Photographic appearance of high-current vacuum arcs prior to and during anode spot formation

This article presents the results of research on the photographic appearance of a highcurrent vacuum arc between butt type copper electrodes a of 30–80 mm diameter and a fixed gap of 10 mm. Current pulses of up to 30 kA peak amplitude at an initial value of (di/dt)₀ from 1–10kA/ms and a duration of approximately 14 ms were applied. Arcs were photographed with a high-speed framing camera, mostly at 10⁴ frames/s. A detailed study of discharge modes in phase transition from a high-current diffuse arc to a constricted arc with an anode spot was conducted. Most of the measurements were obtained at a peak current slightly in excess of 10 kA for electrodes of 55 mm diameter. It was found that at peak current exceeding moderately the threshold value of the onset of anode spot formation, the arc is characterized by the following main features: the formation of an anode spot and an anode plasma jet occurs concurrently with a local concentration of cathode spots; the anode spot is, most often, formed on the electrode edge; the coexistence of very varied structures of spots on the cathode; the lack of considerable constriction of the cathode discharge; the pseudo-periodic shrinking and expansion of the area occupied by cathode spots; the existence of a relatively dark space separates the anode plasma jet from the plasma sheath near the cathode surface; the plasma space distribution in the interelectrode gap is non-uniform and non-stationary.This work was supported by State Committee for Scientific Research within the research project No. 3 P40101507.     

The Time-Resolved Characterization of Erosion Products from High-Current, Copper Vacuum Arcs

A vacuum arc at high enough current can produce gross melting on electrode surfaces as a consequence of anode spot formation and other high-current electrode phenomena. Erosion from the electrodes under this condition is much more rapid than at low-current (where material loss occurs principally from the cathode) and is a process that is presently poorly understood. The present work is aimed at characterizing the erosion products from cathode and anode surfaces during high-current arcs on copper electrodes for single half cycles (60 Hz) arcs having peak currents of 30 kA. Fully open gap lengths were approximately 18 mm. Among the findings were the following. a) Erosion rate determined by electrode weight loss was approximately 8 mg/C of arcing. b) Droplets ejected from the electrodes had masses varying from a few tenths to a few tens of micrograms and velocities typically up to 40 m/s, although higher velocities are seen. c) The greatest number of droplets are produced at, or just after the current peak, and higher droplet velocities are seen in this same time interval. d) Erosion in vapor form detected in the plane of the cathode surface and moving radially is a maximum just after the peak of current and is relatively abundant. Such vapor is essentially absent in the anode plane.     

Observation of tungsten particles in vacuum arcs by laser inducedfluorescence

We have clarified the relation between the decay of tungsten ion density in the vicinity of current zero and vacuum arc mode in high current period by using a laser induced fluorescence method in tungsten vacuum arcs of 60 Hz sinusoidal current with the peak value of 3.3, 6.7, and 9.8 kA. In the case of 6.7 kA, the arc mode was the anode spot mode. Because of the generation of the anode spot, the tungsten ion density near the anode was higher than near the cathode and the density near the anode was about ten times as high as the case of 3.3 kA which was the diffuse mode. In the case of 9.8 kA, which was the intense arc mode, the density near the anode was not significantly different from the case of 6.7 kA. The density near the cathode was higher than near the anode and tungsten ions were observed till about 30 μs after current zero while they disappeared at current zero in the other cases     

Arc voltage of a fast triggered vacuum gap

At high current, the performance of triggered vacuum gaps (TVGs) is limited by constriction of the vacuum arc. Several concentrated modes can be defined, i.e., foot point, anode spot, and intense arc mode. In all cases, small, luminous, high-temperature spots (from melting to boiling temperature) appear on the anode surface. In accordance with the anodic mode, the arc voltage has different characteristics (quiet and low or with high-frequency noise). The arc voltage is measured for different electrode configurations for a conduction time of 27 μs and for peak current up to 45 kA. For small gap distances (1-1.5 mm), the arc voltage is quiet and low (20 to 30 V) and almost independent of the peak current. For greater distances, the arc voltage increases with the distance and the peak current. If the peak current is higher than the threshold interruption current, the arc voltage is high and noisy. These overvoltages, with a frequency of about one megahertz, can reach more than 1000 V. These overvoltages disappear completely after about 15 μs, and the voltage decreases     

