joints |
| 0 | 0 | 0 | 1 | Flashover without power follow-through |
| 0 | 1 | 0 | 1 | The arc with power follow-through or persistent sparking |
| 0 | 0 | 1 | 1 | Arc with power follow-through or persistent sparking |
Codes of RR Rogers for dissolved gas analysis in 1974
| Gas Ratio | Range Description | Code |
| CH₄/H₂ | Not greater than 0.1 | 5 |
| Between 0.1 and 1.0 | 0 | |
| Between 1.0 and 3.0 | 1 | |
| Not less than 3.0 | 2 | |
| C₂H₆/CH₄ | Less than 1.0 | 0 |
| Not less than 1.0 | 1 | |
| C₂H₄/C₂H₆ | Less than 1.0 | 0 |
| Between 1.0 and 3.0 | 1 | |
| Not less than 3.0 | 2 | |
| C₂H₂/C₂H₄ | Less than 0.5 | 0 |
| Between 0.5 and 3.0 | 1 | |
| Not less than 3.0 | 2 |
Diagnostic chart of Roger’s Dissolved gas analysis
| CH₄/ H₂ | C₂H₆/ CH₄ | C₂H₄/C₂H₆ | C₂H₂/C₂H₄ | Diagnosis |
| 0 | 0 | 0 | 0 | Normal deterioration |
| 5 | 0 | 0 | 0 | Partial discharge |
| 1/2 | 0 | 0 | 0 | Slight overheating below 150°C |
| 1/2 | 1 | 0 | 0 | Overheating 150-200°C |
| 0 | 1 | 0 | 0 | Overheating 200-300°C |
| 0 | 0 | 1 | 0 | General conductor overheating |
| 1 | 0 | 1 | 0 | Winding circulating current, overheated joints |
| 1 | 0 | 2 | 0 | Core and tank circulating currents |
| 0 | 0 | 0 | 1 | Flashover without power follow-through |
| 0 | 0 | 1/2 | 1/2 | Arc with power follow-through |
| 0 | 0 | 2 | 2 | The arc with power follow-through |
| 5 | 0 | 0 | 1/2 | Partial discharge with tracking (note CO) |
Other authorities use varied ratios for dissolved gas analysis as the British standard 5800 which resembles IEC-599 uses only three of the ratios discussed above omitting the ethane to methane ratio.
Drawing conclusions from carbon monoxide and carbon dioxide in dissolved gas analysis can be very misleading as it was considered earlier their source was from overheating cellulose or paper insulation but it is now recognized that carbon dioxide and carbon monoxide can evolve from normal oxidation of the insulating oil too and in a considerable amount. Therefore, it is more reliable in dissolved gas analysis to ignore these gases.
| Code of the range of ratios | |||||
| C₂H₂/C₂H₄ | CH₄/ H₂ | C₂H₄/C₂H₆ | |||
| Ratios of characteristic gases | |||||
| < 0.1 | 0 | 1 | 0 | ||
| 0.1-1 | 1 | 0 | 0 | ||
| 1-3 | 1 | 2 | 1 | ||
| > 3 | 2 | 2 | 2 | ||
| Case No. | Characteristic Fault | Typical Examples | |||
| 0 | No fault | 0 | 0 | 0 | Normal aging |
| 1 | Partial discharges of low energy density | 0, but not significant | 1 | 0 | Discharges in gas-filled cavities |
| 2 | Partial discharges of high energy density | 1 | 1 | 0 | As above, but leading to tracking or perforation of solid insulation |
| 3 | Discharges of low energy (see Note 1) | 1 →2 | 0 | 1 →2 | Continuous sparking in oil between bad connections of different potential or to floating potential; breakdown of oil between solid materials |
| 4 | The thermal fault of medium temperature range 300°C – 700°C | 1 | 0 | 2 | Discharges with power follow-through; arcing breakdown of oil between windings or coils, or between coils to earth; selector breaking current |
| 5 | Thermal fault of low temperature < 150°C (see Note 2) | 0 | 0 | 1 | General insulated conductor overheating |
| 6 | The thermal fault of low-temperature range 150°C – 300°C | 0 | 2 | 0 | Local overheating of the core due to concentrations of flux; increasing hot spot temperatures; small hot spots in the core; shorting links in the core; overheating of copper Due to eddy currents, bad contacts/joints (pyrolitic carbon formation); up to core and tank circulating currents Severe overheating leading to potential failure |
| 7 | Thermal fault of medium temperature range 300°C – 700°C | 0 | 2 | 1 | |
| 8 | Thermal fault of high temperature > 700°C (see Note 4) | 0 | 2 | 2 | |
Druval’s Triangle
Apart from the Rogers ratio, the Dissolved Gas Analysis or DGA results can also be interpreted by the Duval triangle method. In this method, the three vertices of the triangle represent percentages of these three gasses CH4, C2H4, C2H2.
Let us take data from a dissolved gas analysis or dga report:
- H2: 16 ppm
- CH4: 1818 ppm
- C2H2: 2789 ppm
- C2H4: 715 ppm
- C2H6: 561 ppm
Now the required data for the interpretation of dissolved gas analysis with the Duval triangle are CH4 =1818, C2H4 =715, C2H2=2789.
Therefore, the percentage of CH4 = 1818/(1818+715+2789)= 34.16%
Percentage of C2H4 = 13.4%
Percentage of C2H2 = 52.4%
The lines are drawn parallel to the lines on the triangle as per the calculated percentages and the intersecting point lies in the area marked as D1.
Following are the marked areas of the diagram:
- PD- Partial discharge
- T1- Thermal fault below 300 degrees Celsius
- T2- Thermal fault between 300-700 degrees Celsius
- D1- Low energy discharge
- D2- High energy discharge or arcing
- DT- Combination of thermal and electrical faults.
Conclusion
In practice, many professionals individuals, and organizations take the dissolved gas analysis or dga of the transformer as an answer to all issues. However, this is not the case. On many occasions, it can potentially create problems than it solves.
Drawing conclusions on the basis of dissolved gas analysis of the first test sample must be avoided. The first sample is to provide an initial analysis of the fault, whether the fault persists or not. The following sample is to reconfirm the findings of the first analysis. The next step is to check the history of the transformer. If it is in service for how much time and inspect the loading characteristics? By checking the history of past sampling, one can conclude if the gas levels are rising steadily or if there is a sudden step up which should be taken seriously.
If the gas has diffused from the diverter switch of the OLTC or the transformer is being topped up with contaminated oil. After checking for all these possibilities one should proceed with the possibility of the fault as per the dissolved gas analysis.
