Knut Brede Liland

Oxidation of paper insulation in transformers

Oxidation ageing experiments on paper insulation and transformer oil have been performed using two different techniques. One is a traditional ageing study wherein degree of polymerization (DP) is measured after conditioning in sealed bottles containing paper submerged in oil, where the partial pressure of oxygen over the oil is varied. Paper is sampled at certain intervals during 8 months and DP-values are measured. The other study was carried out by means of a microcalorimeter; continuously measuring the heat flow from the paper-oil sample. We find that the degradation reaction rate is not linear with oxygen concentration and that the activation energy of oxidation is lower than what has earlier been found for hydrolysis. The microcalorimeter appears to mimic ageing characteristics for oxidation and can potentially be a useful tool for quick ageing investigations.

Study of oxidation and hydrolysis of oil impregnated paper insulation for transformers using a microcalorimeter

An isothermal microcalorimeter has been used to study ageing of oil impregnated kraft paper. Under the assumption that the heat flow is proportional to the ageing rate of paper it is found that activation energy seems to be lower for oxidation than for hydrolysis. This observation corresponds well with results obtained by traditional ageing methods. The results have also been confirmed by measuring the changes in DP of the samples used in the calorimetric measurements. Comparing degradation of aged and unaged oil impregnated paper in air gave a higher heat flow for the unaged paper, but the activation energy for the processes remained the same. The process was also studied with another cellulosic material and in addition varying size of the sample holder of the calorimeter, giving similar results. Hence, since the microcalorimeter appears to give the same qualitative ageing characteristics for oxidation and hydrolysis as more time consuming methods, calorimeter could be a useful tool for quick ageing investigations.

Esterification of low molecular weight acids in cellulose

The interaction between low molecular weight acids dissolved in mineral oil with cellulose has been studied. The formation of esters as a function of temperature has been demonstrated using attenuated total reflection infrared (IR) spectroscopy. Higher temperatures lead to a shift towards ester formation compared to lower temperatures. This reaction might be reversed by exposing the cellulose to an excess of water at elevated temperatures. However, this hydrolysis of the esters will not be complete, leaving some of the initially free acids trapped in the paper as esters. Regarding degradation of the cellulose, only free acids with a carbonyl group available will enhance the break-up of the glycosidic bonds in the cellulose polymer. Therefore it seems more appropriate to use cold water extraction in order to quantify the amount of free acids in the paper/pressboard.

Oxidation of Cellulose

Oxidation ageing experiments on paper and transformer oil insulation has been performed using two different techniques. First, we used oxygen pressurized bottles with oil-paper isolation where oil and paper are sampled at certain intervals during 8 months and degree of polymerization is measured. Second, a microcalorimeter measuring continuously the heat flow from the paper-oil sample was used. We find that the degradation reaction rate is not linear with oxygen concentration and that the activation energy of oxidation is different from what has earlier been found for hydrolysis. The microcalorimeter seems to mimic ageing characteristics for oxidation and can potentially be a useful tool for quick ageing investigations.

Ageing of oil impregnated thermally upgraded papers

Investigation of the difference in the ageing process for several thermally upgraded and non-upgraded papers in transformer oil was performed. We observed that the level of nitrogen in the papers is crucial for the resistance towards hydrolysis. One of the papers in the experiment did not fulfil the upgrade specification with respect to the nitrogen level (data sheet). For hydrolysis of wet oil impregnated samples in argon atmosphere the upgraded papers seem to degrade less than non-upgraded papers. The paper with highest level of nitrogen degrades less than those with lower levels. The Insuldur process seems to be the best way of thermally upgrading the paper (Upgrade 1) and this also gives the highest level of nitrogen. For one of the upgraded samples (Upgrade 2) the nitrogen disappears completely after startup and this paper behaves as non-upgraded for the highest temperature. Dry oil impregnated paper (0.2% water content) under argon atmosphere does not seem to be hydrolysed and ages at a very slow rate. In the case of oxidation of dry oil impregnated samples (0.2% water content) in air there are initially no significant difference between non-upgraded and upgraded papers. However, the upgraded papers seem to have an improved performance after the water production from ageing becomes significant. The oxidation activation energy for non-upgraded paper is lower compared to hydrolysis and for upgraded paper this difference is smaller.

