Electric power switches

A review is given of the physics and engineering behavior of arcs in vacuum and axial gas blast under the conditions found in high-voltage circuit breakers. The topics included are: the zero energy switch; current limiting switches; arc control; the gas-blast circuit breakers; the low-current gas-blast arc; arcing with ablation; interruption dynamics; arc modeling; vacuum as a switching element; the vacuum arc; the constricted arc; and the properties of SF₆     

Nature of convection-stabilized DC arcs in dual-flow nozzlegeometry. I. The cold flow field and DC arc characteristics

An experimental investigation of the steady-state low current air arcs in a dual-flow nozzle system is presented. The cold flow field with no arc was determined for various nozzle geometries, i.e. two- and three-dimensional and orifice nozzles, and nozzle pressure ratios. Supersonic flow separation and oblique and detached shock waves were observed in the flow field. Using a finite-element computer program, the Mach number contours were determined in the flow field for various nozzle-gap spacings and pressure ratios. In addition, the DC arc voltage and current measurements were made for an electrode gap spacing of ≈5.5 cm and current levels of I≈25, 50, and 100 A for the three nozzle geometries. The arc voltage and arc power increased rapidly as the flow speed increased from zero to sonic velocity at the nozzle throat. The shock waves in the converging-diverging nozzles resulted in a decrease in the overall resistance by about 15%     

Study of the Arc and of the Interaction between Arc and Operating Mechanism in SF6 Puffer Breakers

The article deals with studies of the arcing performance of SF6 puffer breakers at currents of up to 80 kA and voltages of up to 200 kV, using an interrupter unit fitted with viewing windows. The studies are mainly concerned with the behavior of the arc at current zero and with restrike phenomena under various conditions. In addition to measurements of the electrical variables and the pressure variation, the tests cover the use of high-speed cameras, spectroscopic diagnostic methods, and schlieren methods. The density fields and flow fields derived from the latter are used to optimize the gas flow. On the basis of the breaker data and those of its operating mechanism a method for computing the essential variables, e.g., pressure, gas flow, back pressure, and contact travel, is developed. The variation of these parameters as a function of the interrupter data for various currents and with due allowance for the interaction between the arc and the operating mechanism must be known if the breaker design is to be optimized. The method uses simplified equations for the gas flow and the arc. The relationship between the gas data (e. g., density, enthalpy, speed of sound, electrical conductivity, and dielectric strength) and the temperature is taken into account by using approximation functions. The temperature of the arc core is matched to the test results.     

Contact temperature and erosion in high-current diffuse vacuum arcson axial magnetic field contacts

We have investigated the surface heating effects of drawn vacuum arcs for several industrial designs of axial magnetic field (AMF) contacts, using near infrared (IR) photography of the Cu-Cr arcing surfaces with an image-intensified charge-coupled device (CCD) camera and an IR pyrometer. This enables detailed contact temperature mapping immediately after a half-cycle of arc current. The very homogeneous temperature distribution observed at current zero stands in contrast to the visually nonhomogeneous high-current diffuse arc, which was studied in separately reported experiments using high-speed digital photography and arc voltage measurements. The peak temperature at current zero increased relatively linearly with the peak current I_P, and reached well beyond the melting range. We combine the temperature maps with a heating model to determine the thermal sheath thickness after arcing and its dependence on I_P. The results suggest that near the interruption limit of AMF contacts, the interaction of the stable high-current arc with the anode and cathode is dominated by processes induced by flowing liquid metal, which redistributes the heat input from the axially concentrated arc over most of the contact surface. Furthermore, the flow of liquid metal off the cathode and anode faces contributes to the overall contact erosion     

The Dynamic Behavior of Nozzle Arcs

The dynamic behavior of ac arcs in a supersonic nozzle has been investigated. The close coupling between the arc and its external flow determines the arc characteristics as well as the flow transients within the nozzle in a gas-blast circuit breaker. For affinely related nozzles, the dynamic behavior of the system is found to be fully determined by two nondimensional parameters: the nozzle coefficient N [17] and the nondimensional frequency ? which is the product of the angular frequency of the current and the time scale of the system. Of these two parameters, ? controls the qualitative behavior of the arc and its external flow. When ? is of the order of 10-2 or less, a quasi-steady period of arcing exists. Arc modeling can then be conveniently divided into a quasi-steady-state (hence dc) phase and a current-zero period.     

The High Current Metal Vapor Arc Column between Separating Electrodes

We observed metal vapor arcs between separating electrodes in a demountable vacuum chamber using high speed photography. The peak values of the ac arc current half-wave ranged from 5 kA to 67 kA. Determination of the arc appearance as functions of arc current and electrode gap revealed that the arc can assume various types of columnar forms when the current at the instant of electrode separation exceeds 7 kA. The duration of the columnar arcing forms is influenced by axial magnetic fields differently for different field strengths. The graphical representation of the results allows prediction of the most probable arc appearance for a given set of operating parameters. A qualitative explanation of the various arc appearances on the basis of balances between magnetic and kinetic pressures is provided.     

