Search

In countries with a high ambient temperature and strong solar irradiation, transformer winding hot-spot temperature may increase over its maximum permissible limit. This can considerably reduce the insulation life of the transformer by enhanced degradation of the paper insulation. According to current loading guides, for each 6 K increase in working temperature, the ageing rate increases with approximately a factor two. Therefore, it is important to take into consideration the impact of the sun on the power transformer thermal behavior. In this paper, a modified hot-spot temperature model is presented to account for the effect of transformer winding temperature rise by solar irradiation. The effects of solar irradiation on transformer winding paper insulation are shown by comparing the degree of polymerisation (DP), the fault probability and the remaining life. Here, the fault probability is defined as the probability that the estimated DP-value at a certain moment in time is below a certain end-of-life criterion (threshold value). An additional winding hot-spot temperature rise of 9 K during the summer and a temperature rise of 6 K during the winter may occur in countries with strong solar irradiation. This may result in a reduction of the remaining lifetime by up to 40%.

The age of the majority of power transformers installed in the western electricity network reaches 30 to 60 years and replacement on short term seems eminent. A technically sound policy concerning the replacement of these assets requires a model that estimates the life expectancies of individual components and from that calculates parameters related to the behavior of a population of assets as a whole. A probabilistic approach is adopted and is applied to thermal degradation of the transformer paper insulation. In this paper, we will focus on the determination of the population reliability from individual reliabilities. These individual reliabilities are based on Arrhenius modeling of paper insulation degradation, including the inherent uncertainty in the parameters involved. A statistical failure model is used to obtain the population reliability figures. The modeling method is demonstrated on two populations of power transformers in The Netherlands to evaluate the different replacement alternatives. Using the model, strategies can be defined to maximize transformer utilization and postpone replacement. The downside is the need to replace the complete fleet in a relatively short time afterwards.

Uncertainty analysis is an essential part of models that estimate remaining lifetimes of power transformers. Uncertainty can be included in modeling the insulation aging process. If the uncertainty becomes too large, the insulation condition can, in principle, be determined by measuring the DP value, and more accurate projections can then be made. A drawback of the model presented here is the required input data, i.e., average values of dynamic load and ambient temperature, and maximum values, because the degradation rate is not a linear function of temperature. This information is not available for most transformers. However, even if the loading data are available only over times when the load approaches the rated value, a reliable estimate of the paper degradation can still be made. The observation that tap-changer failure is still the major transformer failure mode supports the simulated results presented above, which show that a transformer failure peak due to paper insulation failure is expected only after several decades of service. Although prediction of transformer performance may be considered speculative because of uncertainties in the model parameters and incomplete service data, it can serve as a tool to compare different maintenance and replacement strategies. Thus annual replacement strategies appear superior to equal loading because the latter provides no guidance on the order of replacement. It has also been shown that transformers should be replaced before the beginning of the failure wave and that variation in annual load growth between transformers leads to spreading of the replacement wave. A disadvantage of a distributed transformer loading is the reduction of the average remaining lifetime due to relatively fast degradation of the more heavily loaded transformers. Comparison of these strategies requires a technical reliability model, which is the foundation of risk-based asset management.

This paper presents an approach for obtaining science-based answers to practical vacuum device problems. The question "How to translate scientific knowledge to design and operation practice" is rephrased to the more productive question "What research activities are required to provide science-based answers to practical problems!" Such activities involve the study of operational mechanisms in realistic, geometries with the help of appropriate diagnostics. Within this framework, recent research developments on vacuum insulator flashover are reviewed, with emphasis on the implications for device performance and design. Secondly, an overview is given of recent research activities performed at the Eindhoven University of Technology. By studying the relationship between flashover mechanisms and insulator performance, and the relationship between insulator performance and design, we aim at improved device performance, the formulation of guidelines, and the development of device diagnostics, for dc and ac vacuum devices.

