The Columnar Vacuum Arc

In the development of vacuum interrupters for interrupting high short-circuit currents, understanding the formation and the control of the columnar vacuum arc is of great importance. When the contacts in vacuum initially part while carrying high currents, the transition from the molten metal bridge to the bridge column arc still occurs. If the current is high enough, the power into the contact surfaces at the arc roots is high enough to fully compensate for the slow expansion of the radius of the bridge column arc and for the material being lost to the surrounding vacuum. The bridge column will then transition into a high-pressure columnar arc, which has properties similar to those of an arc in air at atmospheric pressure: see Figure 2.40. The stationary arc roots cause considerable erosion of the contact. Figure 2.41 shows erosion data taken by Mitchell [120] for an arc between

An example of a plasma plume after the decay of the high-current columnar vacuum arc, 5p before the 14kA peak ac current falls to current zero

FIGURE 2.39 An example of a plasma plume after the decay of the high-current columnar vacuum arc, 5p before the 14kA peak ac current falls to current zero: na is the neutral metal vapor density and n, is the ion density [119].

The columnar vacuum arc

FIGURE 2.40 The columnar vacuum arc.

Vacuum arc erosion as a function of the arc current for disc-shaped (butt), Cu contacts [120]

FIGURE 2.41 Vacuum arc erosion as a function of the arc current for disc-shaped (butt), Cu contacts [120].

Comparison of arc erosion of butt contacts in air and vacuum [120, 121, 122]

FIGURE 2.42 Comparison of arc erosion of butt contacts in air and vacuum [120, 121, 122].

disc-shaped or butt Cu contacts. For currents below «10kA the erosion is entirely from the cathode spots and gives the usual value of about 100 pgC-1. Once the columnar arc forms there is a rapid increase in erosion from both the cathode contact and the anode contact i.e., both contacts now have similar erosion rates. Figure 2.42 compares the erosion of these Cu contacts in vacuum with the erosion rate given by Turner & Turner [121] for arcs in air and for the actual, high-current, arc erosion of W-Ag contacts also operating in air [122]. Figure 2.43 presents a summary of the energy balance

Energy balance for the columnar vacuum arc

FIGURE 2.43 Energy balance for the columnar vacuum arc.

at the contacts for this stationary columnar vacuum arc. As can be seen from Figure 2.42 the erosion of the contacts for a given current by this arc can have considerable variability. Fortunately, the condition of a stationary columnar vacuum arc almost never occurs in practical vacuum interrupters, because vacuum interrupter designers have developed contact structures to control it. I will discuss this further in Chapter 3 in this volume.

Detailed analysis of the stationary columnar vacuum arc appearance between opening discshaped or butt contacts is rather complex and there has been very little model development for this vacuum arc. In a series of experiments Heberlein and Gorman [123] have presented a comprehensive description of the columnar vacuum arc’s development between opening Cu-Cr (25wt%), 100mm diameter contacts at ac current levels from ЮкА to 67kA (peak). Typical current and travel records are shown in Figure 2.44. The arc is photographed with a high-speed movie camera running at a speed of five to eight frames per millisecond. Exposure time of the individual frames is 50ps. In each movie the contact gap is determined as a function of arcing time by measuring the distance between the contacts in each frame of the movie. The frame where a first bright spot appears is identified as the beginning of arcing and is correlated to the instant of a voltage jump of 20V on the arc voltage record. Thus, it is possible to correlate the movie frames with particular arc phenomena, changes in arc voltage, current level and contact gap. By photographing many opening operations and a large number of ac current levels the “Appearance Diagram” shown in Figure 2.45 is developed. Here the arc appearance from the photographs is given as a function of contact gap and circuit current. They developed the following descriptive names for the observed columnar arcs, which are illustrated in Figures 2.45 and 2.46.

