No description available.

The relationship between impulse breakdown stress and stressed oil volume has been extensively investigated for uniform field electrodes. In large high voltage transformers it is essential to extend this relationship to paper insulated non-uniform field situations which are more frequently encountered in practice. This paper develops the stated relationship by experimental tests, using insulated cable to plane configurations, and statistical methods. These tests show that the impulse breakdown strength of an insulated cable to plane or nonuniform field system has a two parameter Weibull distribution, when the oil is considered to be the weak-link of the system. A non-linear regression analysis is then used to find that the Weibull parameters for insulated non-uniform field electrodes are approximately the same as the parameters for the uninsulated uniform field electrodes provided their stressed oil volumes are equivalent. Statistical calculations are used to derive the fundamental relationship between the Weibull parameters for any stressed oil volume and the Weibull parameters for the unit oil volume. With this relationship it is possible for design engineers to approximately determine the probability of electrical breakdown of any insulated cable to plane configuration in large power transformers.

A dissertation submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Master of Science in Engineering in the High Voltage Research Group School of Electrical and Information Engineering Johannesburg, June 2017 South Africa / Servitude availability in space-constrained built-up areas within the Johannesburg or Central Load Network (CLN) poses every-day challenges for power system engineers. Strengthening the backbone 88/275 kV transmission system within the CLN becomes even more difficult when multi-circuit transmission lines are required for increased power transfer capabilities. When uprating is considered to increase the power transfer capability, the withstand levels of existing external insulation demands an optimisation to find a new stress versus strength balance that allows reliable operation of substations at higher voltages. The research includes primarily an investigative simulation study to evaluate the current Eskom available design clearances in terms of their withstand capability when subjected to over-voltage transients. Two voltage range classes were evaluated and the results are discussed. For voltage range 1, it was found that the over-voltage stress was low enough to allow for a higher nominal operating voltage while maintaining the existing clearances. For voltage range 2, existing clearances are also found to be conservative and smaller safety margins will most likely be acceptable. From a transient analysis evaluation, voltage uprating is considered as a very attractive option to increase the power transfer capability of existing substations. Current Eskom clearances for 88 kV and 275 kV are expected to perform well during transients generated in uprated systems. Electrode grading to improve the field gradients in the substation will require attention to increase gap factors. Additional surge arresters are considered to be a cost effective solution to control over-voltages throughout the whole uprated substation. The physical modification of substations to replace strung conductors with tubular conductors, ensuring sufficient outage time to refurbish and rebuild with new equipment will be the most challenging part of uprating existing substations. / MT 2017

No description available.