Pseudospark discharge in long time current pulse: electricalcharacterization and recovery phenomena

At high currents, the performance of triggered vacuum gaps is limited by the constriction of the arc discharge on the anode. In a symmetrical system within flat electrodes, a motionless anode spot appears in front of the trigger spark. A pseudo-spark system within electrode cavities and specific gas pressure provides a good mechanism for avoiding anode spot phenomena. It appears possible to obtain a high intensity diffuse discharge. In this paper, we present the experimental measurements of the instantaneous voltage and discharge current, (peak current less than 50 kA, discharge time between 20 and 300 μs). These measurements allow us to determine the dynamical resistance, mean voltage and energy lost during the conduction phase. Consequently, we refine our knowledge of discharge development. Additionally, we present a complementary investigation on quenching behavior     

Anode Melting in a Multicathode-Spot Vacuum Arc

Melting of the anode surface in a multicathode-spot vacuum arc is expected when the incident energy flux is not balanced. The anodic energy influx is proportional to the arc-current collected by the anode and melting of the anode should be observed when peak arc-current exceeds a critical value. In this work, the critical peak arc-current Ipt was measured, and its dependence on anode and cathode materials was determined. The arc was sustained between two parallel cylindrical electrodes, 14 mm in diameter and spaced 4 mm apart. The almost critically damped current pulse lasted for 30 ms with a 6-ms rise time to peak value. Peak currents were in the range of 500-2300 A. In most of the experiments the anode material differed from that of the cathode. In the runs where the cathode-anode materials were Cu-Al or Mo-Cu, respectively, the time dependence of a spectral line intensity radiated by the anode atoms located in the plasma near the anode surface was recorded. We found that Ipt depended on both the anode and cathode materials. Thus for an Al anode and Al and Cu cathodes, Ipt equaled to 1100 and 900 A, respectively. In arcs with a peak current larger or equal to Ipt, a sudden jump of the spectral line intensity was observed. In all experiments, even when strong melting of the anode was observed, the arc-voltage stayed quiescent and in the range 15-35 V, suggesting that no anode spot was formed.     

Soft X-ray emissions by high-current vacuum discharges

Soft X-ray emissions by Z-pinch discharge between vacuum spark electrodes are examined. Mainly aluminum and molybdenum were utilized to generate soft X-rays, and emissions resulted for pulse currents of about 150 kA. These emissions increased as the discharge current increased. A large X-ray intensity was obtained when a spherical anode and a conical cathode were used. A stable spot plasma was produced for an electrode separation distance of 2 mm. For molybdenum, the X-ray emission of the 0.52-nm line and continuous X-rays of 0.65-1.2 nm were observed. The X-ray continuum was assumed to be a recombination continuum. For aluminum alloy containing magnesium, the 0.62-, 0.66-, 0.72-, 0.78-, and 0.92-nm lines were observed. The experimental results indicate that vacuum spark is a potential exposure source for X-ray lithography     

水中脉冲放电金属电极烧蚀机理

采用板-板电极, 在放电间隙距离为2 mm、放电电流峰值为22 kA条件下, 对黄铜、钨铜电极的烧蚀特性进行了对比研究。利用高精度天平测量放电过程中的电极质量损失, 分别获取了阴极、阳极及总的平均烧蚀速率。通过放电后电极表面微观形貌、微观元素组成的分析及液体中金属离子的含量分析, 对水中脉冲放电金属电极的烧蚀机理进行了探讨。结果表明, 水中脉冲放电时, 钨铜电极的抗烧蚀性能明显高于黄铜电极。黄铜电极的主要烧蚀是以中心的大量孔洞及其边缘的波纹结构为表现形式的液体金属的溅射;钨铜电极的突出物及较平整的表面暗示了气相侵蚀的作用。以电弧的焦耳热效应为催化剂, 钨铜与水的电化学反应更为强烈, 因此电化学腐蚀是水中放电电极烧蚀的形式之一。     