Liquid insulation of IGBT modules: Long term chemical compatibility and high voltage endurance testing

To facilitate operation of power electronics for subsea operation at ambient pressure components have to be submerged in a liquid. Equipment and schemes for testing of long term properties have been developed. Several techniques were used to investigate compatibility between a silicone gel and various insulation fluids. Equipment for electric endurance testing of power electronic components at dc stress under controlled temperature and humidity was developed. Performance of high voltage diodes covered with various insulation liquids, coatings and gel covering were studied.

Absorption of water in silicon gel

Water is an important enemy of an electric insulation system. Water vapour can be absorbed by electrical insulation materials (liquids or solid) and migrate in between the materials to achieve the same relative humidity equilibrium everywhere. Water can reduce withstand voltage of the insulation liquid, initiate partial discharges in wet solids and increase dielectric losses in solids like epoxies, gels, PCB cards and produce dielectric heating. In insulation liquid filled system the water content of the liquid should never be allowed to reach saturation. One failure scenario is when typical oil with significant water content is cooled to a temperature where the water content is above the saturation content, and then water will precipitate and may condensate at cold insulating gaps and give breakdown due to electric field enhancement. Gel samples were investigated in climate chambers in various conditions of humidity and temperature until an equilibrium was reached. Diffusion and absorption of water were measured at different intervals. The aim of the study was to find an appropriate technique/methodology to follow the absorption of moisture in gel as a function of time. Several techniques were tested to determine the moisture content in the gel such as Karl Fischer, weight differences measurement, freeze drying and capacitive relative humidity sensors giving contrasted results. The weight difference method showed “anomaly” indicating bound water, even after equilibrium (measured with the other methods) was reached the weight continued to increase. The use of humidity sensors covered by gel appears to be the most reliable technique.

Determine the compatibility between electrical insulation fluids and solid materials using a micro-calorimeter

Compatibility between solid and liquid electrical insulation materials has been studied using an isothermal microcalorimeter. This work compares the micro-calorimeter measurements with a simple volume and mass increase for the solid material also capable of indicating incompatibility and compatibility in some cases. The measurements reported here on a known incompatible system (XLPE and mineral oil) and a compatible system (XLPE and silicone oil) illustrate how such measurements could be performed. For these materials we conclude that a micro-calorimeter can be used to characterise compatible/incompatible phenomena for electrical material combinations. It proves to be a fast and alternative method potentially capable of screening compatibility between many different material combinations giving an indication/warning of undesirable reactions. Such a methodology must in general be supported by results from other kind of analytical measurements in addition. It could however be used as a first screening method targeting potential promising new material combinations for electrical apparatuses. The most promising material combinations found from such measurements then could later be investigated further with other methods.

Axial water ingress in watertight MV XLPE cable designs

When the outer sheath of a polymeric cable becomes severely damaged, humidity can enter the insulation system potentially increasing the relative humidity above a critical value (70% RH) making initiation and growth of water trees possible. The main purpose of this work is to determine how fast water vapour will diffuse axially in the cable. Sensitive relative humidity and temperature sensors were placed within the outer sheath at different axial positions. After drying (evacuation) a hole was cut at the cable end facilitating water ingress. Numerical calculations of axial water diffusion were performed using Comsol. The results so far show that the axial water vapour diffusion in the cable is slow and dependent of the air gap close to the swelling tape. After 130 days the humidity at 0.5 m had increased by about 40%, and the sensor at 1 m had increased by 20%. Numerical calculations of the water diffusion in the same section show a slower increase. The actual axial liquid water penetration is yet not determined. The numerical calculations show that this is an important factor, as the calculations are in more agreement with measurements when adjusting the position of the water front.