Cette thèse s’inscrit dans le cadre de l’amélioration du comportement électrique des isolateurs de lignes haute tension ; l’objectif est d’assurer une meilleure fiabilité et qualité d’alimentation en énergie électrique. Ce travail a été motivé par la nécessité de répondre à trois questions liées au comportement des isolateurs verre en zone polluée. La première porte sur la recherche d’une méthode permettant de calculer la tension de contournement des chaînes polluées selon le type d’isolateur et ses caractéristiques. La deuxième question concerne la différence de comportement entre les isolateurs en verre et les isolateurs en porcelaine de type « outerrib » ; ce type d’isolateurs présente une forme spécifique adaptée aux environnements à forte pollution. Les tensions de contournement ainsi que les trajectoires de l’arc sur les isolateurs en verre sont très différentes de celles observées avec les isolateurs en porcelaine. Et la troisième question est relative à la défaillance des isolateurs recouverts de silicone lors des essais en chocs (des impulsions de tension) à front raide. En effet, les isolateurs recouverts d’une couche de 0.3 mm (ou plus) de silicone hydrophobe explosent lorsqu’ils sont soumis à des impulsions de tension à front raide d’amplitude très élevée pendant un temps très court. Différents mécanismes pouvant être à l’origine de l’explosion/éclatement des isolateurs recouverts d’une couche de silicone sont discutés. Il ressort des différents tests et analyses que le mécanisme le plus probable semble être la fragmentation par plasma. En effet, suite à l’application d’une tension à front raide, d’amplitude très élevée, des canaux (fissures) microscopiques prennent naissance là où le champ électrique est le plus intense. L’application répétitive des chocs de tension conduit au développement de décharges dans ces canaux (rupture diélectrique de l’air) c’est-à-dire des arcs (canaux de plasma) qui se développent/propagent dans le volume de l’isolateur. La puissance déchargée (c’est-à-dire l’énergie stockée dans les condensateurs du générateur en des temps très courts) dans ces canaux à chaque choc étant très élevée, elle conduit à l’explosion de l’isolateur après quelques chocs (parfois 5 ou 6 suffisent): c’est la fragmentation par plasma. / This thesis deals with the improvement of the electrical behavior of insulators of high voltage lines; the objective is to ensure better reliability and quality of power supply. This work was motivated by the need to answer three questions related to the behavior of glass insulators in polluted areas. The first one concerns the search for method for calculating the flashover voltage of polluted chains according to the type of insulator and its characteristics. The second question concerns the difference in behavior between glass insulators and "outerrib" porcelain insulators; this type of insulator has a specific shape adapted to environments with high pollution. The flashover voltages as well as the trajectories of the arc on glass insulators are very different from those observed with porcelain insulators. And the third issue is the failure of silicon-coated insulators during shock tests (pulses) with a steep front. Indeed, insulators coated with a layer of 0.3 mm (or more) of hydrophobic silicone explode when subjected to very high amplitude steep-edge voltage pulses for a very short time. Different mechanisms that may be responsible for the explosion / puncturing of insulators covered with a layer of silicone are discussed. It appears from the various tests and analyzes that the most probable mechanism seems to be plasma fragmentation (cracking). Indeed, following the application of a steep front voltage, of very high amplitude, microscopic channels (fissures) originate where the electric field is most intense. The repetitive application of impulse voltages (shocks) leads to the development of discharges in these channels (breakdown of the air), i.e.; arcs (plasma channels) which develop / propagate in the volume of the insulator. The discharged power (i.e.; the energy stored in the capacitors of the generator in a very short times) in these channels (cracks) at each shock being very high, leads to the explosion of the insulator after some shocks (5 to 6 sometimes): it is the fragmentation by plasma or plasma cracking.

Thesis. 1978. Ph.D.--Massachusetts Institute of Technology. Dept. of Physics. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Vita. / Bibliography: leaves 225-239. / Ph.D.

Since the discovery of the long-range ferromagnetic order in two-dimensional and multi-layered van der Waals crystals, and the observation of a nontrivial topology of the magnon bulk bands in the chromium trihalides, the bosonic honeycomb lattices have drawn significant attention within the condensed matter community. In this thesis, we employ a Heisenberg model with a Dzyaloshinsky-Moriya interaction in a honeycomb ferromagnetic lattice to study the properties of bulk and edge spin-wave excitations (magnon). By the Holstein-Primakoff transformations in the linear spin-wave approximation, the spin Hamiltonian is written as the bosonic equivalent of the Haldane model for spinless fermions. We present a simple bosonic tight binding formalism which allows us to obtain analytical solutions for the energy spectrum and wavefunctions. We investigate three basic boundaries in the honeycomb lattice: zigzag, bearded and armchair, and we derive analytical expressions for the energy band structure and wavefunctions for the bulk and edge states, and with both zero and nonzero Dzyaloshinsky-Moriya interaction. We find that in a lattice with a boundary, the intrinsic on-site interactions along the boundary sites generate an effective defect and this gives rise to Tamm-like edge states. If a nontrivial gap is induced, both Tamm-like and topologically protected edge states appear in the band structure. The effective defect can be strengthened by an external on-site potential, and the dispersion relation, velocity and magnon density of the edge states all become tunable. We also investigate the bond modulation in the bosonic Haldane model, where by introducing a Kekule bond modulation and with the analysis of the gap closing conditions and the bulk band inversions, we find a rich topological phase diagram for this system yet to be discovered. We identify four topological phases, verified by a numerical calculation of the Chern number, in terms of the Kekule modulation parameter and the Dzyaloshinsky-Moriya interaction. We present the bulk-edge correspondence for the magnons in a honeycomb lattice for both armchair and zigzag boundaries. We believed that our study in this thesis will be important for possible applications of magnons in data process devices such as magnonics.