We discuss a number of practical implications from recent studies on HVDC design concepts for vacuum components. These studies dealt with microwave tube technology. The conclusions, however, are valid for a wide range of components. The goal of this work is to provide a scientific basis for the design of HVDC vacuum components. From a study of breakdown and emission mechanisms, and from the measured insulating performance of many different geometries, we have derived guidelines for the design of, for example, insulators and cables. It is further shown how conditioning procedures and operating conditions (operating pressure, insulator charging) should be reflected in the design. We discuss a number of practical implications regarding insulator design, conditioning, vacuum vs. air operation, HV cables in vacuum and potting. >

This paper presents the remaining lifetime calculation of power transformers paper insulation and consequently of power transformers. The calculations are performed based on two models, which are related to the thermal degradation of the cellulose winding paper insulation: the common IEC loading guide and a paper degradation model. The paper insulation model's prediction can be improved by involving data from furfural analysis. The remaining lifetime is extracted from the fault probability (reliability) of the paper insulation. The two models are brought together, to aid the asset manager in the decision making process. A probabilistic approach is used, which can be coupled to analysis in terms of risks, benefits, costs, and availability by the asset manager.

As part of a study on HV design concepts for microwave tubes, a number of different insulator designs have been studied. Analysis of the measured DC current, partial discharge activity and breakdown voltage shows that surface charging of insulators is a key mechanism in the breakdown process and in the conditioning process. Insulator parameters are not only the breakdown voltage, but also the conditioning speed and the sensitivity to gas exposure or charge leakage. In all these respects insulators with a field enhancement at the anode are superior. Field enhancements at the cathode are less harmful if stepped insulator shapes are chosen. Effective conditioning requires at least a limited number of breakdowns. With sufficient conditioning breakdowns, all insulator geometries tested reached a breakdown field exceeding 12 kV/mm. >

The age of the majority of power transformers installed in the western electricity network reaches 30 to 60 years and replacement on short term seems eminent. A technically sound policy concerning the replacement of these assets requires a model that estimates the life expectancies of individual components and from that calculates parameters related to the behavior of a population of assets as a whole. We will illustrate the approach by applying it to a well-known degradation process: thermal degradation of the transformer paper insulation. In this paper we focus on the determination of the population reliability from individual reliabilities. These individual reliabilities are based on Arrhenius modelling of paper insulation degradation including its inherent uncertainties. A statistical failure model is used to obtain the population reliability figures. We demonstrate the modelling method by applying it to an existing population of power transformers in the Netherlands, to evaluate the expected replacement wave. Further, the model is applied to analyse alternative replacement scenarios.

The age of the majority of power transformers applied in the western electricity network varies between 25 and 50 years. Depending on the load history and time of operation, replacement on short term is imminent. A technically sound policy concerning the replacement of these assets must be based on knowledge of (i) the life expectancy or reliability of individual components, (ii) how these failure probabilities cumulate to a replacement wave, and (iii) how to manage an expected replacement wave. The population reliability is obtained from individual transformer reliabilities using Arrhenius based modelling of paper insulation degradation. This modelling technique includes measures to cope with inherent uncertainties in available data. Population reliability figures are obtained using an adapted k-out-of-N failure model. The modelling method is applied to existing populations of power transformers in The Netherlands, to evaluate their expected replacement wave.

A report is presented on the analysis and improvement of HV traveling-wave tube (TWT) components for use in spacecraft. The investigation includes a theoretical study, involving electric field and pressure calculations, and an experimental one, involving partial-discharge measurements as a supplement to the standard voltage withstand tests. Improvement of the designs under study has been achieved mainly by the reduction of the cathode electric fields, thus reducing primary emission. A problem inherent in most space tubes is the fact that they are subjected to preflight tests in air, although designed for use in the satellite vacuum. This concerns primarily the outside connections and the HV cable. Because vacuum and air operation impose different and sometimes conflicting design requirements, these preflight tests are not representative of the HV performance in orbit and may even cause insulation degradation as a result of partial discharges. The methods used are presented, discussed, and illustrated, and, based on the results, some general conclusions and recommendations are formulated regarding the design, operation, and testing of HV-TWT components. >

