US3601647A - High power electron discharge device having anode with improved heat dissipation capability - Google Patents

Abstract

An electron discharge device having an anode comprising a plurality of hollow pipes. The pipes have relatively thin, continuous walls of uniform dimension and generally may be of any cross-sectional geometry. The pipes are adjacently disposed in a circular or linear array such that the electron-intercepting surfaces thereof are oblique to the direction of the electron beams. The angles of incidence of the electron beams on the pipe surfaces preferably have a minimum average value. A cooling medium is circulated through the interiors of the pipes.

Description

United States Patent [56] References Cited UNITED STATES PATENTS [72] Inventor Fred George Hammersand East Petersburg, Pa. 802,783

[2]] Appl. No.

Filed [22] Feb. 27,1969 [45] Patented Aug. 24, 1971 [73] Assignee RCA Corporation Assistant Examiner-E. R. LaRoche Anarney-Glenn H. Bruestle e d dl n mmazm P a m m h h T g S w. mm i w m h e T 8 s e .m P

ABSTRACT: An electron discharge device havin comprising a plurality of hollow tively thin, continuous walls of uniform generally may be of any cross-sectional E w m DT m n R AD mm m DD-Y mr OII nuuw HBF C A Ewenu E w "w N NC-m N s w Nplll II-h G IAI Hub 4 .H.

[52] U.S. 313/30, are adjacently disposed in a circular or linear array such that 313/32, 313/35, 313/39, 315/538 the electron-intercepting surfaces thereof are 0 Int. direction ofthe electron beams. The an blique to the gles ofincidence of the H0lj1/42, HOlj 19/36 ..313/24,30,

electron beams on the pipe surfaces preferably have a minimum average value. A coolin g medium is circulated [50] Field of through the interiors of the pipes.

PATENTED ALIG24 I97! SHEET 1 OF 3 mm M0 Wm m m H G. w F

' SHEET 2 OF 3 IN YEN T 08 Fred 6. Hammersand A f TOR! Y PATENIEH Aus 24 Ian sum 3 0r 3 Fig. 7.

l N VEN TOR Fred G Hammersand ATTORNEY IIIGII POWER ELECTRON DISCIIARGI'J DEVICE IIAVING ANODE WI'III IMPROVED IIIEAT DISSIPATION CAPABILITY BACKGROUND OF THE INVENTION 1. Field of the Invention This invention relates to electron discharge devices and particularly to anodes for these devices. One type of electron discharge device where this invention may be utilized is a multielectrode, superpower beam power tube.

2. Description of the Prior Art A multielectrode, superpower beam power tube is one in which directed electron beams are used to increase substantially the power-handling capability of the tube. In one tetrode of this type, there is a plurality of cathodes surrounding a central anode, with both a control grid and a screen grid located between the anode and each cathode, the control grid being located closer to the cathode.

The anode may be, for example, a single hollow cylinder of relatively large diameter that is centrally located within the tube and made of conductive material, such as copper. Electrons emitted by the cathodes impinge upon the anode surface at substantially right angles. The cathode may be comprised of thermo'emissive filaments of thoriated tungsten, for example, mounted on supporting means. The cathode filaments may be arranged in a cylindrical array, concentric with the anode and close to the outer periphery of the tube. In some tubes, however, the anode and cathode positions may be reversed so that the cathode is centrally located.

When electrons emitted by the cathode are collected on the anode surface, there is a dissipation of electron kinetic energy. The kinetic energy is converted to heat at the anode surface so that it is necessary to cool the anode. Where the anode is hollow, cooling may be done by circulating a cooling medium through the anode interior. At high beam current densities and/or short beam pulse duration, the absence of adequate heat dissipation causes the anode surface to be heated to temperatures which may vaporize the anode material and/or induce mechanical fatigue. For the foregoing reasons and because higher anode heat dissipation allows higher peak RF and/or longer duty levels, it is desirable to maximize such heat dissipation. Where the anode is hollow and a coolant is circulated through the anode interior, previous attempts to increase the anode heat-dissipation capability have included fluting the interior, or back, surface of the anode. The fluted surface is intended to increase the anode area which is in contact with the coolant, thereby permitting greater heat transfer from the anode. However, fluting of the interior surface of the anode wall has provided only a limited improvement, especially under relatively short beam pulse conditions. This results from the short beam pulses causing heat to be generated on the impinged anode exterior surface more rapidly than heat can be transferred through the anode wall to the interior surface, notwithstanding the fluting. This is attributable to the fluting having very little effect in increasing the effective" surface area of the anode. The effective surface area is defined as the average of the respective areas of those two uninterrupted concentric surfaces of the anode wall which (i.e., the surfaces) are separated by the shortest distance through the anode wall. In the case of a cylindrical anode with a fluted interior surface and an uninterrupted exterior surface, the effective surface area is the average of both the area of the exterior surface and the area of the interior surface exclusive of the fluting.

