Search

A three-dimensional (3-D) model for multispecies vacuum arc considering ionization and recombination process has been built in this article. With the model, the arc behaviors and species characteristics subjected to actual spatial magnetic fields of cup-shaped axial magnetic field (AMF) contact and transverse magnetic field-axial magnetic field (TMF-AMF) contact are investigated and compared. Furthermore, the validity of the model is verified by comparing with the experiments. Simulation results show that contact structures affect the species distributions of the arc by changing magnetic field and current density distributions. More triple charged ions are generated through ionization under higher electron temperature so that the triple charged ions gather where current constricts. The heat flux density is large in the current contraction region, where the contribution from triple charged ions is large as well. Spatial magnetic field distribution affects the Lorentz force acting on ion components. The force distribution is relatively uniform and the species are subject to inward force on the arc edge under cup-shaped AMF contact. For TMF-AMF contact, the double charged ions are subject to the force pointing to the TMF contact arms, whereas the triple charged ions are subject to the force pointing to the arc center. In addition, the consideration of arc species can improve the accuracy of the simulation results of the heat flux distribution, leading to a good agreement of the erosion condition between experiments and simulation.

As a special phenomenon under transverse magnetic field (TMF) contacts, vacuum arc rotation deserves more attention. In this article, rotation characteristics of three typical TMF contacts, i.e., swastika-type, cup-shaped, and double-TMF contacts, are systematically studied. According to arc images, recorded by a high-speed camera, physical terms related to arc rotation such as rotation duration and rotation speed are clearly defined. The magnetic field and Lorentz force are calculated and the force balance of the moving arc is analyzed. On this basis, the effects of arc ignition mode, peak current, and contact type on arc rotation are investigated. Arc voltage noise as well as arc splashing outside caused by rotation is also discussed. The results show that the duration from ignition to rotation in two-point arc ignition (TPAI) mode is usually the shortest because of the rapidest constructed column formation. As the peak current increases, arc rotation becomes easier. Due to stronger TMF, the arc under double-TMF contacts has the longest rotation duration and the highest rotation speed.

In this article, the model including cathode spot process and deuterium desorption process on zirconium deuterium cathode is developed. Five kinds of particle fluxes on cathode surface and three mechanisms affecting deuterium diffusion in cathode are considered. The model not only describes the deformation of cathode metal, but also describes the variation of deuterium concentration with time. On pure zirconium cathode, the mean charge state is larger but on zirconium deuterium cathode, the total ion number density is larger since there are additional deuterium ions, and simulation results show the discharge currents are very close on the two cathodes, the energy flux density on ZrD0.66 cathode is larger so cathode spot crater develops faster on ZrD0.66 cathode. Ablation rates caused by different mechanisms are calculated and the results show in cathode spot development process, mass loss caused by evaporation is ignored, most of the liquid metal will resolidify on cathode surface, and droplets are the main reason for cathode ablation. The ablation rates due to the ejection of droplets are close to experimental results. The ratios of deuterium ion from different mechanisms in this process are also calculated and it is 93.5% in simulation. Simulation results are in agreement with the other researchers’ works.

In this article, the plasma decay in the postarc stage of vacuum circuit breakers and the influence of the transverse magnetic field (TMF) on this process are investigated with a 2-D particle-in-cell model. In this model, distributions of electric field and plasma are solved by Poisson’s equation and Newton’s law, respectively, in 2-D Cartesian coordinate system. The distributions of charged particles and the velocity of the sheath development in the simulation without magnetic field in this article agree well with those in the simulation solved in cylindrical coordinate system in the published work. The TMF in the postarc phase leads to asymmetric distributions of charged particles, which slows down the plasma decay process in the postarc phase. In addition, the TMF results density fluctuations in the plasma and makes the plasma discontinuous and cluster, which can disturb the ion sheath near the postarc anode and result the local electric fields in the plasma. Electrons can be accelerated and gain high velocities in these local fields. These effects of TMF will slow down the dielectric recovery in the postarc stage of vacuum circuit breakers and may increase the risk of rebreakdown in the postarc stage when the vacuum circuit breaker interrupts large fault currents.

The objective of this article is to investigate the effect of contact material on the formation and transition of high-current anode spots and anode plumes. A series of experiments are carried out in a demountable vacuum chamber at three current levels of 3, 4, and 6.5 kA. Plate contacts made of CuCr30 and CuCr40 with a diameter of 10 mm are adopted. Arc images are captured by a high-speed camera and synchronized with measurements of the arc current and voltage. The experimental results show that the three typical anode discharge modes, namely footpoint, anode spot type 1, and anode spot type 2, are formed at a smaller gap distance as the experimental current increases, and this tendency is independent of the contact material. The arc voltage is higher and the anode spot appears earlier and lasts longer under the CuCr40 contact than under the CuCr30 contact. As a result, the footpoint is formed at a smaller gap distance, whereas anode spot type 2 is formed at a larger gap distance. As a special anode discharge mode, an anode plume usually occurs between anode spot type 2 and arc current zero in various shapes at different locations on the anode surface. The anode plume of the CuCr40 contact appears at a larger gap distance and the corresponding threshold current and accumulated arc energy are also larger. Due to a stronger anode activity, the anode plume lasts a longer time under the CuCr40 contact than under the CuCr30 contact.