  • 1. The Bridge Column Arc: I have described this in Section 2.2. i.e., for currents less than about 5kA the bridge column gradually increases in diameter until the diffuse arc forms
  • 2. The Diffuse Column Arc: When contacts are separated at current levels between 7kA and 15kA the bridge column transitions into a diffuse column. The diffuse column diameter increases linearly with current from approximately 8mm to 20mm. The characteristics of this type of vacuum arc will be discussed in Section 2.5, the Transition Vacuum Arc
  • 3. The Constricted Column Arc or Columnar Arc: When the instantaneous current exceeds a value between ЮкА and 20kA, the diffuse column suddenly constricts and becomes very
Current and contact opening curves as the vacuum is photographed with a high-speed camera [123]

FIGURE 2.44 Current and contact opening curves as the vacuum is photographed with a high-speed camera [123].

The “Appearance Diagram” for the columnar vacuum arc at various currents and contact spacings [123]

FIGURE 2.45 The “Appearance Diagram” for the columnar vacuum arc at various currents and contact spacings [123].

The appearance of the high-current columnar vacuum arc

FIGURE 2.46 The appearance of the high-current columnar vacuum arc

luminous with well-defined boundaries (constricted column). The strong radiation intensity gradient at the column boundaries is demonstrated by the fact that the column diameter remains essentially constant for a wide variation of the photographic film exposures using neutral density filters. The column is cylindrical with a slight constriction in front of the anode; column diameter is typically 10mm and is relatively insensitive to the value of the instantaneous current. When contacts are separated at current values larger than 15kA, the columnar arc transitions directly from the bridge column arc

4. The Plasma Jet Column Arc: This arc is comparable to the constricted column in intensity, but it is wider where it attaches to the contact and it has a constriction in its center section. Its resulting appearance is thus two cones meeting in their apexes. Typical dimensions for the column diameters are 12mm at the anode, 10mm at the narrowest point, and 20mm at the cathode, but the diameters increase for currents above 36kA

5. The Anode and Cathode Plasma Jets: Further increase of the electrode gap finally results in separation of the anode jet from the cathode jet

In Figure 2.45 the dashed line in the diagram shows one representative arcing sequence. Here after contact separation at approximately 29kA, a constricted column arc forms; this changes into a jet column when the gap reaches a value of approximately 4mm. At a gap of 9.5mm, the jet column breaks up into an anode jet and a cathode jet, which subsequently disappear. When the current falls below a value of 18kA, the anode spot dies away and a diffuse arc forms. The boundaries between the regions of different appearances are the averages of a large number of points, and some uncertainty exists regarding the exact values of gap and current at which the actual transitions will occur. Further research on this vacuum arc between opening butt contacts by Zalucki et al. [124] generally show similar arc appearances. Figure 2.47 summarizes their appearance data for the vacuum arc between opening butt contacts during an ac current half cycle. Abplanalp et al. [125] using a spectroscopic technique have measured the temperature and the pressure of a columnar vacuum arc between Cu contacts. A representation of their data is shown in Figures 2.48 and 2.49 for 50 Hz currents 10kA, 20kA, and 30kA. As might be expected the azimuthal magnetic field provides a strong pinch force on the plasma column. At 30kA this results in a higher arc pressure than at ЮкА. Surprisingly the average temperature for the three currents are similar. Both the pressure and the arc temperature decline as the current approaches zero.

Observed sequences of vacuum arc appearance modes for a high-current vacuum arc on butt contacts [124]

FIGURE 2.47 Observed sequences of vacuum arc appearance modes for a high-current vacuum arc on butt contacts [124].

The arc pressure of the columnar vacuum arc during the passage of the ac current ШкА, 20kA. and 30kA [125]

FIGURE 2.48 The arc pressure of the columnar vacuum arc during the passage of the ac current ШкА, 20kA. and 30kA [125].

The arc temperature of the columnar vacuum arc during the passage of the ac current ШкА, 20kA. and 30kA [125]

FIGURE 2.49 The arc temperature of the columnar vacuum arc during the passage of the ac current ШкА, 20kA. and 30kA [125].

For the vacuum interrupter designer, the “Appearance Diagram” gives a good qualitative understanding of expected vacuum arc modes between large area contacts switching high ac currents in vacuum. As this columnar vacuum arc is generally stationary on butt contacts, the arc root regions will be strongly eroded. At an ac current zero the contact regions close to the arc roots would remain at a high temperature and metal vapor would continue to evaporate into the contact gap. Above a certain current level, therefore, the performance of a vacuum interrupter with butt contacts would be degraded. I will discuss this in Section 4.2.2. Thus, in a practical vacuum interrupter it is important to control this arc by either forcing the arc roots to move over the contact surfaces or developing a way to ensure a diffuse vacuum arc at high currents. Contact designs for the control of the high- current vacuum arcs will be discussed in Chapter 3 in this volume.