In view of the coming major asset replacement wave of power grid components, new lifetime prediction tools are developed to assist asset managers in making technically and economically sound decisions. This paper discusses the modeling technique based on the concept of quality parameters applied to paper insulated systems. The concept of quality parameters was recently introduced for power transformers to obtain a measurable quantity which relates the actual degradation process to an observable quantity. The present paper discusses a general approach for modeling other power components along the same lines. In case of a power transformer, oil samples can be taken and analyzed in order to estimate the DP (Degree of Polymerization) value, which is linked to the mechanical strength of the insulation material. This quantity then serves as quality parameter. Using a probabilistic approach based on the physical degradation mechanism a remaining lifetime model was developed. This resulted in a scheme, which is flexible in the sense that any new available data can be included to enhance the prediction reliability. The question addressed here is whether a similar approach can be applied to other power components as well. In particular we have investigated the possible use of quality parameters which are not directly correlated to one degradation mechanism. We will define different kinds of quality parameters and propose a component reliability model that incorporates all of these.

In this paper the concept of an integral transformer lifetime model is presented. The model provides the best possible prediction of future behaviour given the data available. It treats remaining life in terms of future failure probability, thereby giving better support to the decision taking process than a mere remaining life estimate. The core of the model is a generic description of ageing processes, coupled to a probabilistic approach. The approach presented utilizes various techniques to reduce the uncertainty that is inherent to modelling processes with incomplete knowledge of the operational history. One technique couples the process model to externally measurable quantities; another technique involves a sensitivity analysis, which shows what additional input data gives the most efficient way to improve accuracy. We will illustrate the approach by applying it to a well-known degradation process: thermal degradation of the transformer winding insulation.

In the coming decades the systems for generation, transmission and distribution of electrical power undergoes significant changes due to technical, economical, social and environmental drivers and developments. In this paper we discuss how these changes affect future maintenance. First of all there is a continuous evolution of maintenance practice as such. Secondly maintenance practice have to adapt to the changes in power system design and operation. We will show that maintenance of the future infrastructure involves more than fixing defects: it starts with the design of components and grid. Maintenance involves not only intelligent corrective and preventive maintenance activities but also preventive action in terms of design and operational practice. Based in this evaluation we conclude that future maintenance involves a smart mix of some or all of 4 maintenance strategies: 1) maintenance free component design, 2) system design including maintainability, 3) effective and efficient maintenance processes and 4) controlled stress system operation

Fast swarm measurements have been performed in hexafluoropropene (C/sub 3/F/sub 6/) with pressures (p) ranging from 20 to 200 Torr and E/p values from 80 to 110 V/cm-Torr. The apparent swarm parameters, derived from a simple model (ionization and attachment only), confirm the pressure dependences reported in the literature. The existence of detachment has been identified from the observed time-resolved avalanche-current waveforms. A model that includes detachment and conversion in addition to ionization and attachment processes has been applied to derive the real swarm parameters in C/sub 3/F/sub 6/. The study indicates that two-body collisional detachment and three-body stabilization of the transient anion C/sub 3/F/sub 6//sup -*/ are responsible for the observed pressure dependence of the dielectric strength of C/sub 3/F/sub 6/. >

The bandwidth limitations which affect gap current measurements, and which are important for pulsed swarm experiments, are analyzed. Such limitations are caused by the electrode geometry, geometry imperfections and the components and equipment used for signal transport and storage. Recommendations are given to optimize the time resolution. The analysis is applied to a pulsed swarm setup for which a time resolution of 1.2 to 1.4 ns has been achieved. >

In this work, a high bandwidth experimental setup is used to measure the inception voltage, the waveform and the apparent charge of partial discharges (PD) in artificial voids in solid insulating materials. The investigations include microvoids, i.e. voids with characteristics dimensions below 100 /spl mu/m. In this paper, the results for PTFE (teflon) are presented. Several simplified models are used to estimate externally measurable PD quantities for given void geometries. A one-dimensional hydrodynamic model is used to obtain better insight into the relevant mechanisms and processes by comparison of measurements and simulation data. Experimental studies show that the measured values for the inception voltage correspond well with calculated inception voltage based on the critical streamer criterion. The hydrodynamic model provides a good qualitative description of the phenomena. The trends of externally measured correspond with those predicted by the authors' models.

