US3471734A - Periodic electrode structure for vacuum gap devices - Google Patents

Description

J A- RICH Oct. 7, 1969 PERIODIC ELECTRODE STRUCTURE FOR VACUUM GAP DEVICES Filed May 19, 1967 3 Sheets-Sheet 1.

/nvem0r Joseph A. Ric/7,

His A/fome y.

J. A. RICH Ucku 7, 1969 PERIODIC ELECTRODE STRUCTURE FOR VACUUM GAP DEVICES Filed May 19, 1967 3 Sheets-Sheet 2 &

50 i F/gA /nvenfor Joseph A. Ric/7 by QQ XL His Afro/nay.

J. A. RICH PERIODIC ELECTRODE STRUCTURE FOR VACUUM GAP DEVICES Filed May 19, l '7 3 Sheets-Sheet 5 Joseph A. Ric/2 v His Afforney- United States Patent 3,471,734 PERIODIC ELECTRODE STRUCTURE FOR VACUUM GAP DEVICES Joseph A. Rich, Schenectady, N.Y., assignor to General Electric Company, a corporation of New York Filed May 19, 1967, Ser. No. 639,843 Int. Cl. 1101i 17/04, 61/06 US. Cl. 313-217 8 Claims ABSTRACT OF THE DISCLOSURE J XB on the current paths between any adjacent opposite electrode pair is substantially zero and bunching of current paths to form destructive anode spots is avoided. Device may carry very high currents with very low current density at any given point.

Related applications The present application is related to my copending concurrently filed applications, Ser. Nos. 639,693 and 639,- 834, and the concurrently filed copending application of James M. Latferty, Ser. No. 639,844.

The present invention relates to vacuum gap devices adapted to operate at high current without the formation of anode spots therein. More particularly, the present invention relates to such devices in which the formation of anode spots is avoided by the configuration and position of the electrodes such that essentially no magnetic field exists within the interelectrode gap. Specifically, a pair of primary arc-electrodes comprising a plurality of electrode members are generally assembled so that the individual electrode members interleave with one another to form a periodic structure which renders magnetic fields normal to the conduction paths within the interelectrode gaps vanishingly small to prevent the formation of anode spots.

In the development of vacuum switches and triggerable vacuum gap devices, a limiting factor to the amount of current which can be drawn by a given structure is the threshold current at which a destructive anode spot is formed. Formation of such anode spots results in erosion of the anode electrodes and melting thereof. Such erosion and melting adversely effect the surface of the device, making the breakdown voltage change from its original value, eventually leading to the failure of the device. This is because erosion of the electrodes leaves irregular surfaces, which irregular surfaces cause perturbations in the electric field, which facilitate the breakdown at lower voltages. In developing prior art vaccum gap devices, many attempts have been utilized in order to keep anode spots from causing such destructive erosion. Generally, these attempts have been along the lines of accepting the fact that anode spots are inevitable and constructing the electrodes of such a configuration and providing the interaction space with magnetic fields, either due to the action of the arc itself, or due to external influences, so that the interaction of the magnetic field with the electric arc causes the arc to move, generally by rotating around the periphery of disc-shaped electrodes. Although these techniques are useful and do in fact result in longer life than vacuum gap devices in which such attempts have Patented Oct. 7, 1969 not been made, much is left to be desired in order to prevent the formation of anode spots.

In my copending application, mentioned hereinbefore, I have set forth my discovery leading to a new approach in the attack on anode spots in vacuum gap devices. Briefly stated, I have found that it is not necessary to accept the fact that anode spots must be lived with. Rather, since the body force or force per unit volume of conducting fluid acting upon any conduction path between a pair of arc-electrodes in a vacuum gap device is governed by the formula where F is body force, B is the magnetic field existing in the interelectrode gap, and J is the current density between these elctrodes, in accord with the general discovery set forth in the copending application, I eliminate or minimize the body force by eliminating or minimizing the normal magnetic force in the interelectrode gap. In one embodiment of the invention set forth in my copending application, I form an inner electrode from a re-entrant cylinder wherein the current path is folded-back upon itself, and form the other electrode from a concentric cylinder surrounding the first electrode. Current flow in the outer cylinder is longitudinal and thus, by Amperes law, current flow within the exterior cylinder results in no net magnetic field within the confines of the cylinder. In the inner-electrode, current in one direction within the re-entrant structure is substantially equal to current in the opposite direction in the other section thereof and, hence, the net magnetic field is essentially zero exterior of the re-entrant inner-electrode. Accordingly, the interelectrode gap, which is a cylindrical annulus between the two concentric cylindrical electrodes, is essentially magnetic field-free and essentially no body force acts upon the current paths between the concentric electrodes. The current paths are, therefore, not caused to bunch-up at one end of the device, thus causing the formation of destructive anode spots.