Modeling and numerical simulation is an important tool for the research of the physical mechanisms of the vacuum arc (VA). Compared with the experiments of the VA, numerical simulation is more efficient and economical. Axial magnetic fields (AMFs) are also the most popular control technology in vacuum interrupters and other application fields. First, the model of VA was reviewed from a relatively simple zero-order one to 2-D one, and then a 3-D one, which gives us a more comprehensive understanding of VA. Second, the influence of spatial distribution of magnetic fields on VA behavior is also studied. VA behavior under different commercial AMF contacts was simulated and compared based on 3-D magnetohydro-dynamic model. Influence of the radial slots and magnetic field generated by adjacent phase and bus bar on arc characteristics has been reviewed. Third, the VA modeling with active anode is reviewed and discussed. Anode vapor can cool arc plasma, and leads to changed distribution of current density and plasma parameters. Model and simulation technology of anode activity have also been reviewed in this paper, from 1-D, 2-D, to 3-D, and then the deformation of anode melting pool is considered. Deformation of anode melting pool will make anode temperature more reasonable. Finally, the challenge and research direction of VA modeling is also given and discussed.

The Robson drift phenomenon is a macroscopic reflection of the movement of a cathode spot (CS) in oblique magnetic field (OMF). Although it was discovered many years ago, its physical mechanism remains unclear. In this paper, the influence of OMF on the Robson drift of CS motion is simulated. A 2-D stepwise model of the movement of a single CS in OMF is established. It is assumed that the probability of generating a new spot in directions of Amperian force, anti-Amperian force, and directions perpendicular to Amperian force, is proportional to the magnetic pressure around the old spot. The drift angle and velocity of the CS in OMF are simulated. The simulation results show that the drift angle of the CS in OMF has liner relationship with the inclination angle of magnetic field, and it has no obvious relationship with the strength of the applied magnetic field and the amplitude of the arc current, which are in a good agreement with the experimental results. Besides, under the same angle of magnetic field, the velocity of spot motion increases linearly with the strength of OMF. Moreover, the simulation results show that the transverse magnetic field component has a much more predominant influence on the CS velocity than the axial magnetic field component, which are also in agreement with the experimental results.

A quench protection switch (QPS) is indispensable to protect the magnet coils from the damage of quench in a superconducting Tokamak. In this paper, a QPS based on artificial current zero is involved. The vacuum circuit breaker (VCB), which is driven by a high-speed electromagnetic repulsion mechanism, is used as the main circuit breaker (MCB). Two kinds of commercial vacuum interrupters (VIs), which have contacts generating axial magnetic field (AMF) and transverse magnetic field (TMF), respectively, are applied. The breaking current with an amplitude of 10~25 kA is generated by an LC oscillating circuit. The countercurrent with frequency in the range of 500–3000 Hz is provided by a commutation branch. The interruption performance of the two kinds of VIs under different breaking current and countercurrent frequencies is investigated. The experimental results indicate that the differences in interruption performance under the two types of contacts are increasing with larger frequencies of countercurrents.

Vacuum interrupter is widely used in the medium-voltage power system. Controlling the arc by the magnetic field is an effective method to improve the breaking capacity. The transverse magnetic field (TMF) forces the constricted arc to rotate to reduce local overheating. In this paper, three kinds of TMF contacts, which are Swastika-type contact, cup-shaped contact, and double-TMF contact that combines Swastika-type and cup-shaped contacts coaxially, are investigated in a demountable vacuum chamber with a high-speed camera. The arc voltage and current are also recorded. The experiments were conducted with circuit currents from 5 to 20 kA (root mean square) with a frequency of 50 Hz. There are four current levels in the experiment, which are 5, 10, 15, and 20 kA, and at least three tests are taken for each current level. The results showed that under low current (I = 10 kA), the arc of the three TMF contacts did not rotate. Under high current (I = 20 kA), it was observed that the arc of Swastika-type and double-TMF contacts rotated in the gap. The arc speed of double-TMF contacts is higher than that of Swastika-type contacts. The cup-shaped contacts have the highest arc voltage and noise among the three TMF contacts. The Swastika-type contacts were eroded most seriously. The simulation results show the force distribution of the arc on different locations. The resultant force makes the arc rotate counterclockwise.

