Kerr Electro-optic Measurements Research Articles

HVDC electrical insulation performance in oil/pressboard composite insulation system based on Kerr electro-optic field measurement and electric field analysis

For the development of HVDC electrical insulation performance in liquid dielectric systems such as oil-pressboard (PB) composite systems in DC converter transformers, a Kerr electro-optic electric field measurement (Kerr-EOM) technique has been well established and will be considered as the key technique to be introduced. In this paper, the Kerr-EOM was experimentally applied to identify the electric field stress distribution and clarify charging activities in oil/PB composite insulation systems, under HVDC and DC polarity reversal (DC-PR) conditions. The physical mechanisms of charge movement and the amount of charge accumulation in liquid/solid insulation systems are also thoroughly investigated and discussed. Moreover, the breakdown discharge initiation by accumulated charges on the PB was observed and discussed together with the field measurement results. Secondly, by combining the Kerr-EOM with the electric field analysis using the Finite Element Method (FEM), the enhancement of the electrical insulation performance for HVDC applications was discussed and the critical testing voltage conditions in DC converter transformers were addressed. This paper is based on the Hans Tropper Memorial Lecture given at the IEEE DEIS, International Conference on Dielectric Liquids (ICDL) in 2017 in Manchester, UK.

Effect of nanoparticles on impulse breakdown performance of propylene carbonate based on modified Kerr electro-optic measurements

Impulse breakdown test results show the breakdown strengths of propylene carbonate modified by Al2O3 nanoparticles and Fe3O4 nanoparticles are improved by 13 % and 20 % than that of propylene carbonate without nano-modification, respectively. In order to explore the nano-modification mechanism, the electric field and space charge distributions in propylene carbonate with and without nano-modification were measured through a modified Kerr electro-optic measurement method. The measurement results show the space charge densities are decreased and the space charge transportations are inhibited in PC-based nanofluids compared with that of pure propylene carbonate due to the space charge adsorption of nanoparticle. Therefore, the streamer propagation in nanofluids is impeded, and the breakdown voltage is improved. The calculation of electric potential outside the nanoparticle indicates that Fe3O4 nanoparticle has stronger space charge adsorption capacity than Al2O3 nanoparticle. Accordingly, the Fe3O4-modified propylene carbonate shows greater improvement of breakdown performance than Al2O3-modified propylene carbonate.

Time-continuous Kerr electro-optic field mapping measurement under impulse voltage using array photodetector

Space charge affects the electric field distribution in high voltage-stressed liquid dielectrics and can cause insulation degradation. This letter reports on research using a Kerr electro-optic measurement system with an array photodetector for Kerr electro-optic field mapping measurements in propylene carbonate under application of switching impulse voltages to investigate the electric field and space charge effects in high voltage-stressed liquid dielectrics. The electric field distribution between parallel aluminum electrodes is measured, and the results are compared with those measured using a charge-coupled device measurement system. The array photodetector measurement system has the advantages of higher sensitivity and higher time resolution because simultaneous optical Kerr measurements can be taken at multiple positions using a multi-element array photodetector. The results show that bipolar homo-charge injection occurs between the parallel aluminum electrodes. In the near future, this measurement system will be used in continuing research into the measurement of space charge effects using low Kerr constant liquids such as transformer oil.

Electro-optic signatures of turbulent electroconvection in dielectric liquids

Signatures of turbulent electroconvection in transformer oil stressed by dc and ac voltages are identified from Kerr electro-optic measurement data. When the applied dc voltage is high enough, compared with the results with no applied voltage, the optical scintillation index and image entropy exhibit substantial enhancement and reduction, respectively, which are interpreted as temporal and spatial signatures of turbulence. Under low-frequency ac high voltages, spectral and correlational analyses indicate that there exist interacting flow and charge processes in the gap. This work also clarifies the meaning of dc steady state and the requirement on ac modulation frequency in Kerr measurements.

Fractal-like charge injection kinetics in transformer oil stressed by high-voltage pulses

Kerr electro-optic measurements are taken to study the transient electrode charge injection in high voltage pulsed transformer oil. It is found that the injection current densities from two stainless-steel parallel-plate electrodes with distinct surface roughness display fractal-like kinetics, i.e., power-law time dependence. Our measurement data agree with numerical simulation results of the time-dependent drift-diffusion model with the experimentally determined injection current boundary conditions. The fractal-like kinetics implies that the electric double layer processes contributing to injection are diffusion-limited. Physical mechanisms based on formative steps of adsorption-reaction-desorption reveal possible connections between geometrical characteristics of electrode surfaces and fractal-like kinetics of charge injection.