Design and manufacture of planar GaAs Gunn diode for millimeter wave application

GaAs-based planar Gunn diodes with AlGaAs hot electron injector have been successfully developed to be used as a local oscillator of 76 GHz in monolithic millimeter-wave integrated circuits.We designed two kinds of structure diode,one has a fixed distance between the anode and cathode,but has variational cathode area,the other has a fixed cathode area,but has different distances between two electrodes.The fabrication of Gunn diode is performed in accordance with the order of operations:cathode defining,mesa etching,anode defining,isolation,passivation,via hole and electroplating.A peak current density of 29.5 kA/cm 2 is obtained.And the characteristics of negative differential resistance and the asymmetry of the current-voltage curve due to the AlGaAs hot electron injector are discussed in detail.It is demonstrated that the smaller size of active area corresponds to the smaller current,and the shorter distance between anode and cathode also corresponds to the lower threshold voltage and higher peak current,and hot electron injector can effectively enhance the radio frequency conversion efficiency and output power.     

Anode-Spot Formation and Motion of Vacuum Arcs

This paper reports about experimental investigations on high-current vacuum-arc phenomena, especially anode-spot formation, arc states, and motion. The presented work was stimulated by lack of information about the transition process from the diffuse low-current mode to the high-current mode characterized by anode spot(s). Optoelectronic measurements, streak photographs, high-speed movies, and correlated arc voltage/current records yielded remarkable results on power-frequency vacuum arcs. Three different high-current vacuum arc modes can be observed beyond a certain threshold current. Which mode appears depends mainly on the momentary electrode distance. The modes are characterized by different anode-spot behavior and interelectrode phenomena. The transition between different arc modes is continuous. The arc modes observed on ring electrodes producing a magnetic blast field are the same as those appearing on butt-type electrodes. Anode-spot formation is preceded by congregations of cathode spots and may be initiated by thermal overload of the anode surface opposite to these cathode-spot clusters.     

Experimental investigations of the currents collected by metalshields in high-current vacuum arcs

This paper presents experimental results of currents collected on the three-element condensation shield connected to the cathode potential in high-current vacuum arcs. The arc current had 900 Hz, 150 Hz, or 50 Hz half-cycle sinusoidal shapes and was conducted between the CuCr40 contacts with a diameter of 50 mm (cathode) and 30 mm (anode) spaced 10 mm apart. Most of the measurements were made for the current of 900 Hz with peak values up to 9 kA. Arc voltage, floating shield potential, and distribution of shield currents were measured. It was found that the current collected by the shield and also the arc voltage and floating shield potential are greater for higher frequency currents, and that they are affected by the arc mode. Considerable shield current is observed during a high-amplitude (HA) oscillation sequence of arc voltage while its mean value is increased. For 900-Hz arcs at the 9-kA peak value (I_am), the ratio of shield current (i_s) to an instantaneous value of arc current (i_a) reaches even 40% near I_am value. In the initial half-cycle period (before the initiation of high-voltage oscillation), the ratio of i_s/i_a increases with i_a and current frequency. A close relationship was found between arc voltage and current distribution on a three-element shield     

A Review of Anode Phenomena in Vacuum Arcs

This paper discusses are modes at the anode, experimental results pertinent to anode phenomena, and theoretical explanations of anode phenomena. A vacuum are can exhibit five anode discharge modes: (1) a low current mode in which the anode is basically passive, acting only as a collector of particles emitted from the cathode; (2) a second low current mode that can occur if the electrode material is readily sputtered (a flux of sputtered atoms will be emitted by the anode); (3) a footpoint mode, characterized by the appearance of one or more small luminous spots on the anode (footpoints are generally much cooler than the true anode spots present in the last two modes); (4) an anode spot mode in which one large or several small anode spots are present (such spots are very luminous, have a temperature near the atmospheric boiling point of the anode material, and are a copious source of vapor and ions); and (5) an intense are mode where an anode spot is present, but accompanied by severe cathode erosion. The are voltage is relatively low and quiet in the two low current modes and the intense are mode. It is usually high and noisy in the footpoint mode, and it can be either in the anode spot mode. Anode erosion is low, indeed negative, in the two low current modes, and it is low to moderate in the footpoint mode. Severe anode erosion occurs in both the anode spot and intense are modes. The dominant mechanism controlling the formation of an anode spot appears to depend upon the electrode geometry, the electrode material, and the current waveform of the particular vacuum are being considered. In specific experimental conditions, either magnetic constriction in the gap plasma, or gross anode melting, or local anode evaporation can trigger the transition. However, the most probable explanation of anode spot formation is a combination theory, which considers magnetic constriction in the plasma together with the fluxes of material from the anode and cathode as well as the thermal, electrical, and geometric effects of the anode in analyzing the behavior of the anode and the nearby plasma.     