Using magnetic fields to control the vacuum arc is a major way to improve the breaking capacity of vacuum interrupters. Besides the single transverse magnetic field (TMF) and single axial magnetic field (AMF), new type of TMF-AMF contact and TMF-TMF contact structures have been put forward to control the vacuum arc in recent years. In this paper, three different structural contacts are used in the experiments. The arc evolution was captured by an high-speed digital video camera, and at the same time, the arc current and arc voltage wave were recorded. The experiment was conducted under current ranging from 5 to 20 kA (root mean square) and 10 mm of gap distance. There are 12 current shots performed on one contact design. There are four current levels in the experiment which are 5, 10, 15, and 20 kA, and three tests are taken for each current level. The results showed that for TMF-AMF contacts, under low current (I = 5 or 10 kA), the arcs of the two kinds of contacts both kept diffused and rotating. Under high current (I = 20 kA), the arc of the TMF-AMF contact with six slots was constricted and rotated on the plate, and the melting was uniform and slight. As for the eight-slot contacts, the arc was constricted but did not rotate which resulted in significant melting in the local region. As for TMF-TMF contacts, the result is similar under low current, while the arc was constricted and rotated quickly on the inner plate under high current. The simulation results show that when the arc core is located on the outer plate; the total Lorentz force is near zero. When the arc core is located on the inner plate, the direction of total force is counterclockwise.

In order to improve the breaking capacity of contacts, it is very important to find a better contact material. In this paper, the triggering experiments of plate contact with different materials are studied to investigate the characteristics of vacuum arc. The current range is from 1 to 9 kA (rms). The gap distance is selected as 3 and 5 mm. The arc appearance and the behavior of cathode spots are captured by high-speed camera with an exposure time of $2~\mu \text{s}$ . The surface topography of contacts after experiments is evaluated by scanning electron microscope. The arc current and arc voltage are measured at the same time by Rogowski coil and high-voltage probe. The diameter of the contact is 25 mm. There are seven kinds of materials (CuCr10, CuCr25, CuCr30, CuCr40, WCu10, WCu30, and CuW–WC) conducted in this paper. Compared with the experimental results, the arc of WCu alloy distributes more evenly on the contact plane with many bright cathode spots. The small gap distance makes the burning point more fixed. The arc of CuW–WC is unstable. During the arc trigger moment to the arc stable burning stage, the radial diffusion velocity of WCu10 cathode spot is the lowest with about 25% of CuCr10. CuCr10 alloy has the worst melting resistance. WCu alloy has good melting resistance and bad toughness. The toughness of CuW–WC is even worse. Under the low current, CuCr30 might be preferred for vacuum interrupters from the cathode spots’ radial diffusion speed, arc appearance, and erosion result. When the current increasing, the arc can easily spread to the entire surface of WCu contact, and WCu10 would be the prefer material with the diffusion arc and good erosion resistance.

The quench protection switch (QPS) is an indispensable component ensuring the safety of the magnet coils of a superconducting Tokamak when a quench happens. In this paper, the second commutation process of a QPS based on artificial current zero is investigated. In this process, the current, which has already transferred from the bypass switch to the main circuit breaker, is forced to commutate from the vacuum circuit breaker (VCB) to the discharge resistor by the countercurrent. An LC oscillating circuit is applied to generate oscillating current to simulate the direct current near its peak which is about 20 kA. The countercurrent with frequencies of 500 and 1000 Hz is provided by a commutation branch. The equivalence of the experimental interrupting process compared with the practical situation is analyzed. Two kinds of vacuum interrupters of the VCB are used in experiments. One kind is equipped with contacts generating transverse magnetic field, and axial magnetic field contacts are used in another one. The experimental results indicate that the initial process of the vacuum arc before injecting the countercurrent has crucial impacts on the interruption performance. The countercurrent with the frequency of 500 Hz is easier to interrupt the current than the case of 1000 Hz.

Based on the commercial ANSYS workbench simulation software, the fluid–structure interaction method is used to calculate the air arc pressure distribution in arc chamber, and its impact effect on the shell during interrupting the high current is simulated and analyzed. The stress and deformation distributions of the arc chamber shell with different materials and thicknesses are obtained. It is confirmed that the maximum stress of the shell with 6-mm thickness decreases by ~25% compared with that of 3-mm thickness shell, and the maximum deformation decreases by ~20%. Four models for thickening different local regions are proposed to optimize the shell structure based on the above analysis, including thickened middle regions of the wall inside/outside and corresponding edge regions of the wall. The simulations results indicate that the structure of the arc chamber shell could be optimized very well to achieve the most uniform stress/deformation distribution using the model with thickening the middle region of the outside wall.

In real power system, when the three-phase short-circuit fault occurs, the vacuum arc in one phase will be influenced by the transverse magnetic field generated by neighbor phases and bus bars. This kind of effect is the main cause of the unstable arc and deflected erosion of contact plates, which leads to the failure of vacuum circuit breakers interruption. The objective of this paper is to get more insight into the influence of external transverse magnetic field (ETMF) on vacuum arc’s behavior. The experiments were conducted in a demountable vacuum chamber with pressure about $10^{-4}$ Pa. The cup-type axial magnetic field contacts were used, whose material was pure copper and the diameter is 35 mm. The uniform ETMF in the arc region was generated by two parallel bulk permanent magnets. The experiments were conducted under different ETMFs (0, 15, and 25 mT) and current levels (1-, 2.5-, and 4-kA rms) with different gap distances (6, 8.5, and 11 mm). The videos of arc column were recorded by a high-speed charge-coupled device camera. Under the action of ETMF, the vacuum arc got deflected. Due to the retrograde motion of cathode spots and Ampere force acting on arc column, the deflection behaviors during three typical periods (that is, initial, peak value, and close-to-current-zero period) were different. Moreover, the deflection level of vacuum arc at current peak value moment was greatly impacted by ETMF and gap distance. Larger the ETMF and gap distance were, higher the deflection level at current peak value moment was. The simulation results of arc deflection based on magnetic-hydrodynamic model were in agreement with experimental results in trends.