Although the class of devices described hereinbefore is a great advance upon the prior art and essentially opens a new field of development for vacuum gap devices, I have found that the precise geometry necessary to obtain a magnetic field-free region in the interelectrode space is diflicult to achieve. Similarly, if and when the device does fail, the entire active portion must be replaced. It is desirable that replaceable electrode elements be incorporated therein in order to serve as a device without completely replacing the electrode structure.

Accordingly, it is an object of the present invention to provide vacuum are devices wherein the electrode configuration greatly increases the current threshold for the formation of anode spots without requiring close tolerances and difficult configurations.

Yet another object of the present invention is to provide vacuum arc discharge devices which are capable of carrying very high currents without the formation of anode spots and utilizing replaceable electrode members in the event of failure.

Still another object of the present invention is to provide vacuum arc devices in which the interelectrode gaps are substantially free of magnetic fields which tend to cause bunching of the electric paths and the consequent formation of anode spots in a configuration that is simple, economical to manufacture and easily maintained.

In accord with one embodiment of the present invention, I provide improved vacuum arc discharge devices having high current thresholds for the formation of anode spots and including a pair of primary arc-electrode assemblies, each of which includes a plurality of electrode members and which are assembled together with the individual electrode members interleaved or interdigitated together so as to provide a plurality of interelectrode gaps, each of which is substantially free from magnetic fields transverse to the path of current conduction between the individual electrode members. Such a structure facilitates the carrying of high currents between adjacent opposite poled electrode members due to the large electrode area without bunching thereof, with the resultant formation of destructive anode spots. In one embodiment of the present invention, conduction between opposite poled electrode members is initiated by the injection of an electron-ion plasma into the interelectrode gap by the pulsing of the trigger electrode assembly. In accord with another embodiment of the present invention electron-ion plasma is created within the interelectrode gaps by the opening of a starter electrode which strikes an initial arc filling the device with plasma.

The novel features believed characteristic of the present invention are set forth in the appended claims. The invention itself, however, together with further objects and advantages thereof, may best be understood with reference to the following description, taken in connection with the appended drawing, in which:

FIGURE 1 is a vertical cross-sectional view of a triggerable vacuum gap device constructed in accord with the present invention,

FIGURE 2 is a vertical cross-sectional view, with parts broken away, of a vacuum switch constructed in accord with another embodiment of the present invention,

FIGURE 3 is a horizontal section view taken along lines 3-3 in FIGURE 1,

FIGURE 4 is a vertical cross-sectional view, with parts broken away, of an alternative embodiment of the trigger vacuum gap device of FIGURE 1,

FIGURE 5 is a horizontal plan view in section illustrating an alternative configuration to that illustrated in FIGURE 3,

FIGURE 6 is a schematic diagram illustrating the field configuration in the vicinity of four electrode members of the device of FIGURE 4,

FIGURE 7 is a schematic illustration of current conduction paths between adjacent electrode members of the devices of FIGURES 1 and 4, respectively,

FIGURE 8 is a vertical cross-sectional view, with parts broken away, of a triggerable vacuum gap device constructed in accord with another embodiment of the present invention,

FIGURE 9 is a horizontal sectional view taken along lines 9-9 of FIGURE 8,

FIGURES 10 and 11 are horizontal section views of alternative structures to that illustrated in FIGURE 9.

In vacuum are devices, the current threshold marking the onset of the formation of anode spots is a function of electrode geometry and electrode material. For a given material, therefore, the formation of anode spots is a function of electrode geometry. In the plane-parallel geometry, frequently utilized in switches in general and vacuum switches in particular, the threshold is relatively low, since a spot is formed at any point at which the current density becomes high, either due to surface irregularities or anchoring of the are due to the interaction of electric currents and magnetic fields. One means of inhibiting the formation of anode spots is to use electrodes having a very large area so that the current paths between the arc-electrodes are diffused over a very large area to lower the current density and prevent the formation of anode spots. Yet another means utilized to avoid the formation of anode spots, or to facilitate the carrying of high currents without destructive action of anode spots, is to utilize electrode configuration and magnetic fields, either caused by the current conduction paths or an external magnetic field, to interact to cause the arc to move over the electrode surface, most generally to rotate about the P p y Of a disc-Shaped electro e,- his tends to keep the burning and erosion at any given point of the anode to a minimum.