Insulators used in the construction of spacecraft are irradiated with high-energy electrons in the space environment and this sometimes causes the insulators to charge to very high voltages. Such charged insulators can generate spontaneous electric partial-discharge pulses of the order of mA to tens of A. These pulses sometimes last enough time to destroy the expensive micro-circuitry present in the spacecraft. In evaluating the threat to the spacecraft due to these discharges, calculation of the resistivity becomes a critical parameter since it determines how accumulated charge will distribute across the spacecraft and how rapidly charge imbalance will dissipate. So far, resistivity values for the insulators for spacecraft applications have been simply imported from tabulated results measured using standard American Society for Testing and Materials (ASTM) and International Electro-technical Commission (IEC) methods. This thesis work provides the details of the charge storage method which has been found to be more appropriate in calculating the resistivity of spacecraft insulators by emulating the space environment better. This method is based on the concept that the resistivity is better measured as the decay of the charge deposited on the surface of an insulator, rather than by the flow of current across two electrodes around the sample which is the case with the classical method of measurements. From the results obtained from the charge storage method, it has been found that the ASTM resistivity values for thin film insulating spacecraft materials have been found to under-predict charge transport values applicable to many spacecraft charging problems, by 10 to 104 times. The charge storage method has only one side of the insulator in vacuum exposed to charged particles, light and plasma, with a metal electrode attached to the other side of the insulator. The chamber for measuring the charge storage decay has been designed with the capability to measure 32 samples simultaneously. The details of the apparatus, instrumentation, test methods, data acquisition methods, and data analysis for measuring resistivity of the spacecraft insulators are given here. Details about the vacuum environment, sample mounting, isolation of the samples, charging of the samples, measurement of the surface charge, rotary motion of the sample carousel, etc., are also given. The report also includes differences between the classical methods and the charge storage method both in terms instrumentation and methodology. The results obtained from both methods are tabulated showing the superiority of the charge storage method. Recommendations for future work are also included.

Advancements in power electronics to higher power levels and faster switching times allow new machine and systems designs, but also create higher stresses on electric machinery insulation. High performance, pulse-width modulated (PWM) inverters are now available for medium voltage drive systems, and are being considered by the U.S. Navy as they move to the "all-electric" ship. If this process is to be successful, a necessary component will be to understand the impact of partial discharge (PD) generation on electric drive systems. Out of the many PD influencing parameters, voltage level, voltage rise-time, switching frequency, and temperature were chosen to be investigated with regards to their influence on PD generation in a comprehensive research project in the Motor Systems Research Facility (MSRF) at Oregon State University (OSU). The tests were performed on representative propulsion coils employing two different 4160 V insulation systems and were evaluated by both an optical and electrical PD detection method. A highly flexible test configuration was developed, capable of adjusting each of the four test parameters independently over a wide range of appropriate values. The developed test program enabled the analysis of the influence of the parameters on the generation of PD, as well as an evaluation of the test coils and PD instrumentation used. It is concluded that, as expected, voltage level is the most significant parameter affecting PD production. However, there is a surprising interdependence of rise-time and pulse-width that requires further investigation. Multiple-cycling tests are seen as appropriate to determine the effect of temperature. Based on the subjective nature of the findings from the test program an improved PD instrument is proposed, which would increase the capabilities and objectivity of the PD detection process. / Graduation date: 2004

In this thesis the concept of inhomogeneous dielectrics is demonstrated for various optical coating applications. Compositionally-varying silicon oxynitride (SiON) dielectric layers, with the refractive index varying as a function of position, are grown by computer-controlled plasma-enhanced chemical vapor deposition (PECVD) using silane, nitrogen, and nitrous oxide reactant gases. Compositionally graded and superlattice-like SiON layers are grown and their compositional profiles are confirmed by Auger electron spectroscopy sputter profiling. Inhomogeneous antireflection coatings and rugate filters, with sinusoidally varying refractive index profiles, are designed and fabricated and their measured spectral responses are found to be in excellent agreement with simulated results. Alternating-current thin film electroluminescent (ACTFEL) devices with multiple layer dielectrics also are designed, fabricated, and the insulating layers are shown to increase the optical outcoupling efficiency of an ACTFEL devices by approximately 14 % compared to that of a conventional ACTFEL structure. / Graduation date: 1993