A time-dependent, hydrodynamic, model of partial discharges in voids in dielectrics is presented. The partial discharge is regarded as a space charge dominated electron avalanche, or streamer. The model incorporates the collisional and radiative processes in the void and at the boundary, and the field distortion due to space charge and surface charge. The coupling with the external circuit is taken into account. Simulations show that, when increasing the Laplacian field, the charge saturates above a threshold, due to field quenching. The threshold field depends on the geometry and is higher than usually quoted inception fields. The saturation charge may deviate by an order of magnitude from values obtained with simplified models. Accurate reproduction of measured waveforms requires a 2D hydrodynamic model.

This paper presents a study on the feasibility of void detection in mm-sized glass-bead medium voltage devices with partial discharges induced by external fields. Based on an evaluation of the field required for discharge inception and the attainable sensitivity it is concluded that voids with sizes in the range of 10-100 /spl mu/m can be detected provided that the setup is compact and well-shielded, and uses a cylindrical geometry with the test-object immersed in an insulating fluid with a high dielectric constant.

In this paper we present the diagnostic approach and the techniques developed by KEMA for condition based maintenance of transformers, switchgear, generators and cables. The diagnostics include UHF and VHF PD-detection techniques, dynamic measurements on moving parts including the electric drives used (switchgear, tap changers) and loss measurements. The condition-assessment techniques involve pattern recognition, fingerprint techniques and the application of performance indicators and knowledge rules obtained from field experience. As an example our approach will be illustrated with two techniques employed to assess the dielectric condition and the remaining life of transformers.

Power equipment such as turbines, generators, transformers, switchgear and cables represent a major capital investment in power stations and grids. With a high duty-cycle, or repeated short duration overloading, aging of such components may be appreciable. In practice, however, many components perform well even after their depreciation term has expired. As a result, extension of service time is a serious option for reduction of investment costs. On the other hand, life extension may increase the risk of failure, depending on the operation and maintenance practice. Adequate condition assessment (CA) enables to determine effective strategies for operation, replacement, thereby economical catastrophes. Before elaborating on CA-techniques, the authors specify three different types of condition relevant to power transmission and distribution equipment: technical, strategic and economical condition.

An experimental study of prebreakdown phenomena in dry air along insulator surfaces is presented. Differences in time-to-breakdown curves indicate a different discharge behaviour. Localized field enhancement is proposed as the physical explanation of the occurring phenomena.

In this paper processes preceding breakdown are described; the so-called prebreakdown phenomena. It deals with a time-resolved optical and electrical study of the avalanche and streamer formation in atmospheric nitrogen and dry air. With the results from experiments and two-dimensional model simulations the formation of avalanches and streamers, and the dominant processes involved are illustrated and discussed. A fast experimental setup is used to measure the discharge current and the optical discharge activity.

In the design of electrical gas insulated systems data on breakdown fieldstrengths is extensively used. Breakdown voltages for gases are usually given by Paschen curves, which serve as an engineering tool in high voltage engineering. This work is a continuation of earlier work1, 2 and deals with a study of avalanche and streamer formation in air along teflon (PTFE) insulators.

This work is a continuation of work presented earlier for nitrogen and dry air1, and deals with a time-resolved study of the prebreakdown phase in SF 6. the goal of this work is to develop and verify discharge models for practical insulating geometries on the basis of physical discharge processes.