Electro-optic precursors of critical transitions in dielectric liquids

Motivated by the search for approaches to non-destructive breakdown test and impurity detection in dielectric liquids, we explore the possibility of early warning of breakdown initiation in high voltage pulsed transformer oil from the data of Kerr electro-optic measurements. It is found that the light intensities near the rough surfaces of electrodes both fluctuate in repeated measurements and vary spatially in a single measurement. We show that the major cause is electrostriction which brings disturbances into optical detection. The calculated spatial variation has a strong nonlinear dependence on the applied voltage, which generates a precursory indicator of the critical transitions.

Space charge dynamics investigation based on Kerr electro-optic measurements and processing of CCD images

The Kerr electro-optic technique is used to investigate electric field and space charge effects using large Kerr constant propylene carbonate as the dielectric liquid with Kerr constant B≈1.22×10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-12</sup> m/V <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> . However, improved accuracy of this technique to measure space charge distributions remains an important goal. This paper describes inversion algorithm research on obtaining accurate space charge distributions from Kerr measurements. The original charge-coupled device (CCD) image of light intensity was converted into grayscale values for improved calculation. The grayscale CCD image was transformed into the frequency domain through a two-dimensional Fourier transformation. Butterworth filtering techniques were employed for edge enhancement algorithms. The most important parameter for calculation of space charge distributions are the integer fringe numbers n in the sequence of light and dark fringes measured by the CCD camera. Comparison of the improved algorithm and traditional methods for calculating electric field and space charge distributions between a pair of stainless steel parallel plate electrodes was investigated. The space charge dynamics were also calculated after processing the CCD images with some improvements. The results showed the electric field was essentially uniform at early times of pulsed high voltages. This uniformity indicated the absence of space charge during the early transient interval, while a bipolar charge injection was found in the central region between electrodes at later times. For longer times the electric field and space charge distributions slowly became uniform again. Grayscale processing and Butterworth filtering played an important role in the improvement of the algorithm for better image quality and edge enhancement.

Investigation of charge behavior in low viscosity silicone liquid by kerr electro-optic field measurement

From the viewpoints of less flammability and environment protection, a new type of electrical insulating liquid for power transformers is strongly required. Low viscosity silicone liquid (20 cSt) is suitable for the requirements owing to the following excellent features; high flash point, low environmental impact concerning pollution and high chemical stability. In this paper, we investigated the charge behavior under dc voltage application by direct measurement of the electric field using Kerr electro-optic method. At first, Kerr constant of 20 cSt silicone liquid was investigated. Then, the time dependence of electric field strength in silicone liquid / pressboard (PB) composite systems was measured. The time constant of the decay was much larger in silicone liquid than in mineral oil. We discussed the charge behavior based on the physical properties of silicone liquid and we suggested that the influence of negative charge injection from a negative electrode¿s surface in silicone liquid played a significant role because of the high volume resistivity.

Fundamental property of electric field in rapeseed ester oil based on Kerr electro-optic measurement

In the future, environmentally friendly insulating oil such as rapeseed ester oil is expected to be utilized as a substitute for mineral oil for power transformers. In addition, electrical insulation design for transformers becomes more important regarding size reduction and enhancement of electric field stress. The fundamental property of the electric field distribution in the rapeseed ester oil and in the oil/solid dielectrics composite insulation system should be clarified. In this paper, we have directly measured the electric field in the rapeseed ester oil with parallel-plane electrode and the rapeseed ester oil/pressboard (PB) composite insulation systems under DC voltage application using the Kerr electro-optic technique. From the results, we compared measurements of electric field and charge behavior for the rapeseed ester oil and the mineral oil and confirmed the applicability of the rapeseed ester oil for use in transformers

DC field measurement in oil / pressboard composite insulation system by electro-optic Kerr effect

An insulating oil/pressboard (PB) composite insulation is generally utilized in oil-immersed power transformers. In such a transformer, charges accumulate at oil/PB interface which changes the electric field distribution depending on time and oil condition. In this paper, we report on directly measured dc electric field in oil/PB composite insulation system using the Kerr electro-optic field measurement technique. From measurement results, we propose a physical model for the charge behavior. We have calculated an electric field distribution using finite element method (FEM) based on the charge movement. We compared the results of measurements and calculations and confirmed that the proposed charge behavior model explains well the experimental results in the case of the polarity reversals as well. As a result, a significant validity of the charge behavior model was confirmed.