Erosion of tube electrodes in vortex plasmatorches

The stationary modes of the end-face tube electrodes operation are under consideration. The formation of cathode and anode jets which cause dramatic increase in the electrode erosion at the arc current growth has been explained. We have considered the ways to increase the life-time of the continuous operation of tube electrodes. The analysis of the thermophysical interaction of the arc spot and electrode surface has enabled us to choose the optimum metal for the electrode.     

Correlation of current quenching and occurrence of metal vapor in apseudospark discharge

The quenching phenomenon, i.e., a sudden interrupt of the discharge current, was investigated in a pseudospark discharge with charging voltage of 2.5 kV, maximum current of 2 kA and discharge duration of 3 μs. The working gas was hydrogen at a pressure of 40 Pa. Concerning electrode material and geometric parameters, molybdenun electrodes were chosen with hole diameters of 5 mm; the electrode distance was 3 mm. In this parameter range, a temporal correlation of current quenching and the occurrence of metal vapor could be detected by means of time-resolved optical spectroscopy. With each current interruption a sudden increase of emission from neutral molybdenum atoms as well as an increase of cathode spot emission, which is spatially localized on the cathode, occurs. Also oxygen ions were observed which show a similar time-dependence, however with a significant delay of the order of 200 ns. The results are discussed in the scope of the mechanism proposed for quenching, i.e., ion depletion in the plasma boundary layer, and the mechanisms occurring in the high current phase of a pseudospark discharge     

Anode jet in a high-current vacuum arc

A stable intense jet with a clear-cut bright sheath has been detected on the anode of a 10-ms-long high-current vacuum arc with a current amplitude of 15 kA. The jet is adjacent to the hot spot of a molten metal on the anode surface. The primary light of the jet is emitted by neutrals. The sheath of the jet is surrounded by an ion-induced diffuse glow. The anode jet arises from interaction between the cathode and anode plasmas. Because of this, the size of the jet inversely depends on the current of the arc and the jet becomes observable only by the end of the current pulse. This object (anode jet with a bright sheath) is well reproducible when the arc is initiated between copper-chromium electrodes. In the case of pure copper electrodes, such objects occur randomly and appear at long projections of the molten metal, where heat release is hampered, and at large drops moving in the interelectrode gap. This means that the anode evaporation intensity is crucial for the appearance of bright-sheath jets.     

强流碳纳米管阴极重复频率发射特性

采用钛氰铁高温催化热解方法可制备发射性能优异的碳纳米管薄膜阴极。当脉冲电场峰值达到30 MV/m时,发射电流密度达kA/cm2以上,对应相对论电子束流强度高达15 kA,等离子体发射机制参与电子束发射过程。以重复频率10 Hz发射模式时,其发射阈值低,束压、束流波形跟随性好,发射稳定性优于石墨阴极。发射发次达到1000后,碳纳米管形态依然完整,界面无脱附。     

Behavior of a High Current Vacuum Arc between Hollow Cylindrical Electrodes in a Radial Magnetic Field