In accord with the invention set forth in my copending application, realizing that the force tending to cause conduction paths between a pair of oppositely-poled arcelectrodes to bunch together and form a high current density which results in the formation of an anode spot to be the result of the vector product of the current and the normal component of the magnetic field, I eliminate or reduce the normal magnetic field so as to reduce the magnitude of the vector production and, hence, the body force acting to propel the current conduction paths together, or to bunch or concentrate the arc discharge into a small region of the discharge volume. In accord with the configuration disclosed in my copending application, concentric cylindrical conductors are utilized with the inner conductor being re-entrant in structure to cause a foldingback of current therein with a net zero contribution to the azimuthal magnetic field in the interelectrode gap. In accord with the present invention, I find that it is not necessary to utilize a concentric configuration for the primary arc-electrodes, but that, rather, a number of electrode members may be arranged in interleaved relationship to form a periodic electrode structure in which magnetic fields transverse to the current conduction paths are minimized or eliminated. Such structures give greater freedom for the construction of vacuum are devices and facilitate the dissembly of a device that has failed to remove one or more electrode members without having to discard the entire device.

FIGURE 1 illustrates a triggered vacuum gap device constructed in accord with the present invention. In FIG- URE 1, triggerable vacuum gap 10 includes an upper electrode assembly 11 and a lower electrode assembly 12 joined with a cylindrical sidewall member 13 which is hermetically sealed to lower electrode member 12 by a dielectric or insulating seal 14. Upper electrode assembly 11 includes a base plate or disc 15 and a plurality of downwardly depending electrode members 16. Lower electrode assembly 12 includes a plurality of upwardly depending electrode members 18 and a base plate or disc 17. Each of the individual, downwardly depending electrode members 16 includes a central post 19 and a concentric cylindrical member 20 which is connected to central post 19 at the inward end thereof by disc member 21. Similarly, each of the upwardly-depending electrode members 18 includes a central post 22 and a concentric cylindrical mber 23 which is joined to central post 22 at the inner end thereof by disc member 24. The periodic structure caused by the interleaving of downwardly-depending electrode members 16 and upwardly-depending electrode members 18 causes the creation of a plurality of interelectrode gaps 25. The active surfaces of the arc-electrode members include the cylindrical members 20 and 23 of electrode members 16 and 18 respectively and the end caps 21 and 24. These materials are prepared from a high purity, high vapor pressure material as for example, copper or any of the materials set forth in Patent 2,975,- 256, to Lee et al., Patents Nos. 2,975,255 and 3,016,436 to Laiferty, and Patent No. 3,140,373 to Horn, and similar materials, alloys and intermetallic compounds which are operative to provide a copious quantity of metallic particles during arcing for the supplying of conduction carriers during operation of the device.

During operation, it is essential that none of the aforementioned vaporized metallic particles be deposited upon the iusulator separating the opposite electrode assemblies. Accordingly, ceramic or other insulating seal member 14 is protected by the bafiling arrangement provided by the lower end of sidewall member 13 and an annular flange 26 projecting upwardly from base member 17 of electrode assembly 12. In operation, a high voltage is connected between terminals 30 and 31 electrically in contact with electrode assemblies 11 and 12, respectively, and in circuit with an electric load tov be protected,

switched, or otherwise controlled. When it is desired to cause the device to change from a non-conducting to a conducting state, a pulse of electron-ion plasma is injected into the volume within device 10 from trigger assembly 27. Trigger assembly 27 includes a trigger anode 28 and a trigger cathode 29 in electrical contact with arcelectrode assembly 12. Trigger electrode 28 may conveniently comprise a metallized ceramic cylinder with a scored gap therein with the metal on one side of the gap connected electrically to trigger cathode 29 and the metal on the other side of the gap connected to a trigger anode lead 32. Although a simplified trigger assembly is shown herein, it is to be understood that any suitable trigger assembly operative to inject a cloud of electronion plasma into the interaction space between the arcelectrode members is suitable. Some such triggers are illustrated, for example, in the copending applications, Ser. Nos. 516,941; 516,942; and 516,943, of J. M. Lafferty, filed Dec. 28, 1965.