Kerr electro-optic measurements of space charge effects in HV pulsed propylene carbonate

The Kerr electro-optic field mapping technique has been used to study space charge effects in HV pulsed propylene carbonate (C/sub 4/H/sub 6/O/sub 3/) using stainless steel parallel-plane, aluminum blade-plane, and aluminum point-plane electrodes. Measurements presented here were taken using a circular polariscope with aligned polarizers at room temperature (T /spl sim/ 20 to 27/spl deg/C). Average peak space charge free electric field strengths were E/sub 0/ /spl sim/ 52 kV/cm for the parallel-plane electrodes, E/sub 0/ /spl sim/ 210 kV/cm at the blade electrode tip and E/sub 0/ /spl sim/ 240 kV/cm at the point electrode tip. The measurement times ranged from 10 /spl mu/s to 1 ms. For the parallel-plane electrodes, the light intensity pictures were analyzed along a line between electrodes in the center of the electrode gap and for the blade-plane and point-plane electrodes the light intensity pictures were analyzed both from the tip of the blade and point, to the plane, and along the plane surface. From the light intensity distribution for the parallel-plane and blade-plane electrodes, the electric field and space charge density were calculated as a function of position and time. The calculated space charge distributions show both positive and negative charge injections from the electrodes as well as bulk charge dissociation and recombination. The light intensity was recorded on Polaroid film for the parallel-plane, blade-plane and point-plane electrode measurements and also with a CCD camera for the point-plane electrodes. A CCD camera was used for the point-plane electrodes to resolve more accurately the gray-scale light intensity distribution because of the short optical path length. All the HV Kerr electro-optic measurements showed significant space charge effects in propylene carbonate because the light intensity distributions differed significantly from the calculated Kerr effect patterns from numerical solutions to Laplace's equation under space-charge free conditions.

Optical measurement of non-uniform electric field vector distribution in a dielectric liquid using triplet measurement system

We have carried out measurements of three-dimensional nonuniform electric field distribution using Kerr electrooptic measurement with a reconstruction technique. A conventional measurement system was composed of He-Ne laser, a polarizer, a quarter-wave plate, an experimental cell, an analyzer, and a photo-detector. For the measurement of an electric field vector distribution, it was necessary to measure light intensity with changing the optical axis of optical wedges, i.e., the polarizer, the quarter-wave plate and the analyzer. Moreover, the intensity measurement must be carried out at many points from /spl theta/ = 0 to /spl pi/ around the electrode system. Therefore it took considerable time to measure the electric field distribution. We have improved the measurement system by a cell with six optical windows and three optical systems. Using the improved measuring system, it was possible to measure the non-uniform electric field vector distribution faster.

Finite element based Kerr electro-optic reconstruction of space charge

Recently we used the onion peeling method to reconstruct the axisymmetric electric field distribution of point/plane electrodes from Kerr electro-optic measurements. The method accurately reconstructed the electric field from numerically generated data. However in the presence of experimental noise the performance was less satisfactory. The measurements were especially noisy and unstable near the needle tip which is also the interesting region since most charge injection initiates here. We develop a new algorithm for Kerr electro-optic reconstruction of space charge in axisymmetric point/plane electrode geometries. The algorithm is built on the finite element method (FEM) for Poisson's equation and will be called finite element based Kerr electro-optic reconstruction (FEBKER) hereafter. FEBKER calculates the space charge density directly to avoid the numerical problems associated with taking the divergence of the electric field, uses single parameter light intensity measurements to enable transient analysis, which otherwise is difficult since multiple parameter intensity measurements are slow due to the rotation of polarizers, and is capable of reconstruction even when the number and/or position of measurements are limited by the electrodes and/or the experimental setup. The performance of the algorithm is tested on synthetic Kerr electro-optic data obtained for an axisymmetric point/plane electrode geometry in transformer oil with specified space charge density distributions. The impact of experimental error is analyzed by incorporating random error to the synthetic data. Regularization techniques that decrease the impact of experimental error are applied. In principle FEBKER is applicable to arbitrary three-dimensional geometries as well.

Comparative study of theoretical Kerr electro-optic fringe patterns in 2-D and axisymmetric electrode geometries

Kerr electro-optic fringe patterns have long been used to study space charge injection and transport phenomena in highly birefringent materials such as nitrobenzene, Most past experimental work has been limited to 1 or 2-dimensional geometries where the electric field magnitude and direction have been constant along the light path such as two concentric or parallel cylinders or parallel plate electrodes. For these geometries the extrema in the fringe patterns can be used directly to find the electric field magnitude and direction. In this work we extend the fringe based Kerr electro-optic measurement technique to a point/plane electrode geometry which often is used in HV research to create large electric fields for charge injection at known location and at reasonable voltages. We calculate theoretical Kerr electro-optic fringe patterns for this point/plane electrode geometry with and without space charge distributions, for which the electric field magnitude and direction vary along the light path. We particularly compare the calculated space charge free optical patterns for the point/plane electrodes to the optical patterns of the 2-dimensional analog blade/plane geometry. We underline the differences and study how these fringe patterns can be used to reconstruct the axisymmetric electric field components in practice.