The motion of cathode spots plays a crucial important role in removing oxide layer on a metal surface by vacuum arc. In this paper, the characteristics of the motion of a single cathode spot on metal surface with oxide layer are investigated experimentally. Experiments are conducted in a detachable vacuum chamber. A hollow copper anode with a hole of $10~\text {mm} \times 10$ mm is used in order to observe the 2-D motion of the cathode spot. The motion of the cathode spot during the descaling process is photographed by a high-speed digital camera with an exposure time of $2~\mu \text{s}$ . Experimental results indicate that there are slow motion and fast motion of the cathode spot during descaling process, and the slow motion is the basic characteristic in most of the descaling time. The probability distribution of the cathode spot’s displacement and resident time, the average velocity, and the movement parameter $S^{2}/t$ are analyzed quantitatively. The results indicate that with the increase of gap distance or the decrease of oxide layer thickness, the cathode spot becomes more active.

Anode material seriously influences the characteristics of vacuum arc and further affects the performance of medium-voltage vacuum switches when the interruption current is high. There are many materials used for electrode manufacture, and different materials are selected for different switches. For a pure metal, its performance usually cannot satisfy the actual requirement. To improve switch’s performance, an alloy is usually used as an electrode material. In this paper, thermal processes of six kinds of metal anodes (including pure metal and alloy anodes) are simulated and researched. The physical parameters of the pure metals all come from experiment results directly or are fitted by the experimental data. The physical parameters of the CuCr alloys are derived from Cu and Cr parameters. Two kinds of temperature calculation methods are used, which are called melting and solidification model and equivalent specific heat method, respectively. Simulation results show that W and Mo anodes have the higher temperature than Cu, Cr, CuCr25, and CuCr50 anodes. A pure Cr anode has the largest melting width and highest saturated vapor pressure and evaporation energy. A Cu anode has the biggest melting depth. A W anode has the smallest melting width and depth. Axial temperature gradient is related to the thermal conductivity, and the Cr anode has the largest axial temperature gradient. The thermal characteristics of CuCr25 and CuCr50 anodes are located between the pure Cu and Cr anodes. There are two melting points appearing in the results of CuCr alloys, and between the two melting points, the alloy anodes are in solid–liquid mixture state.

Based on a steady-state 3-D magnetohydrodynamic (MHD) model, the high-current vacuum arc (HCVA) under combined effect of actual magnetic field (MF) and external transverse MF (ETMF) is simulated. The actual MF is generated by cup-type axial magnetic field contact system commonly used in commercial vacuum circuit breakers. The ETMF may cause the deflection of arc column, which is the main reason of the contact deflected erosion. According to some experimental results, the electron temperature in HCVA is assumed to be uniform and equal to 3 eV. Therefore, the MHD model is simplified by neglecting the electron energy equation to improve the simulation efficiency. With the three conservation equations (mass, momentum, and energy) of ion flow coupling solved, the spatial distributions of some flow parameters can be obtained. The influence of all three components of the MF is inserted by solving the magnetic transport equations sequentially. Proper boundary conditions are set on the cathode and anode side, which separated the cathode spots mixing region and anode sheath region from computation domain, respectively. Under the influence of the ETMF, the deflection of the plasma flow can be predicted, which may be helpful to understand the mechanism of the contact deflected erosion.

In this paper, the influence of an external transverse magnetic field, in parallel with the cathode surface, on the ignition probabilities of a cathode spot (CS) in the Amperian and retrograde directions is developed assuming that the ignition probability of a new CS is proportional to the magnetic pressure around the old one. Furthermore, both stepwise and statistical models are established to simulate the random and retrograde motions of a single CS of vacuum arc, assuming that the 2-D motion of the CS in the $xy$ plane is equivalent to the synthesis of two independent 1-D movements along the $x$ - and $y$ -axis, respectively. With both models, the random and retrograde motions of a single Cu CS are simulated step by step and statistically, respectively. It was found that stepwise simulation results agree well with the statistical results, and they are also consistent with the relevant experimental results. The difference in retrograde motion velocity between the present stepwise model and that assuming that a new CS can be ignited in any direction around the old one is discussed.