In operation, when a pulse of electron-ion plasma is injected into the interaction space, the plurality of interelectrode gaps 25 between individual electrode members 16 and 18 become the site of a number of small arcs or conduction paths with the conduction paths spreading rapidly over the many broad areas presented by the closest surfaces of the individual electrode members. The conduction paths are limited to the interelectrode gaps 25 because these are the shortest distances between any points in the device 10 which are electrically at the potential of the primary arc-electrodes. Although FIGURE 1 of the drawing is not meant to be exact scale, it is to scale in the respect that it clearly represents that the distance between the outermost, upwardly-depending electrode members 18 and the base plate of upper arc electrode assembly 11 is much greater than the interelectrode gaps 25. Similarly, the distance from downwardly-depending arc-electrodes 16 to the lower plate 17 or arc-electrode 12 is much greater than the distance of interelectrode gaps 25, as also is the distance between the outwardly-disposed downwardly-depending electrodes 16 and annular flange 26 which protects insulator 14 from Sputtering and short-circuiting.

Once the device 10 has become conducting with a plurality of conduction paths between adjacent oppositelypoled individual electrode members, the principles enunciated hereinbefore of the substantial elimination of magnetic forces within the interaction space becomes apparent. The conduction paths in any individual arc-electrode member, as for example, downwardly-depending arc-electrode members 16, are downwardly in cylinder 20 and upwardly in central post 21. Since these currents are substantially equal and in opposite directions, the magnetic field external of the electrode member due to current conduction paths therein is substantially zero. Similarly, in upwardly depending electrode member 18 current concluction is upwardly in cylindrical member 23 and downwardly in central post 22. As with electrode member 16, the substantially equal and opposite conduction paths and magnitude thereof causes a substantial cancellation of the external magnetic field due to the conduction path therein and the entire assembly between the individual arc-electrodes is substantially field-free. Although sidewall members 13 surrounds the entire interaction space, there is no conduction of electric current therein and no effect upon the magnetic field, either favorable or unfavorable results from the presence of member 13.

In the substantial absence of magnetic field in the interaction space between the arc- electrode members 16 and 18, there is substantially no body force tending to bunch the numerous conduction paths between the individual electrode members and thereby anode spots are avoided and very high currents may be carried before what minimal field remains is etfective to cause any bunching. Accordingly, a very great current conduction may be obtained with no destructive arcing or erosion of the arc electrodes. In the event that destructive arcing does occur, however, it may occur between only a particular pair of electrodes. In accord with the invention, it is a simple matter to cause each of arc- electrode members 16 and 18 to be individually removable as for example, with a screw-thread from base members 15 and 17 to make it possible to dissemble device 10 at seal 14, remove any damaged or eroded arc-electrode member and re-assemble the device, evacuate and re-seal. In devices in accord with the present invention it is possible to conduct currents of the order of hundreds of thousands of amperes at voltages of 50,000 to 100,000 volts in devices having a volume of approximately one cubic foot with substantially no arcing or erosion of the electrode members.

FIGURE 2 of the drawing illustrates a vacuum switch constructed in accord with the present invention. In FIG- URE 2, like members to those of FIGURE 1 are identified with the same reference numerals. Vacuum switch 40 of FIGURE 2 comprises a first upper electrode assembly 11 and a second lower electrode assembly 12 having a plurality of downwardly-depending electrode members 16 and upwardly-depending electrode members 18, respectively, as in FIGURE 1. Downwardly-depending electrode members 16 are comprised of a central post and a concentric cylindrical member capped with a planar disc 21, as are upwardly-depending electrode members 18. The electrode members are interleaved with one another in alternating fashion, so as to form a periodic structure as in the device of FIGURE 1. The envelope enclosing the interaction space of device 40 comprises upper electrode assembly 11 and lower electrode assembly 12 joined with a cylindrical, insulating dielectric sidewall member which may, for example, be fabricated from a high temperature glass such as Pyrex or Vycor or a high dielectric strength ceramic, as for example high density alumina, or a fosterite. Sidewall member 41 supports an insulator shield 42 which is supported by a flange 43 imbedded in an annular bead 44 which is integral with the inner portion of cylindrical sidewall member 41. Means to provide a quantity of ionized particles to cause breakdown between arc- electrode members 16 and 18 are provided in the form of a starter electrode 45 mounted upon an actuating rod 46, which is reciprocably movable and in contact with disc 21 of individual downwardly-depending electrode member 16 by means of a Sylphon bellows 48 which is suitably fastened to the outer periphery of an aperture 49 in lower base plate 17 and similarly fastened in hermetic seal by plate 48 to actuating rod 46. Starter electrode 45 is conveniently constructed of a refractory, low vapor-pressure material, as for example, tungsten or molybdenum. To initiate an arc, a force is applied to actuating arm 46, withdrawing starter electrode 45 from plate 21, causing an initial arc to be struck. Electrode 45 is completely withdrawn and seated within orifice 49 in plate 17. As the electrode 45 is withdrawn from plate 21, the arc is preferentially caused to spread out into the spaces between oppositely-poled electrodes and almost instantaneously the arc spreads out between the individual electrode members 16 and 18. Although one starter electrode is shown, any number may be used.