Kerr electro-optic theory and measurements of electric fields with magnitude and direction varying along the light path

Differential equations that govern light propagation in Kerr media are derived when the applied electric field direction and magnitude vary along the light path. Case studies predict Kerr electro-optic fringe patterns for the specific case of point/plane electrodes. We apply the characteristic directions theory of photoelasticity to understand these fringes. We also study birefringent media with small Kerr constant, in particular transformer oil. For this case we show that simplifications in the characteristic parameter theory is possible, resulting in simple integral relationships between the characteristic parameters and the applied electric field. We use these simple relationships to extend the ac modulation method to measure the characteristic parameters for small Kerr constant media. Measurements of the characteristic parameters using the ac modulation method are presented for point/plane electrodes in transformer oil. The measurements agree reasonably well with space charge free theory for infinite extent electrodes for which analytical expressions are available. We finally employ the 'onion peeling' method to reconstruct the axisymmetric electric field magnitude and direction from the measured characteristic parameters and compare the results to the analytically obtained electric field.

Optical, electrical and electromechanical measurement methodologies of field, charge and polarization in dielectrics

A selective review is presented of the use of optical, electrical, and electromechanical measurement methodologies of the electric field, charge, conductivity, and permittivity in dielectrics. Kerr electro-optic measurements are presented to demonstrate how volume charge distributions can distort the electric field distribution significantly and how the field and charge distributions depend on dielectric and electrode materials and geometry, and on voltage magnitude, polarity, and time duration. A new class of interdigital dielectrometry sensors is reviewed which from one side can measure profiles in dielectric permittivity and conductivity and related physical properties such as moisture content. Electromechanical devices such as the absolute charge sensor, a flow loop, and the couette charger (cc) with a rotating cylindrical electrode are reviewed for their applications in liquid flow electrification measurements.

Kerr electro-optic field measurement and charge dynamics in transformer-oil/solid composite insulation systems

We measured the electric field in transformer oil/solid composite insulation system under dc voltage application using a Kerr electro-optic measurement system. When a pressboard was inserted at the center between two parallel-plane electrodes, the electric field at the mid-point between the pressboard and the cathode decreased with time. On the other hand, the electric field at the mid-point between the pressboard and the anode increased initially with time, reached a maximum value and then fell down. We measured also the electric field before and after polarity reversal of the dc voltage. The results revealed that the electric field after polarity reversal reached a magnitude 2.7/spl times/ as much as the average dc applied field. We interpreted the above results on the time transition of the electric field in terms of a charge dynamic model, considering the difference in charge accumulation speed on the pressboard between positive and negative charges. This model was verified to be applicable to the time transition of the electric field in oil/solid composite system and BTA added oil/solid composite system as well.

Comparative High Voltage Impulse Measurement.

A facility has been developed for the determination of the ratio of pulse high voltage dividers over the range from 10 kV to 300 kV using comparative techniques with Kerr electro-optic voltage measurement systems and reference resistive voltage dividers. Pulse voltage ratios of test dividers can be determined with relative expanded uncertainties of 0.4 % (coverage factor k = 2 and thus a two standard deviation estimate) or less using the complementary resistive divider/Kerr cell reference systems. This paper describes the facility and specialized procedures used at NIST for the determination of test voltage divider ratios through comparative techniques. The error sources and special considerations in the construction and use of reference voltage dividers to minimize errors are discussed, and estimates of the measurement uncertainties are presented.

Electrooptic Measurement of Three-Dimensional Nonuniform Electric Field Mapping in Nitrobenzene

An electrooptic method is useful for measuring electric fields without disturbing them. Our investigations have been concerned with measurements of electric field distribution. Three-dimensional nonuniform electric field distributions were measured in nitrobenzene using the Kerr electrooptic measurement. The field distributions to be measured were produced in a sphere-to-sphere electrode system. The optical system consisted of a He–Ne laser, a polarizer, a quarter-wave plate, an experimental cell, an analyzer, and a photodetector. When a negative high voltage pulse was applied to the electrode system, the light intensity transmitted to the analyzer was measured using the photodetector. Intensity measurements were carried out at many points from θ= 0 to π around the electrode system. From the data obtained, the electric field distribution, i.e., the strength and direction of each electric field vector in the measured plane, was reconstructed using a computed tomography (CT) technique. The measured distributions were compared with the theoretical ones calculated using an analytical method.

Localized multi-alternating-current modulation Kerr measurement of nonuniform three-dimensional electric fields

One of the electric-field measurement methods without intrusiveness uses the Kerr electro-optic effect. This method can be widely applied to gas, liquid, and solid materials, which are transparent and electrically birefringent. However, this method has not been adequately developed as yet for practical measurement of nonuniform three-dimensional electric fields. We describe a Kerr electro-optic measurement method for nonuniform three-dimensional dc and ac electric fields that employs localized multi-ac modulation, taking into account the effect of the observing (laser-beam) direction on the apparent electric field.