Application of High Current and Current Zero Simulations of High‐Voltage Circuit Breakers

This paper reports on the use of computational fluid dynamic (CFD) simulations to predict the interruption behaviour of high‐voltage circuit breakers (HV‐CB) using the self‐blast principle. Two different levels of accuracy of the arc model are proven to be sufficiently accurate for simulating the high‐current phase and the period around current zero (CZ). For the high‐current phase, a simplified equivalent model of the arc is implemented to predict the pressure build‐up, and even more important to accurately trace the hot gas from the arcing zone into the exhausts and the heating volume. A detailed analysis of the gas mixing in the heating volume for different arcing times and current amplitudes showed the optimum geometrical design of the heating volume. For the CZ phase, a more detailed arc model is needed including the effects of ohmic heating, radiative energy transfer, and turbulent cooling fully resolved in space and time. The validation with experiments was done and shows good agreement which justifies the use of the implemented model. With it, scaling laws varying only one parameter at a time (pressure and applied current slope) were derived and confirm previously found empirical laws. This is of particular interest, as it is very difficult to derive such scaling laws from experiments where the scatter is always very large and where it is impossible to vary only one parameter at a time. The influence of the most important geometrical parameters of the nozzle on the interruption performance is shown. In addition to previous experimental indications of this, the simulation reveals that turbulent cooling on the arc edge is the main reason for the difference in interruption performance. Moreover, the exact spatio‐temporal build‐up of arc resistance and with it the detailed understanding of the arc interruption process is possible and shown here for the first time. These simulations enable us to predict HV‐CB performance and to minimise the number of development tests and are routinely used in new development projects. (© 2006 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

A rotating arc gas pump for circuit breaking and other applications

A rotating arc circuit breaker is described which uses an auxiliary current source to generate the magnetic field for driving the arc. Test results obtained using optical fiber measurement systems have shown that there are three main arcing phases. Initially the arc rotates at an essentially constant but low velocity, subsequently its velocity oscillates between this and much higher values, and finally the are plasma may become diffuse in nature. Test results obtained with dielectric strength probes have indicated that a unidirectional flow of arc heated gas is generated. The flow is away from the moving contact of the interrupter so promoting good dielectric strength in this critical contact region. The combination of the optical fiber and dielectric probe results indicates two possible modes of gas pumping represented, respectively, by a fan and a piston-type action of the arc. Simplified analytical models for both modes are developed with predictions obtained showing good agreement with the experimental results. Discussion of experimental results suggests that the transition from oscillatory velocity changes to diffuse arcing represents an important parameter for scaling the geometries of future interrupters and arc heaters     

Experimental Investigation on Arc Phenomena in SF6 Puffer Circuit Breakers

A detailed observation of an arc in a model puffer-type SF6 gas circuit breaker in the current range between 10 and 50 kA (rms) has been carried out. It was found that the arc column remained stable on the center axis during the high-current region, then became turbulent near current zero. It was found that the time interval during which the turbulent arc was observed decreased with increasing values of the peak current. These phenomena indicated that the thermal effects of high-current arcs remain even at current zero. It also was observed that the arc diameter at the nozzle throat outlet was smaller than that at the throat (29 mm), even at a current as high as 70 kA (instantaneous), and that the boundary of gas flow at a downstream region had a very large diameter when the arcs were present. However, around current zero the boundary diameter became as small as that without arc.     

Mechanisms for Temperature Decay in the Freely Recovering Gas Blast Arc

Energy loss mechanisms for the extinguished gas blast arc channel in free recovery are defined and their comparative magnitudes are explored for both N2 and SF6 gases. The arc channel temperature decay rate is found to follow at least two time constants: one corresponding to the transit time of the channel gas and the other to a later period. In addition, the influence of the gas pressure and of the initial conditions of the arc channel and the surrounding hot gas mantle at current zero on the decay rate of the channel temperature are investigated.     

Mass spectrometry of high pressure arcs in air and SF6

For the development of SF₆ pufferbreakers it is important to gain insight into the chemical composition of the SF₆ gas during the high-current period as well as close to current zero. For this reason a mass-spectrometric sampling system is presented which allows dynamic measurements of chemical reactions occurring in high-pressure arcs. Results of mass-spectroscopic investigations on a dc and a pulsed arc are presented.     

Time-Resolved Laser Doppler Measurements in the Underexpanded Jet of a Model Gas-Blast Circuit Breaker under Arcing Conditions

A differential laser Doppler system has been used in conjunction with a self-triggered oscillographic raster display system for measuring particle concentrations and flow velocities in a model gas-blast circuit breaker. The flow velocities have been correlated with particle size to enable extrapolation to be made to determine the true plasma flow velocity. The results have been deconvoluted to take account of lateral movements of the arc column. The measurements have been taken with air as the host arcing medium and a flow through a 25-mm-diameter orifice sustained by an upstream pressure of about 7 bar (downstream pressure 1 bar). The arc was sustained by sinusoidal cuffents of peak values 3 and 8 kA and a frequency of 85 Hz. These two currents corresponded to conditions when the model circuit breaker successfully and unsuccessfully interrupted the arc-sustaining current, respectively. The results show that in the case of the 8-kA arc, the flow pattern is severely distorted from that under nonarcing conditions, suggesting that this is a contributory reason for the poorer perfonnance of the model circuit breaker at this higher current level.     