In this paper, high-current vacuum arc (VA) breaking experiments in vacuum interrupters were conducted using two kinds of high-capacity (larger diameter) axial magnetic field (AMF) electrodes with contact gap distance changing from 10 to 40 mm. Triggering arc mode and fixed gap distance were adopted. Experimental results showed that during the initial diffusion stage, the arc voltage of high-current VA got a peak value. The change in high-current VA appearance would also affect the curve of arc voltage. With the increase of contact gap distance, the voltage of high-current VA increased and the VA near the anode appeared contracted, and its shape was changed from cylinder to cone. In the shorter gap distance, for cup electrodes, the plasma jets mix with each other and the arc column appears like a bright cone, whereas for coil electrodes, the plasma jets are clear and the column appears like a cylinder, and the transfer of arc appearance affects arc voltage distribution. Moreover, arc twist and rotation often occur under longer gap and higher current, which is due to the larger radial magnetic field component generated by AMF coils, and arc voltage appeared high-frequency and high-amplitude noise, and ultimately resulted in the distortion of the arc current in the current experimental conditions (lower charging voltage of capacitor).

In this paper, the relationship between the external transverse magnetic field (TMF) and the ignition probability of new cathode spot in different directions is developed assuming that the ignition probability of a new cathode spot is proportional to the magnetic pressure around the old one. Furthermore, a method is established to simulate the 2-D motion of a cathode spot step-by-step. With this approach, stepwise simulations of the 2-D retrograde motion of a single Cu cathode spot are carried out in a large range of flux density of TMF ( ${B}_{\rm {em}}$ ). Simulation results indicate that, if the self-generated magnetic field ${B}_{\rm {sm}}$ exceeds the applied magnetic field ${B}_{\rm {em}}$ , i.e., ${B}_{\textrm {em}}\le {B}_{\rm {sm}}$ , there is a linear relationship between the velocity of the retrograde motion and ${B}_{\rm {em}}$ , in certain range of ${B}_{\rm {em}}$ , after which the velocity of the retrograde motion increases slowly and nonlinearly with further increase of ${B}_{\textrm {em}}$ . This phenomenon agrees well with relevant experiment results. In the linear stage, the dependence of velocity constant on the spatial step and the temporal step of the motion of cathode spot is investigated as well. Moreover, when ${B}_{\rm {em}}>{B}_{\rm {sm}}$ , the velocity of cathode spot decreases with the increase of ${B}_{\textrm {em}}$ , even the macroscopic motion is still in the retrograde direction. This effect might also contribute to the reversal of cathode spot motion from the retrograde direction to the Amperian direction at strong TMF besides the predominant effect of the deflection of arc column toward the Amperian direction.

Trigatrons have been widely used in many fields. However, the performance of trigatrons typically degrades as operational time increases. Among many influence factors, electrode erosion is a primary cause of trigatrons failure. In this paper, the peak current and pulsewidth of trigatrons are in the order of kiloampere (kA) and microsecond $(\mu{\rm s})$ , respectively, with an electrode size in the order of millimeters (mm). The discharge images show that the arc appearance looks filamentary in shape during the early period of the discharge. As the discharge process progresses, the arc expands and moves radially to the side surface of the anode. Simultaneously, electrode erosion experiments are conducted on the trigatrons in an atmospheric environment. Experimental results show that the erosion level becomes more serious with an increase of gap distance and discharge current. The weight loss of Mo and W electrodes with erosion times are very close to a linear distribution, however, weight loss rate of W-Cu electrodes increases more dramatically with an increase of erosion times. The erosion of WCu electrode is more pronounced than Mo and W electrodes the volume erosion rate of W is smaller than Mo, but the mass erosion rate of W is higher than Mo. Many cracks always appear on the W and Mo electrode surfaces, which are due to frequent and intensive interaction between the expansion and shrinkage of the electrode material.

The expansion speed of cathode spots (CSs) in the initial expansion stage of high-current triggered vacuum arc was investigated experimentally. Experiments were conducted with butt contacts in a demountable vacuum chamber. Images of CSs were photographed with a high-speed digital camera with exposure time of 2 μs. The uniform constant axial magnetic field (AMF) was supplied by an external magnetic field coil. The combined influence of contact gap of 2-10 mm and AMF of 0-70 mT on CSs expansion speed in the initial expansion stage was mainly investigated. Investigation results indicated that the variation of contact gap could influence the CSs expansion speed by changing the plasma density inside the cathode arc root in two ways: loss of arc plasma and constriction of arc column. AMF could weaken these two effects caused by the variation of contact gap. In addition, experimental results indicated that there existed a critical current for a given AMF. If the arc current peak was less than the critical current, the given AMF could inhibit the arc column constriction effectively both at contact gap of 8 and 10 mm, which could maintain the dependence that CSs expansion speed increased with contact gap; otherwise, the constriction of arc column played a major role in the influence on CSs expansion speed. In this case, the CSs expansion speed was lower at contact gap of 10 mm than that at 8 mm.

This paper analyzes how removing oxide layer on metal surface by vacuum arc could be a green substitute for the traditional acid-washing and mechanical cleaning technique widely used in the metallurgical industry, which is harmful to the environment. In this paper, the effects of vacuum arc on removing oxide layer on the surface of carbon-steel tubes, which were used as the cathode, is investigated experimentally. External coils are used to generate transverse magnetic field (TMF) driving cathode spots to rotate on tube surface. Experiments are conducted in a vacuum chamber with Ar as the protection gas under different TMF flux densities. Arc images are captured by a high-speed digital camera with an exposure time of 2 $\mu{\rm s}$ . The influence of transverse magnetic flux density on the characteristics of arc and descaling effect is mainly investigated.