It will be noted from FIGURE 2 that device 40 utilizes a non-conducting sidewall member 41 with an associated shield member 42 and that the device of FIGURE 1 utilizes a metallic sidewall member 13. Since the sidewall members bear no significance to the electrical characteristics of the device, the device of FIGURE 1 may be constructed with the insulating sidewall member of the device of FIGURE 2. Similarly, the device of FIG- URE 2 may be constructed with the metallic sidewall member of the device of FIGURE 1. Actual structural details depend upon the intended use and the environment of the device.

FIGURE 3 of the drawing illustrates a plan sectional View of the device of FIGURE 1 taken along section lines 3-3. In FIGURE 3, the periodic structure of upwardly depending electrode members 18 and downwardlydepending electrode members 16 may readily be seen. It is to be noted that the interelectrode distance 25 represented by arrows A, which are the conduction paths between adjacent, oppositely-poled individual electrode members, are the shortest distances between any opposite polarity member of the device. Similarly, the distance between upwardly-depending electrode members 18 and metallic sidewall member 13, which is at the same electrical potential as downwardly-depending electrode members 16, is much greater than the interelectrode spacing 25. It may further be seen from the illustration of FIG- URE 3 that each individual electrode member (other than those of the periphery of the periodic array) is surrounded by four symmetrically-located, oppositely-poled electrode members. Since the azimuthal magnetic field around any individual electrode member which would be orthogonal to conduction paths between adjacent electrode members is substantially zero due to current conduction therein and the folded-back structure thereof, the orthogonal magnetic field within the interaction space within the device is substantially zero, and substantially zero net body force is eifective upon any given conduction path to cause a bunching thereof and formation of destructive, erosive anode spots.

FIGURE 4 illustrates an alternative structure for a triggerable vacuum gap device constructed in accord with the present invention. In FIGURE 4, the gap device includes a first upper arc-electrode member 11 and a second lower arc-electrode assembly 12 joined in hermetic seal with insulating dielectric sidewall member 41, as in FIGURE 2. Sidewall member 41 is protected from the deposition of sputtered or evaporated metallic particles and short-circuiting thereof by a shield member 42 which is imbedded by a flange 43 in an annular bead 44 on the inner surface of sidewall member 41. Upper electrode assembly 11 includes a base or end wall plate 15 and a plurality of downwardly-depending electrode members 50, which are solid, but which may be hollow provided it has sufiicient thermal conductivity, and are terminated at end 51. Lower arc-electrode assembly 12 includes a flat base plate 17 and a plurality of upwardly-depending solid electrode members 52 terminated at ends 53. The downwardly-depending electrode members 50 and the upperwardlydepending electrode members 52 of upper and lower electrode assemblies 11 and 12, respectively, are interleaved between one anotehr to form a periodic structure as in the devices of FIGURES 1 and 2. Means for producing an ionized electron-ion plasma to cause the device of FIGURE 4 to be rendered conductive is provided in the form of a trigger electrode assembly 25, similar to that of FIGURE 1 of the drawing. Means for connecting the device in circuit with an electric load to be switched, protected, or otherwise controlled, is provided by means of terminal lugs 30 and 31.

FIGURE of the drawing illustrates, in horizontal plan view, a section taken through lines 5-5 of FIG- URE 4. This plan view of the device of FIGURE 4 illustrates square or rectangular symmetry, rather than circular symmetry as is illustrated in the device of FIGURE 3. It should be appreciated, however, that the square or rectangular symmetry of FIGURE 5 may be utilized with the devices of FIGURES 1 and 2 and the circular symmetry of FIGURE 3 may be utilized with the device of FIGURE 4. In FIGURE 5, it may be readily seen that the periodic structure illustrated in FIGURE 3 is maintained herein, with each of upwardly-depending electrodes 52 being surrounded by a plurality of downwardly-depending electrodes 50 (except at the periphery of the array). Similarly, it may be seen that the inter-electrode spacings 25 are smaller than any spacings between any individual electrode member and any other member having the same potential as the oppositely-poled electrode members. It

may also be noted that the individual electrode members are solid, rather than composed of a concentric structure as in the devices of FIGURES l and 2 illustrated in plan view in FIGURE 3.

