| There is a general agreement that in service conditions the quality of mineral insulating oils gradually deteriorates under the impact of electrical, thermal and environmental stresses. It is also widely accepted that only the incipient electrical failures such as hot spots and partial discharges are responsible for the gassing of oil. Knowing that the resulting fault gases dissolve in the oil, the technique of Dissolved Gas Analysis (DGA) was developed to detect incipient failures in the transformer. DGA has now become a standard in the utility industry throughout the world and is considered to be the most important oil test for insulating liquids in electrical apparatus. More importantly, an oil sample can be taken at anytime from most equipment without having to take it out of service, allowing a "window" inside the electrical apparatus that helps with diagnosing and trouble-shooting potential problems. This thesis intends to show that the gassing of oil is a more complex phenomenon.;In order to emphasize the role played by contaminants in the gassing of oil, fundamental investigations were undertaken. The amount of gases evolved under the impact of electrical stress (ASTM D6180) by a sample of new and aged oil with/without paper was accurately measured along with some physicochemical properties, to assess the relationship between the cause and the symptoms of oil or oil-paper insulation deterioration. The outcome of these investigations provided experimental evidence that the chemical composition of hydrocarbon blend, the oil born decay products and the solid insulation are also contributing factors to oil gassing. Since this finding may affect the diagnostics predicted by some DGA techniques, some thorough investigations were performed. New, aged oil and reclaimed aged oil samples were submitted to thermal and electrical stresses (considering various scenarios) and the dissolved gases analyzed by chromatography. Three of the most used DGA techniques, namely the Duval's Triangle Roger's and Dörnenburg's ratios were implemented in Labview based software to predict the diagnostic. The obtained results provide experimental evidence that oil born decay products may affect the diagnostics predicted by some DGA techniques. Although such a research is still in a preliminary stage, some very stimulating results have been obtained. |