The Transition Vacuum Arc

Schulman and Slade [126] have given a full description of this vacuum arc mode. In their experiments the arc appearance has been photographed for disc-shaped, Cu-Cr (25 wt%) butt contacts opening an ac current for peak current values from 2.8kA to 15.7kA. The observed arc modes are

Observed sequences of vacuum arc appearance modes for a transition vacuum arc on butt contacts [126]

FIGURE 2.50 Observed sequences of vacuum arc appearance modes for a transition vacuum arc on butt contacts [126].

illustrated in Figure 2.50. For low separation currents, i.e., Is < ЮкА, a bridge column exists after the rupture of the molten metal bridge. As previously discussed in Section 2.2, this column begins as a high-pressure arc between the contacts. This arc gradually expands until it reaches a pressure of less than about 0.5 x 10s Pa [25], when it begins a much more rapid expansion into a diffuse vacuum arc. If the peak current Ip < 6kA, the expanding arc becomes fully diffuse for the remainder of the half cycle. For the diffuse mode, the rate of erosion from the contacts cannot keep pace with the rate of loss of plasma and neutral vapor from the arc to the surrounding vacuum. If Ip > 6kA, the diffuse-type arc between the still closely spaced contacts changes after ~lms into a diffuse column.

For opening current Ip > 6kA, as expected, the bridge column has a higher pressure when it reaches the end of its duration. At this stage of its development, the bridge column will not form a diffuse arc, although it would have been expected to keep expanding until it did so. Instead, the bridge column changes directly into a diffuse column. The reason for this development is apparently related to the instantaneous current level, the contact diameter and the overlapping plasma plumes above the cathode spots. When the diffuse vacuum arc collapses into the diffuse columnar mode, the diffuse column begins to form at the position of the original bridge column arc roots, or at the position of a particularly intense cathode spot. This could be expected to result from a continued heating of the localized hot area of the anode.

This diffuse column mode is characterized by its limited current range (I < » 15kA) and the high rate of vapor loss through its effective surface area. This new equilibrium condition can only be maintained over a limited range of pressure, which is roughly 0.2 x 105 Pa to 2 x 105 Pa. The contact erosion pattern of the diffuse column arc is quite unique and reveals lightly melted craters. There is no gross erosion or jets of material from the contacts. Analysis of the cathode crater shows that it consists of many cathode spots, which have been confined within it. Thus, for the transition vacuum arc, the cathode spots do form and initially move apart after the bridge column arc, but then reach a position where they no longer repel each other. Since this occurs at currents greater than several kiloamperes, the effects discussed in Sections 2.3.1 and 2.3.2, are forcing the development of this transition arc. Now the intercontact plasma has a high enough density to experience many inelastic collisions and metal particles that enter this plasma column can be evaporated. The resulting metal vapor will, in turn, become ionized and contribute to an increase in the plasma density. Conditions are achieved where the azimuthal magnetic field that results from the circuit current flowing in the plasma confines the plasma to a well-defined column. The cathode spots now are restricted from expanding beyond the boundaries of this plasma column. The work by Kimblin [127] may also give a clue to the structure of the transition vacuum arc. He shows that as the ambient pressure in an arc chamber increases from about 10~4 Pa to above 104 Pa, the cathode spots tend to remain within a fixed location, the cathode spot erosion decreases and the ion current collected on a surrounding shield biased to the cathode potential also decreases. At present, however, there is no adequate analytical model of the transition to this type of vacuum arc.

As the circuit current goes to zero in an ac circuit, the plasma plumes cease to strongly overlap at a given current and a given contact diameter. When this occurs the character of the plasma changes back to be essentially collisionless and the vacuum arc returns to the diffuse mode. Experimental observations of the diffuse column vacuum arc show that it always returns to the diffuse mode as the current goes to zero.

 
Source
< Prev   CONTENTS   Source   Next >