With the increase of interruption frequency of high-current vacuum arc (HCVA), the relaxation time of arc plasma is close to characteristic time of arc current and transient characteristics of vacuum arc plasma become more obvious. In this paper, based on a 2-D transient magneto-hydrodynamic model, HCVA under different interruption processes is simulated and analyzed. The simulation results of power-frequency (50 Hz) current show that transient simulation results are the same as those of the steady-state model, as characteristic time of current is much longer than relaxation time of plasma. For fast current drop stage in dc interruption process, simulation results show that higher frequency makes current drops faster and ion number density larger; besides, residual plasma density between electrodes at current-zero moment is higher, and interruption becomes more difficult. Finally, transient characteristics of high-frequency HCVA is also simulated and analyzed. Simulation results show that change curve of ion number density lagged behind current curve; higher ion number density in the second 1/4 cycle can lead to reignition more easily.

In this paper, the influence of an axial magnetic field (AMF) on the expansion speed of cathode spots (CSs) in the initial expansion stage of high-current triggered vacuum arc is investigated experimentally in a demountable vacuum chamber. Butt contacts with contact plate material of oxygen-free high-thermal-conductivity Cu and CuCr25 were used in experiments. Images of CSs are photographed with a high-speed digital camera with an exposure time of 2 $\mu{\rm s}$ . The uniform constant AMF in the intercontact region is supplied by external magnetic field coil. Experimental results indicate that, in an arc stabilized by AMF, CS expansion speed also increases with the increase of $di/dt$ , which is the same as that in arc burning freely. In addition, quasi-quantitative analysis of the experimental results show that the bending and lengthening of CS motion trajectory in AMF are not sufficient to explain the decrease of CS expansion speed with AMF. Another reason might be related to the inhibitory effect of AMF on the loss of arc plasma, which could also contribute to the decrease of CS expansion speed.

With a quasi-steady 3-D MHD model, time-dependent axial magnetic field vacuum arcs (AMFVAs) characteristics in commercial vacuum interrupters for half-wave interruption of power-frequency (50 Hz) current are simulated. In a current time-dependent simulation, the influence of AMF lagged from arc current is considered in the arc model. 3-D spatial distributions of many important plasma parameters and electrical characteristics in AMFVAs can be obtained, such as ion number density, ion temperature, electron temperature, plasma pressure, current densities along different directions ( $x$ , $y$ , and $z$ ), electric fields strength along different directions, and so on. Simulation results show that there exists a significant spiral-shaped rotational phenomena in the AMFVAs and the rotational velocity at the moments (9, 8, 7, and 6 ms) after 5 ms (the current peak value moment) is always larger than that before 5 ms (1, 2, 3, and 4 ms) because of the lagged AMF. This kind of plasma rotational phenomena also can be observed by high-speed arc photographs. In the future, the asymmetric appearance in high-current vacuum arc will be simulated based on a 3-D model.

In this paper, a transient 3-D model of the anode activity in a high-current vacuum arc (HCVA) was established. The melting, flow, evaporation, and solidification of the anode material were included in this model. Based on this model, the thermal and flow process of the anode in an HCVA under axial magnetic fields was simulated and analyzed. Simulation results showed that the maximum anode temperature appeared near 7 ms. The maximum rotational velocity and melting radius appeared near 9 ms. This meant that the anode was still in the melting, flow, and evaporation status near current zero moment, which was easier to cause reignition of vacuum interruption. Through 3-D modeling and simulation, a more visualized anode process can be understood. In the future, unsymmetrical anode phenomena will be researched and analyzed by this 3-D model.

In a drawn vacuum arc, the arc sequence begins with the bridge column arc formed after the rupture of the molten metal bridge. This column evolves into the transition mode, which consists of a central column with few or no cathode spots outside the column, and then develops into the fully diffuse mode. In this paper, the influence of arc ignition position on arc characteristics in the transition mode of a drawn vacuum arc was investigated experimentally in a demountable vacuum chamber. Practical cup-shaped axial magnetic field (AMF) contacts were used in the experiments. Arc characteristics were investigated with the aid of a high-speed digital camera with an exposure time of 2 μs. Experimental results indicated that the characteristics of the arc and arc voltage, particularly the peak in the arc voltage waveform, in the transition mode of a drawn vacuum arc in cup-shaped AMF contacts were closely related to the arc ignition position.