FIGURE 6 illustrates a schematic illustration of the field configuration surrounding an assembly of four juxtaposed electrodes 50 and 52 as in FIGURE 5 of the drawing. This view is taken with all four electrodes in section. In FIGURE 6, interelectrode gaps 25 exist between oppositely-poled electrodes 50 and 52. Similarly, since there is no re-entrant structure in the device of FIGURE 4 to cause a folding-back of current conduction path and a net zero magnetic field exterior of each electrode, it is apparent that there will be some magnetic field. It has been found, however, that the periodic array of these electrode structures does not depart markedly from the ideal array in the ideal field configuration obtained in the devices of FIGURES l and 2. In FIGURE 6, the annular area 60, immediately surrounding each of electrodes 52 represents a moderately dense azimuthal magnetic field in the direction indicated by the arrows therein (assuming current into the paper). Similarly, the annular section 61, immediately surrounding electrodes 50 represent a moderate azimuthal magnetic field in the direction of the arrows as shown. Because of the opposite direction of the magnetic fields in the interelectrode gaps the majority of the space 62 existing outside of annular regions 60 and 61'is substantially magnetic field-free and the center portion of the array 63 is exactly magnetic field-free. Although this field configuration has some effect upon the conduction paths between electrode members 50 and 52, it is not as drastic as one may think and the principles of the present invention may still be substantially realized in such a structure. This will be evident from a consideration of FIGURE 7.

In FIGURE 7 a schematic representation of current conduction paths between an upper electrode assembly 11 and a lower electrode assembly 12 having only one individual downwardly-depending electrode member 50 and one upwardly-depending electrode member 52 is shown. Current paths within arc- electrode members 50 and 52 are represented by arrows C. If electrode members 50 and 52 were of the reentrant type as illustrated in FIGURES 1 and 2 of the drawing, the conduction paths between electrode members 50 and 52 would be the sheath enclosed within arrows D. It is apparent, in view of the great area encompassed by the sheath over electrodes 50 and 52 that the principles of the present invention are realized almost ideally in that conduction paths are greatly spread out over the electrode surfaces with essentially no bunching. With the structure as illustrated in FIGURE 4 and FIG- URE 5 of the drawing, the moderate azimuthal magnetic fields 60 and 61, immediately surrounding the individual electrode members 52 and 50, respectively, tend to exert a moderate body force upon the conduction paths in the immediate vicinity of the electrodes. Since, however, these body forces acting upon the current paths are opposite in direction, the conduction paths are urged in opposite directions in the immediate vicinity of electrodes 50 and 52, causing the current paths to occupy the shaded area represented at E in FIGURE 7. Although there is some degree of bunching of the conduction paths in the solid configuration illustrated in the device of FIGURE 4 and further illustrated in FIGURES 6 and 7, it should be readily apparent that the bunching is not sufiiciently great as to cause such an increase in current density sulficient to cause the formation of destructive anode spots at all except relatively high currents. While it is conceded that the current carrying capacity of devices constructed in accord with the embodiment of FIGURE 4 of the drawing does exhibit a slightly lower threshold for formation of anode spots, this threshold is nevertheless substantially higher than that found in conventional are devices and, to a large measure, the advantages of the present invention are achieved with a great simplicity of construction utilizing the construction illustrated in FIG- URE 4.

FIGURE 8 of the drawing illustrates still another alternative embodiment of the present invention. In FIG- URE 8, a triggerable vacuum gap device 70 includes an upper electrode assembly 11 and a lower electrode assembly 12, connected in hermetic seal with a cylindrical insulating sidewall member 41. Upper electrode assembly 11 includes a base plate or disc 15 and a plurality of downwardly-depending individual electrode members 16, each of which is a thin planar vane and which are arranged radially about the longitudinal axis of device 70. Lower electrode assembly 12 comprises a base plate or disc 17 and a plurality of upwardly-depending individual electrode members 18, each of which like the electrode members 16, is a substantially thin planar vane, and which are radially disposed about the longitudinal axis of device 70.

Upon assembly of the device, the vanes 16 and the vanes 18 are interleaved between one another so as to form a periodic structure with alternate vanes connected to one electrode assembly and other alternate vanes connected to the opposite electrode assembly. Assembly is made so that the interelectrode distance between any given pair of oppositely-poled electrode members 16 and 18 is equal to the same distance between any other pair of oppositely-poled electrode members 16 and 18 within device 70. A vapor shield 42, supported from a flange 43 inbedded in annular bead 44 on the inner surface of cylindrical sidewall member 13 protects the inner surface thereof from becoming covered with metal lic particles and short-circuiting. Contact to the electrode assemblies is made through terminal lugs 30 and 31, respectively. As in the embodiment of FIGURE 4, the individual vanes which constitute the electrode members are shorter than the length of the device so that the closest distance between any portion of the oppositelypoled electrode assemblies or any other material at the same potential is the interelectrode gap existing between adjacent electrode members. Conduction between electrode assemblies 11 and 12 is initiated by trigger electrode assembly 27, which has the identical or functionally similar structure to that of the trigger electrode assemblies of the devices of FIGURES l and 4. In operation, the device 70 is rendered conductive by an electrical pulse to trigger lead 32, to cause the injection of an electron-ion plasma into the space between individual electrode members 16 and 18 to cause electrical breakdown therebetween.