Experimental observations were conducted on the behavior of a high current vacuum arc on cylindrical electrodes in a radial magnetic field. The arc was sustained between the ends of two cylindrical Cu electrodes, 54-mm diam and 1.5-mm wall thickness separated by 5 mm. Arc current pulses with peak values in the range 4-15 kA with a half amplitude full width (HAFW) duration of 8 ms were investigated with radial magnetic fields proportional to the instantaneous current with proportionality constants of 4.0 and 6.5 × 10-6 T/A. The arcs were photographed simultaneously with a streak camera and by a high speed framing camera and the arc voltage was recorded on a digitizing transient recorder. The results indicated that the arc in this geometry, both with and without an imposed radial magnetic field, can be characterized by three development stages: a) arc formation, b) diffuse arc along the electrode perimeter, and c) simultaneous existence of several concentrated arc columns. When a radial magnetic field was imposed two changes were noted: 1) the arc appeared somewhat more distributed in that a greater number of constricted columns were observed, and they were distributed more evenly; and 2) the constricted columns moved in the J? × B? direction with velocities in the range 5-35 m/s.     

The plasma in high-current pseudospark switches

The discharge behavior and the erosion rate of pseudospark switches for high currents (50-150 kA) and pulse lengths of several microseconds were investigated for different electrode materials. A capacitor discharge (3.3 μF) without any load was used at a maximum voltage of 30 kV. Side-on optical investigations were performed either with a streak camera or a fast shutter camera. Using metal electrodes, the discharge ignites on axis, then widens up radially and burns homogeneously at the edge of the central apertures. After about 500 ns, a stable anode spot is observed on the plane electrode surface (at currents exceeding 45 kA), the location of which is statistical. The discharge is transformed to a metal vapor are discharge and the erosion rate increases by more than one order of magnitude. With semiconductor electrodes (i.e., silicon carbide), a different discharge behavior is observed, After ignition on axis, the discharge burns homogeneously on the whole carbide surface. No contraction to a small area occurs in comparison to metal electrodes. The reignition of later current half cycles starts at the triple point metal-carbide-gas. Then the discharge again spreads homogeneously over the total carbide surface. The erosion rate is about two magnitudes lower in comparison to metals. We assume that the current is conducted in a thin surface sheath which is heated to more than 2000 K     

Cathode spot dynamics on pure metals and composite materials inhigh-current vacuum arcs

An investigation has been carried out of cathode spot dynamics in a triggered vacuum arc in a demountable chamber. A rectangular current pulse of 1-5 kA, 1-5 ms has been used. Sufficient statistics were collected. The expansion of a cathode spot ring on a clean, pure metal surface was corroborated to be a retrograde movement in the self-magnetic field which obeys the same law as the movement of a single spot in an external magnetic field. The influence of a contact gap of 0.5-8 mm and current on the dynamics of cathode spots was investigated. The gap dependence of the proportional coefficient between the spot velocity and magnetic field in the case of a pure copper cathode was obtained. A phenomenon was discovered, where a group of cathode spots form in the short arcs on the CuCr cathodes after a transition diffuse arc stage. The follow-up investigation revealed that a close interrelation exists between the cathode and anode processes in short arcs. This interrelation is responsible for the appearance of the discovered phenomenon. Short-circuit performance tests conducted for a commercial vacuum interrupter proved cathode spot group formation to be responsible for the interruption failure at short contact gaps     

Self-extraction of negatively charged carriers from anodic vacuumarcs

Measurement and magnetic analysis of self-extracted negatively charged carriers (NCCs) from anodic vacuum arcs are presented. They flow to charge collectors made of stainless steel, which are electrically connected to a cathode. When a 60-Hz sinusoidal arc current of 40 kA is burned on 20-mm-diameter copper electrodes spaced 4 mm apart, a negative current of approximately 900 A flows to a cylindrical collector surrounding the arc. The floating collector potential relative to the cathode is measured, and the mean energy of the NCCs is estimated to be greater than 40 eV. It is difficult for the NCCs to flow from anodic vacuum arcs when an anode is made of zinc or cadmium, the atoms of which have electron affinities of less than 0 eV. A magnetic filter of about 500 G, which is placed between the arc and a 30-mm-diameter circular collector, does not affect the NCC's flow from a 4-kA arc burned on copper electrodes. It is possible to extract a large amount of negative copper ions from the anodic copper vacuum arcs