The initial expansion process in a drawn vacuum arc and the influence of axial magnetic field (AMF) were investigated experimentally in a demountable vacuum chamber. Arc characteristics were investigated with the aid of a high-speed digital camera with an exposure time of 2 μs . In a drawn vacuum arc, the arc sequence begins with the bridge column arc formed after the rupture of the molten metal bridge. This column evolves into the transition mode, which consisted of a central column with few or no cathode spots (CSs) outside the column, and then into the fully diffuse mode. Experimental results indicated that in transition mode arc, the expansion process could be characterized by the appearance of CSs outside the central column, and could be classified into two patterns, “slow” expansion and “quick” expansion according to the characteristics of the formation and motion of new CSs (conducting channels) outside the central column of arc. The influence of AMF and its distribution on the expansion process was also investigated. Investigation results indicated that AMF had two contrary effects, i.e., inhibiting effect and prompting effect, on the initial expansion stage of drawn vacuum arc. Furthermore, saddle-shaped AMF could encourage the arc transition into diffuse mode more effectively than bell-shaped AMF.

The method of magnetic field calculation and its influence on simulating the motion of cathode spots (CSs) in a vacuum arc were investigated in this paper. A method based on coordinate transformation and magnetic field superposition principle was proposed. In this method, the influence of current passing through an electrode rod and a contact plate was taken into account. Comparison was made between the proposed method and two simplified methods which were on the basis of the Biot-Savart law. Moreover, simulation results of CS motion with magnetic field calculated by different methods were shown and compared. Investigation results indicated that current passing through the contact plate and electrode rod could significantly influence the magnetic field and thus influence the simulation results of CS motion. Furthermore, it was shown that the proposed magnetic field calculation method had higher calculation accuracy with acceptable consumption of calculation time.

Removing an oxide layer on a metal surface by vacuum arc has been preliminarily used in the metallurgical industry as an environment-friendly substitutor for the traditional acid-washing technique. In this paper, the influence of protection gas pressure on descaling process is investigated experimentally. Arc images taken by a high-speed digital camera indicated that the descaling rate was closely related to the motion pattern of cathode spots in different gas pressures. The decrease of descaling rate with the increase of gas pressure could be partly attributed to the formation of cathode-spot group and its less mobility at high gas pressure. The influence of gas pressure on the power density of cathode spot was also analyzed. It seemed that higher gas pressure would lead to the decrease of the ratio of power input into the cathode spot to the total arc power, which could also contribute to the decrease of the descaling rate.

Experiments on a drawn vacuum arc with forced current zero, which was formed by injection of a countercurrent with a frequency of 1 kHz, were conducted in a detach able vacuum chamber. The prospective peak main current was 3.5 kA. Arc characteristics were investigated with the aid of a high-speed digital camera with an exposure time of 2 μs. In this paper, the influence of axial magnetic field (AMF) distribution and anode material on the evolution of vacuum arc in the initial expanding stage is mainly investigated. Three types of electrodes were used in the experiments. One of them was simple electrodes with butt contact plates, and the other two types were specially designed electrodes generating a conventional bell-shaped AMF profile and a saddle-shaped AMF distribution, respectively. OFHC copper and WCu10 were used as anodes for the electrodes with bell-shaped AMF. The appearance of vacuum arc in the forced current-zero stage is also investigated.

Much research has been done on the motion of low-current vacuum-arc cathode spots (CSs). The relationship between the motion velocity of CSs and the magnetic field, particularly the relationship between the direction of CS motion and the ratio of Bz/Bt, has been revealed. However, little investigation on the motion of CSs in high-current vacuum arcs has been conducted before. In this paper, the motion of CSs of high-current vacuum arcs under different axial magnetic fields was experimentally investigated. Simultaneously, calculation of the motion of CSs based on the empirical formula has been done, and comparison with experiments and analysis were given. Experiments were conducted in a detachable vacuum chamber with an arc current of 10 kA (root mean square). Images of CSs were photographed with a high-speed digital camera with an exposure time of 2 μs. With computer-aided numerical processing, the spatial distribution of the CSs was obtained, based on which the magnitude and direction of the CS motion velocity were calculated, and the CS motion trajectories around the current peak were plotted. Results indicated that the properties of the CS motion in high-current vacuum arcs are similar to those in low-current vacuum arcs.

The theoretical and experimental studies of cathode plasma jets in vacuum arc and the effect of magnetic field on them have been under way for several years. In this paper, different axial magnetic field electrodes were tested in a vacuum chamber, obvious individual cathode plasma jets were observed, and the cathode-jet inclination was also detected. Based on the numerical calculation of magnetic field in interelectrode region and the measurement of cathode-jet inclination in experimental results, it is proven that the inclination angle of composite magnetic field (the combination of the axial, the azimuthal, and the radial components) is consistent with the inclination angle of cathode plasma jets, which indicates that the cathode plasma jets in the arc column flow along the magnetic-field direction when the high-current vacuum arc is diffused or even constricted slightly. Meanwhile, it was also found that plasma jets far from the cathode surface mixed with each other and became ambiguous when the electrode gap increases to a certain distance.