FIGURE 9 illustrates, in horizontal section along the lines 9-9 in FIGURE 8, a cross-sectional view of the triggerable vacuum gap 70 illustrated in FIGURE 8. As may be seen from FIGURE 9, downwardly-depending electrode members 16 and upwardly-depending electrode members 18 are alternately interleaved in an annular pattern and define a plurality of interelectrode gaps 25. Trigger assembly 27 is visible at the center of base plate 17. Since the electrode members of the device of FIGURE 8 are not folded-back to cause the elimination of all magnetic fields, between electrode members 16 and 18, the principles upon which the present device operates is somewhat modified from that of the device of FIGURES 1 and 2, and resembles more nearly the operation of the device of FIGURE 4, although there is a substantial difference. As with the device of FIGURE 4, there is a small region of magnetic field in the region immediately surrounding each of the arc- electrodes 16 and 18, electrical conduction within each of the electrode vanes being in the same direction. Since, however, the arcing path illustrated as arrow A in FIGURE 9 is acted upon by magnetic forces that are in opposite directions in the immediate vicinity of the respective arc- electrode members 16 and 18, the distribution of the current paths between, any individual arc-electrode pair is rather similar to that illustrated at E in FIGURE 7. The field configuration which obtains in the device of FIGURE 4 and which is 10 illustrated schematically in FIGURE 6 of the drawing does not exist, however, because there is not such a periodic structure in which each electrode is surrounded on all four sides by an electrode member of the opposite polarity.

In all other embodiments of the present invention the force acting upon the current path between adjacent electrode members has been minimized (except at the region immediately surrounding the electrode members in the embodiment of FIGURE 4) by causing the magnetic field that is orthogonal to the current to be minimized or eliminated. Thus the product approaches zero. In the present embodiment of the invention, essentially the same result will be obtained by allowing B to exist, but by rendering B substantially parallel to J. Thus, for example, since I B is the vector product if the two vectors are parallel or substantially parallel, or even antiparallel, the product equals or approaches zero. In the device of FIGURE 8, such is the case. Turning for the moment to FIGURE 9, it may readily be seen that any two adjacent electrode members are surrounded by a substantially equal magnetic field due to current flow in the same direction in the adjacent electrode vanes. Since, however, with respect to the current represented by arrow A, the radial components of the magnetic fields due to current conduction in respective electrode members substantially cancel in the intervening space betwen the electrodes and the azimuthal component of the magnetic fields due to current conduction in adjacent electrode vanes is additive. As a result of this characteristic of the magnetic field in this configuration of the present embodiment, the resultant magnetic field (except in the immediate vicinity of each electrode vane) is an azimuthal field which is substantially parallel with current conduction paths between adjacent arc-electrodes. Accordingly, due to this configuration the product approaches zero and substantially no body force (other than that which causes the so-called rail-gun configuration illustrated at E in FIGURE 7) is operative upon current conduction paths between the arc-electrode members. As a result current conduction path bunching at the anode is avoided and destructive anode spots do not form. Accordingly, the configuration illustrated in FIG- URE 9 and incorporated in the device of FIGURE 8 is operative, with a relatively simple construction, to achieve substantially the same result as the more complicated structures of FIGURES 1 and 2, and to permit the carrying of exceedingly large values of current within the device 70, without causing a high current density to exist at any place within the device to cause the formation of destructive anode spots. Thus, the current threshold for the formation of anode spots is greatly increased and the current carrying capacity of the device is exceedingly high as compared with prior art devices.

The foregoing is particularly advantageous, since the construction of the device of FIGURE 8, like that of the device of FIGURE 4, is relatively simple and needs no complicated arrangement requiring close tolerances. Similar to the device of FIGURE 4, the individual electrode vanes may be made removable so that the device may be dissembled by breaking the seal in the vicinity of upper or lower base plate, replacing one or more arcelectrode members and re-assembling the device with a new seal and placing the device back in service for a very small fraction of the cost it would take to fabricate an entirely new device.