Visualization and characterization of high-current diffuse vacuumarcs on axial magnetic field contacts

We have investigated the behavior of drawn vacuum arcs for several designs of axial magnetic field (AMF) contacts using high-speed digital photography and arc voltage measurements, As the peak current was increased, a gradual transition occurred in the arc appearance from a multiple cathode-spot arc to the high-current diffuse mode, and then to a high-current diffuse columnar mode. Two relatively simple models based on the literature are used to explain the results. The first is an empirical criterion for using the arc voltage behavior to determine the maximum arc current for which an AMF geometry can produce a high-current diffuse mode from the initial bridge column arc. The second model predicts the highest arc current that can be forced into a fully diffuse mode for given values of the AMF and the contact arcing radius. The predictions of these models are compared to our experimental and analytical results     

Experimental determination of the impulse response of cascade arcsusing a correlation technique

Low-current cascade arcs are one of the simplest configurations for investigating the behaviour of low-temperature arcs. In cascade experiments, arc conditions may be strictly controlled, in which case the results become reproducible, allowing detailed measurements of high resolution to be made. The response of cascade arcs to a small change in the current gives insights into the physical processes taking place in an arc, such as ionization/recombination, diffusion and energy transfer between the electrons and the heavy particles. In this work, a new method of measuring time constants is described, in which correlation techniques utilising a pseudo-random binary sequence are employed to deduce the impulse response of the arc. This type of experiment provides a useful check on the validity of the collision models used for the reaction processes in the arc plasma and it is planned to investigate other gases in the future. In this paper, the experimental arrangement and the data analysis are discussed and experimental results are presented     

Similarity Relations for the Electric Arc in Forced Axial Flow

The conservation equations of mass, momentum, and energy in differential form, Ohm's law, and the experimentally determined dependence of the interruption capability of the arc on current shape are employed to obtain similarity relations for high pressure electric arcs in forced axial flow around current zero. The similarity relations are then applied to assess the validity of laminar and turbulent flow models for the arc by comparing model predictions with experiment. It is found that the laminar flow model quite often predicts arc behavior contrary to experiment, while the turbulent flow model predictions are much more consistent with experiment. Moreover, the similarity relations should also be useful in exploring arc behavior under circumstances not discussed in this work.     

Unipolar Arcs

The state of the arc investigations in tokamaks is reviewed. The results of the model experiments with unipolar arcs are discussed. The conditions of the arc ignition are examined. Theoretical considerations of arcing on the tokamak walls and in the model experiments are given.     

The DC Arc in a Supersonic Nozzle Flow

The boundary layer integral method at its second level of approximation has been used to study the DC arc in a supersonic nozzle flow. It is shown that with the inclusion of the arc momentum balance, the critical point of the flow is, generally, not the sonic point of the external flow. The speed, at which a disturbance propagates relative to the external flow, is in general supersonic and is dependent on the arc conditions. The arc model is capable of predicting the axial electric field, the arc size and the axial pressure distribution as a function of current. For affinely related nozzles, the solution is determined by a parameter N, which is related to zt, the stagnation condition and the nominal current density at the throat (I/At). Numerical results are given for a particular nozzle shape although the method of analysis is general. Practical implications as regards nozzle design for a gas blast circuit breaker are briefly discussed.     

Experimental and numerical studies of arc restrikes in low-voltagecircuit breakers

This paper is devoted to the study of the arc restrike phenomenon in low-voltage circuit breakers. We focused our interest on the type of arc restrike that can be described as a sudden reignition in the arcing contact region while the arc was situated in the quenching chamber a few tens of microseconds before it occurs. Our experimental investigations have established that the critical arcing contact region is still crossed by a so called residual current on the order of several amperes. A gas temperature around 4200 K was derived from electrical measurements in this region before the arc restrike occurrence, We also demonstrate that the restrike takes place through the growth of the remaining current of several amperes in the arcing contact region. A numerical approach was carried out with a two-dimensional hydrodynamic code in order to simulate the gas behavior in the arcing contact region before and during the arc restrike phenomenon. The same temperature as the measured one is calculated just before the restrike. It is demonstrated that the current density appears to be the most sensible quantity. A critical value of 5 A/cm² was calculated     

Vacuum breakdown on metal surfaces

The unipolar arc model is described. Experimental proof that unipolar arcing represents a discharge form which easily leads to explosive plasma formation is provided. Using a laser-produced plasma, it has been demonstrated that unipolar arcs ignite and burn on a nanosecond time scale without any external electric field being applied. Similar unipolar arc craters have been observed on the cathode surface of a pulsed vacuum diode with an externally applied field of 0.5 MV/cm. The experimental results show that cathode spots are formed by unipolar arching. The localized buildup of plasma above an electron-emitting spot naturally leads to a pressure gradient and electric field distribution which drives the unipolar arc. The high current density of a unipolar arc provides explosive plasma formation