In this paper, high current vacuum arc (HCVA) characteristics and anode melting pool (AMP) flow phenomena under different conditions are researched and analyzed. Particularly, rotation phenomena of AMP are observed and analyzed. First, the influence of different arc currents on arc column and anode activities is researched. Experimental results show that the rotation phenomena of AMP become more significant when arc currents are very high. Then, the influence of electrode configurations on the AMP rotation is investigated. For larger h/D value electrode, vacuum arc can become more unstable and uncontrollable, the rotation of anode melting region cannot be found; for smaller h/D, the vacuum arc is more stable and the vacuum arc is controlled effectively by the axial magnetic field (AMF), also significant rotation of AMP has been found. For a smaller anode electrode, anode melting and rotation of AMP is more serious, which is due to the same energy inputs smaller electrode. Finally, the influence of pure Cu and CuCr30 electrode on anode melting and flow is compared and analyzed. Experimental results show that AMP rotation with pure Cu material is more significant than that with CuCr30 material for the same arc current interruption. Liquid macroparticles' diameters with pure Cu electrode are significantly larger than that with CuCr30 materials.

It is well known that the melting of electrodes (mainly anode melting) in vacuum arc can increase the metal vapor density around current zero and even lead to interruption failure. In order to clarify the anode activities and their influence on arc appearance and interruption capacity, series experiments of cup-shaped axial magnetic field copper electrodes were conducted. Obvious anode melting was detected; the liquid copper flowed on the contact plate of anode and formed a clockwise swirl flow. The appearance of anode melting is likely to correlate to the transition of arc mode from high-current diffuse mode to high-current diffuse column mode. The melting of anode was severer than cathode and was influenced by the distribution of cathode spots. Various kinds of copper particles at macroscopic level can be seen in arc column. Even at the interruption limit, the majority of melted copper of anode sputtered out of gap in form of liquid droplets or was pressed into the cup of anode, the copper vapor evaporated into arc column only accounted for a few portion and no obvious anode jets was found due to large plasma pressure in arc column.

The influence of the distribution of axial magnetic fields (AMFs) was investigated experimentally with three pairs of specially designed testing electrodes generating a conventional bell-shaped AMF profile and different saddle-shaped AMF distributions. The characteristics of high-current vacuum arcs (HCVAs) up to 20 kA (rms), e.g., the shape of the arc column and the distribution of cathode spots, were investigated with the aid of a high-speed digital camera with an exposure time of 2 mus. Experimental results further manifested the distinct influence of AMF distribution on the characteristics of HCVAs. Saddle-shaped AMFs could resist the constriction of HCVAs more efficiently than traditional bell-shaped AMFs. Furthermore, cathode spots in saddle-shaped AMFs distributed more uniformly than that in bell-shaped AMFs. It was found that AMFs in the central region of saddle-shaped AMFs should not be too weak. Otherwise, cathode spots would be out of control in the central region as observed in experiments. The ldquooptimalrdquo minimal AMF in the central region was estimated to be about 3-4 mT/kA.

In this paper, the initial expansion process after trigger of vacuum arc was investigated experimentally with two typical axial-magnetic-field (AMF) distributions, i.e., bell-shaped AMF generated by traditional cup-shaped AMF electrodes and saddle-shaped AMF generated by specially designed coil-type AMF electrodes. The motion of cathode spots (CSs) in the initial expansion process was investigated by high-speed digital camera with exposure time of 2 mus. Experimental results indicated that CSs expanded faster under saddle-shaped AMF than under bell-shaped AMF at 5 (rms), 10, 15, and 20 kA. Furthermore, the motion of CSs slowed down between 15 and 20 kA in the case of bell-shaped AMF, whereas CSs tended to quicken up with the increase of arc current under saddle-shaped AMF. Based on the images of CSs, it was proposed that, besides direct influence of AMF on the retrograde motion of CSs, the concentration of CSs at relatively high current and strong central AMF could also inhibit the outward motion in the initial expansion process.

Experiments were conducted in a detachable vacuum chamber with conventional cup-shaped axial magnetic field (AMF) electrodes and three types of AMF electrodes with different nonuniform AMFs. The radial distributions of the AMF were varied by introducing ferromagnetic material into the cup-shaped AMF electrodes. The vacuum arc column and the distribution of cathode spots were photographed with a high-speed charge coupled device (CCD) camera. Experimental results indicate that a nonuniform AMF with higher axial magnetic flux density at the contact periphery than that at the center (saddle-shaped AMF) can resist arc constriction and increase the cathode and anode utilization ratios more effectively than the ordinary bell-shaped AMF profile.

Trace of oxygen adding to the helium atmospheric pressure plasma jet can make the jet behave differently, such as dimming the plume and changing the jet length as well as the discharge current. Moreover, with certain specific ratio of oxygen/helium, double-bullet phenomenon could be captured by intensified charge couple device. The speeds of the bullets are estimated to be several kilometers per second.

Based on a 3-D steady magnetohydrodynamic model, the high current vacuum arc subjected to a realistic spatial magnetic field generated by the cup-type contacts was simulated. The influence of the realistic spatial magnetic field profile on plasma's rotation behavior was analyzed. The distribution of the azimuthal velocity on anode side under realistic spatial magnetic field profile was quite different from that under uniform axial magnetic field.