This work deals with a time-resolved optical study of the avalanche and streamer formation phases leading to breakdown in atmospheric nitrogen. We present the results obtained for nitrogen, from experiments and two-dimensional model simulations. This model is used to obtain a better insight in the relevant mechanisms and processes by a comparison of measurements and simulation data. The trends of externally measured quantities correspond with those predicted by our model.

In this paper we present measured partial discharge current waveforms for micrometer voids in PTFE (teflon) and PC (polycarbonate). From the time-resolved partial discharge current waveforms detailed information on discharge mechanisms is obtained. The experimental studies show that the measured values for the inception voltage correspond well with calculated inception voltages based on the critical streamer criterion. The trends of externally measured quantities correspond with those predicted by our models and seem to hold their validity for micrometer voids.

In this contribution we discuss a number of practical implications from recent studies on high- voltage design concepts used in microwave tube technology. We show how the insights gained can be used in the design of, for example, insulators and cables, and how conditioning procedures and operating conditions (operating pressure, insulator charging) should be reflected in the design.

In this paper, we will present the results of a numerical study of the avalanche to streamer transition in an overvolted uniform field gap in atmospheric N2 and dry air. The model used is fully two-dimensional in the solution of the representative set of continuity equations, as well as in the determination of the space-charge field, gas-phase photoionization and cathode photoelectron emission. The spatio-temporal development of the electron density as well as the temporal development of the cathode and anode directed streamer velocities are presented and will be discussed.

A time-dependent, hydrodynamic, model of partial discharges in voids in dielectrics is presented. The partial discharge is regarded as a space charge dominated electron avalanche. The model incorporates the relevant collisional and radiative processes in the void and at its boundary, and the field distortion due to the presence of space charge and void surface charges. Also the coupling with the external circuit is taken into account. Model simulations show that, as a result of field quenching, the associated charge versus applied field shows saturation above a certain threshold. The threshold field depends on geometry and initial electron number, but is higher than the inception values usually quoted. The saturation charge may deviate by a factor of 2–10 from values obtained with simplified models.

As part of a study on high-voltage design concepts used in microwave tube technology, a number of different insulator designs were studied. This work demonstrated the presence of surface charge and its effect on the voltage holdoff performance and the conditioning process. Insulators may be characterized by the breakdown voltage, the conditioning speed, and the dependence on surface charge. Insulators with anode field enhancements gave the best performance. Field enhancements at the cathode side were less harmful if stepped insulator shapes were chosen. Effective conditioning required at least a limited number of breakdowns. With sufficient conditioning breakdowns all insulator geometries tested reached an averaged breakdown field exceeding 12 kV/mm.<>

Positive synergism in binary gas mixtures is studied by means of fast time-resolved current measurements with a time resolution of 1.4 ns. In this paper we present experiments in pure 1-C3F6, in pure c-C4F8 and in an optimized mixture of 40% 1-C3F6 and 60% c-C4F8. We have experimentally verified that the mixture of 1-C3F6 and c-C4F8 has a higher rate of conversion from unstable negative ions to stable ones than either of the two constituent gases. This enhanced stabilization rate is responsible for the observed synergetic effect. The conclusions are supported by an extended avalanche model which includes electron detachment and stabilizing ion conversion processes, next to the commonly assumed ionization and attachment processes.

Voltage transients generated in GIS by the operation of circuit breakers or disconnect switches induce high voltages across interruptions of the enclosure. In this paper we discuss such interruptions and their necessity. We present fast measurements of voltage and current at the GIS/cable-interface of a 150 kV substation, where the enclosure is interrupted to permit current measurement. Circuit breaker operation causes a single voltage transient across the interruption with an amplitude up to 40 kV, a duration of 200–300 ns, and a risetime of about 5 ns. Disconnect switch operation causes multiple transients, each having a smaller amplitude and a longer risetime. The voltage across enclosure interruptions is adequately suppressed, and breakdown is completely eliminated, when the interruption is bridged with resistors. The resistors do not affect the current measurement.

