Electrical discharges in ferrofluids based on mineral oil and novel gas-to-liquid oil

2021 
Abstract Ferrofluids consisting of stabilized iron oxide nanoparticles and insulating oils have emerged as a promising substitute for liquid dielectrics in electrical engineering applications. Recent enhancements of electrical insulating liquids rely on preparation of ferrofluids on emerging insulating oils available on the market. The present work reports on a comparative experimental study of insulating properties of a conventional mineral oil (MO)-based ferrofluid and a ferrofluid based on novel insulating oil produced by a gas-to-liquid (GTL) technology. The ferrofluids prepared on the two oils are subjected to rigorous experimental investigation of dielectric breakdown and partial discharges. The experiments are conducted on 4 MO-and 4 GTL-based ferrofluids with equal concentrations of magnetite nanoparticles. Measurements of partial discharges according to IEC 60270 are complemented with high frame–rate photography with positive streamer analysis. Based on the statistical analysis, it is found that MO-based ferrofluids exhibit superior breakdown performance to GTL-based ferrofluids, even though the pure GTL oil exhibits slightly higher mean breakdown voltage than the pure MO. A deeper analysis revealed a significantly greater number of partial discharges in GTL ferrofluids. The positive streamer size is lowered with increasing nanoparticle concentration only in MO-based nanofluids. Differences in the physical properties of the two oils, such as density, viscosity and permittivity, are considered in the interpretation of the different dielectric performance of the two ferrofluids. The experimental comparison leads to better understanding of the breakdown mechanism, and lay the foundations for proper selection of physical properties of a base oil for high performance insulating ferrofluids.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    32
    References
    4
    Citations
    NaN
    KQI
    []