A limiting factor to the effectiveness of the configuration of the device of FIGURE 8, illustrated in plan view in FIGURE 9, is the lack of parallelism between the adjacent arc- electrode members 16 and 18. This may readily be remedied by rendering each arc- electrode member 16 and 18 wedged-shaped, either in a solid wedge as illustrated in plan view in FIGURE 10, or in a bent V-type wedge as illustrated in plan view in FIGURE 11. Other than the dimunition of the number of vanes utilized in the attainment of the preferred parallel symmetry of the interelectrode spaces 25, devices constructed with the plan view illustrated in FIGURES and 11 are substantially the equivalent of that illustrated in plan view in FIG. 9, although improved operating characteristics and higher current thresholds for the formation of anode spots are obtained.

From the foregoing, it should be evident that I have provided a new approach to the concept of raising the threshold for the formation ofanode spots in vacuum are devices such as triggera'ble vacuum gaps and vacuum switches by utilizing a periodic structure for the individual arc-electrode assemblies within which a plurality of individual arc-electrode members are interposed interdigitally or interleaved between the similar electrode members of the opposite arc-electrode assembly to cause the creation of a periodic structure which defines a plurality of interelectrode gaps. In these gaps the magnetic field orthogonal to the path of current flow between the broad areas of the individual arc-electrode members is substantially eliminated or minimized, to prevent any body force being operative to cause bunching of current conduction paths and the formation of destructive anode spots.

Although the invention has been disclosed with respect to specific embodiments, numerous modifications and changes may be made. Thus, for example, it is within the scope of the present invention that the individual electrode members, although illustrated with certain crosssectional areas, mainly circular, may be square, triangular, polygonal, or ay other configuration which provides a broad area of substantially parallel interelectrode spacings to increase the number of conduction paths between opposite electrodes and minimize the current density. Similarly, circular symmetry may be replaced with a plurality of elongated parallel electrode members which alternate and which may be folded or vanes, as desired. Many other modifications and changes will occur to those skilled in the art. Accordingly, I intend, by the appended claims, to cover all such modifications and changes as fall within the true spirit and scope of the present invention.

What I claim as new and desire to secure by Letters Patent of the United States is:

1. A vacuum arc discharge device adapted to carry high currents without the formation of anode spots comprising:

(a) an hermetically sealed envelope evacuated to a pressure of 10- tort or less;

(b) a first primary arc-electrode assembly disposed Within said envelope and including a first plurality of spaced substantially parallel cylindrical electrode members extending substantially normal to a first endwall member;

(0) a second primary arc-electrode assembly within said envelope and including a second plurality of spaced substantially parallel cylindrical electrode members extending substantially normal to a second endwall member and interleaved in alternating parallel spaced relationship between the spaced electrode members of said first arc-electrode assembly;

(d) said first and second spaced cylindrical electrode members providing during operation a vector prodnot a J X which is insignificantly small in the spacings between electrode members of first and second electrode assemblies where J =current density in the arc discharges between any given pair of opposite electrode members and B=magnetic field between any given pair of electrode members due to current paths within said electrode members when said device is in a current conduction condition;

(e) means for causing an electric arc break-down to be established between said primary arc-electrode assemblies; and

(f) means for connecting said arc-electrode assemblies in circuit with an electric load.

2. The device of claim 1 wherein said first electrode assembly includes an upper base member and a plurality of downwardly-depending electrode members and said second electrode assembly includes a base member and a plurality of upwardly-depending electrode members.

3. The device of claim 2 wherein each of said upwardly-depending and downwardly-depending electrode members is comprised of a central post and a concentric cylindrical member which is joined to said central post at the end thereof that is remote from said base member.

4. The device ofclaim 2 wherein each of said electrode members is in the form of a solid member projecting from said base member.

5. The device of claim 2 wherein said periodic array constitutes a pattern in which each electrode member of said first electrode assembly is surrounded by a plurality of symmetrically arrayed electrode members of said second electrode assembly except at the periphery of said device.

6. The device of claim 2 wherein said device exhibits circular symmetry about a longitudinal axis.

7. The device of claim 2 wherein said device is a triggerable vacuum gap device and the means for supplying an electronion plasma therein is a trigger electrode assembly.

8. The device of claim 2 wherein the device is a vacuum switch and the means for supplying an electron-ion plasma therein is a starter electrode adapted to establish a starter arc discharge therein.

References Cited UNITED STATES PATENTS 3,356,893 12/1967 Lafierty 3l5ll1 3,356,894 12/1967 Lafierty 315-111 JAMES W. LAWRENCE, Primary Examiner R. F. HOSSFELD, Assistant Examiner US. Cl. X.R.

Images