The authors analyze the avalanche growth in electronegative gases which exhibit detachment and conversion processes in addition to ionization and attachment processes. The study is based on an evaluation of the model of H.F.A. Verhaart and van der Laan (1982). Humid air is taken as an example to demonstrate, with this model, the effects of detachments and conversion processes on the avalanche electron distribution across the gap and on the external current waveform. The model is also used to determine swarm parameters from fast swarm experiments. The results show that detachment and conversion affect the electron distribution in the avalanche and can thereby influence the streamer breakdown threshold. The neglect of these process in the analysis of swarm experiments results in measured values for the effective ionization coefficient that can may substantially differ from the true values. >

A report is presented on theoretical and experimental methods suitable for evaluating the high-voltage (HV) performance of spacecraft traveling-wave-tube (TWT) components. The field calculations, pressure calculations, and experimental methods used are presented, discussed, and illustrated, and some general conclusions and recommendations are formulated regarding the design, operation, and testing of spacecraft HV TWT components. >

A free burning stationary arc in cesium-seeded argon at normal pressure has been investigated. Different spectroscopic techniques have been used to obtain profiles of electron density and electron temperature for discharge currents between 1 and 4 A, at a wall temperature of 520 K. The energy balances of electrons and of heavy particles have been evaluated. They describe the arc behaviour well.

The relation between the motion of charges in an electrode gap and the current induced in the external circuit is described on the basis of energy considerations and on the basis of the Ramo-Shockley theory. It is shown that both approaches are applicable, and mutually consistent, not only for gas-filled gaps but, if applied properly, also when the gap is partially filled with a solid dielectric. Energy considerations are especially useful for two-electrode gaps. If more than two electrodes are involved, the Ramo-Shockley theory has to be invoked. Based on the theoretical description, fast measuring systems with a subdivided measuring electrode can be constructed (typically 400 MHz bandwidth, 1 ns risetime). Fundamental and electrotechnical bandwidth limitations are discussed. The derived techniques and insights are applied to perform and interpret time-resolved measurements of currents induced by avalanches in homogeneous fields corona discharges in inhomogeneous fields and partial discharges in voids. >

The method of electron density determination from continuum emission is often used in the nonstationary plasma of an MHD-generator because of its simplicity. An assumption in the analysis is that recombinative radiation forms the main contribution to the continuum. Next to this the method carries with it some other uncertainties. In this work the method is compared experimentally with the method of Stark broadening measurement of spectral lines, using a stationary argon cesium discharge. Both techniques involve corrections for plasma inhomogeneities. Agreement within 20% is found. The contribution of Cs 2 molecules to the continuum emission, which is significant in saturated vapor at moderate temperatures (500–1000 K), is estimated to be of minor importance in the generator plasma.

A model, describing the time-dependent behaviour of a noble gas MHD generator plasma, has been set up. With this model it is possible to calculate the relaxation for ionization or recombination as a response to a stepwise temperature development, once the initial and final conditions are given. In the model radiative transitions and a deviation from Maxwellian electron distribution are included. Radiation causes an enhancement of both the ionization relaxation time and the recombination relaxation time. A non-Maxwellian electron distribution results in an increase of the relaxation time for an ionizing plasma because of a depletion of the high energy electrons. A decrease of the relaxation time for a recombining plasma is caused by an excess of high energy electrons. The relaxation time is strongly dependent on the seed ratio and the temperature step.

Avalanche currents have been measured by means of a time-resolved swarm method in c-C/sub 4/F/sub 8/ at pressures between 0.7 and 27 kPa. The recorded avalanche current waveforms are, in comparison with those of most insulating gases, unusual. Whereas, in most gases, avalanche current waveforms are characterized by a clear distinction between the electron contribution (high amplitude, short duration) and the ion contribution (low amplitude, long duration), the waveform in c-C/sub 4/F/sub 8/ is significantly different. During the first electron transit time an electron component can only be observed at low pressure (below 1.3 kPa) and even then is strongly intermixed with the ion contribution. Above approximately 1.3 kPa the waveform seems completely ion-dominated. >