Surface voltage of bushings, covered conductors and breakdown in needle tipplane electrodes

Abstract

Surface voltage measurement of an insulated (silicone rubber) overhead conductor assumes importance for understanding corona, radio interference voltages, and design purposes. Surface voltage measurement of a bare overhead conductor is quite simple as a voltage divider can be directly connected. However, there are challenges in the measurement of surface voltage on an insulating surface of an insulated covered conductor using voltage dividers. In case of ac measurements, the capacitance parameter dominates in deciding the potential distribution. In this work, it is demonstrated that the measuring system itself will alter the voltage division between the capacitance of the cylindrical insulated conductor and that of surrounding air acting as a capacitive divider. The voltage distribution is shown to be drastically influenced by the measuring system capacitance. In view of these issues, a novel experimental method is proposed for the measurement of surface voltage of an overhead insulated covered conductor using cylindrical strips. In general, this method can be applied to any cylindrical insulated conductor and for the calibration of sensor-based measurements. The proposed empirical method is tested by a validated simulation of the entire system, and it is proven analytically. The analytical, experimental, and simulation results are in close conformity. However, in case of a general insulator surface, the use of a cylindrical strip may not be possible due to the reason that the voltage at different points on the surface of insulators can be different (e.g., string insulators of overhead line). Also, while measuring potential points near the earthed surface, the use of a cylindrical strip might electrically short different potentials, which is not desirable. Considering these facts, a unique method is presented, using which the surface voltage of any insulating surface such as insulator strings or transformer bushings can be measured. The work is based on analytical derivations, a more generalized method of measurement, applicable to any point on an insulating surface of arbitrary geometry is proposed using circular-disc strips. The proposed experimental method is applied to insulator strings and transformer bushings and validated by simulation of the entire system. The simulation and experimental results are in excellent agreement. Generally, for power equipment insulation, the process of damage accumulation that starts due to the presence of the defects eventually leads to power equipment failure. In power cables, conductor protrusions or metallic defects in insulation are important among these defects. The electric field at the tip of sharp protrusions increases multifold due to the small radius of curvature at the tip. The high electric field at the tip initiates electrical treeing under sinusoidal voltages and leads eventually to insulation breakdown. These defects can be modeled by using needle-plane geometry. Therefore, the study of needleplane geometry assumes importance for the evaluation of electric stresses at the tip of these defects. Here, electric field and space charge accumulation have been investigated for needle-plane system (defect) in silicone rubber insulation under dc voltage conditions. Prolate spheroidal coordinate system, believed to be closer to needle-plane system, has been used for solving the governing differential equations, numerically, for space charge and electric field distributions. Unlike past works, in which, space charge at needle tip was assumed, either qualitatively or quantitatively, in this work, space charge formation is estimated using nonlinearity of material properties alone. A comparison with previous methods based on concentric spherical electrode approximations reveals that the results are different for prolate spheroidal system at different nonlinearities. Interesting results on the role of nonlinear conductivity on the dc electric field and space charge accumulation at needle-tip are presented. Also, needle tip-plane breakdown experiments are conducted. Further, for DC systems with line-commutated converters (LCC), where to reverse the power flow, it is necessary to change the voltage polarity. It is known that high stresses may occur at the conductor immediately after reversing the polarity of an external voltage source and if not properly designed, stresses the cable leading to insulation failure. In view of this, a 2D axisymmetric numerical model is presented to compute the timedependent field, space charge, and current density distribution in needle plane geometry under polarity reversal DC. Interesting results on the effect of polarity transition time, applied polarity time step on the above-mentioned distributions have been addressed and presented. The pronounced effect of nonlinear conductivity on the time-dependent field distribution is also demonstrated. In addition, breakdown experiments are conducted on tip-plane electrode system for different tip radii and for different applied time step polarity reversal, and the results put forward a rational and practical estimate of breakdown field.