Another attempt to increase the anode heat-dissipation capability involves pyramiding the anode exterior, or front, surface so as to increase the surface area that is impinged by the electron beam and thereby reduce the power density experienced by a unit of anode impinged surface area. Pyramiding is done by cutting two intersecting series ofV grooves in the electron-impinged anode surface so that pyramid-shaped islands project from the anode. The effective surface area of an anode having a pyramided exterior surface and an uninterrupted interior surface is the average of both the area of the interior surface and the area of the exterior surface exclusive of the pyramiding This approach is not entirely satisfactory because, though the impinged surface area is increased, the effective" surface area presented to the heat flow away from the impinged surface is substantially the same as that without pyramiding. As a result, there is not improvement in the transfer of heat from the impinged anode surface. Also, the sharp corners at the bases of the pyramides provide points of stress concentration and are often the location of physical failure of the anode structure. Hence, pyramiding and fluting of the anode surfaces respectively increase the total area that intercepts electrons and the total area in contact with the coolant circulated through the anode, but neither results in any significant increase in the effective surface area of the anode.

SUMMARY OF THE INVENTION The novel electron discharge device includes an anode which is comprised of a plurality of adjacent, and preferably contiguous, heat conducting pipes having relatively thin, continuous walls of substantially uniform dimension. The pipes may be of triangular, circular, or other cross-sectional configuration, and may be arranged in a circular, linear, or other array. The pipes are disposed such that their respective external surfaces are oblique to the direction of the electron beam or beams. The pipes are disposed such that an electron beam is intercepted by an anode surface or surfaces, on the average, at as small an angle as possible. It is preferred that each elec' tron beam is equally intercepted by two adjacent pipes at the region of their juncture, and that the width of each electron beam is less than the sum of the respective breadths of the two pipes intercepting that beam.

By disposing the anode surface obliquely to the direction of the electron beam, the electron beam is intercepted over an anode area which is dimensionally greater than the cross-sectional area of the beam. By using anode pipes of relatively thin walls and of relatively uniform dimensions, the effective surface area of the anode, as well as the interior surface area of the anode in contact with a coolant passed therethrough, is substantially equal to the increased external surface area used for intercepting the beam. Thereby, the: novel device exhibits improved heat dissipation capability.

Some advantages of the novel device are elimination of expensive fluting and/or pyramiding of the device anode; improved power-handling capacity of the device and improved anode life.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. I is a fragmentary perspective view of the novel electron discharge device.

FIG. 2 is a sectional view of the novel electron discharge device along the axis 2-2.

FIGS. 3, 4 and 5 are fragmentary sectional plan views of arrangements of pipes of various cross-sectional configurations which may be used in the novel electron discharge device.

FIG. 6 is a sectional plan view of two pipes to provide an analysis of the invention.

FIG. 7 is a fragmentary sectional view of another embodiment of the invention comprising a linear array of pipes, wherein similar reference numerals are used for similar structures to those shown in FIGS. I and 2.

DESCRIPTION OF THE PREFERRED EMBODIMENT FIGS. I and 2 illustrate a tetrode tube It) containing a circular anode, generally designated Ill, surrounded by a concentric circular array of cathodes 112. The cathodes 12 are, for example, filaments of thoriated-tungsten or filaments of nickel alloy which bear matrices (not shown) of pressed and sintered nickel powder. The cathode filaments extend between supporting means (not shown). Where such nickel alloy filaments are used, the matrices are impregnated with a coating of alkaline earth compounds to provide electrons. The alkaline earth compounds may consist essentially of thermally-decomposed barium, strontium, and calcium carbonates. Each cathode 12 is located within the channel 14 of a vertically disposed, U-shaped grid member 16 which serves to confine the electrons emitted by its associated cathode 12 to a relatively narrow sheet beam 20 (for purposes of simplicity, only one electron beam 20 is illustrated, that in FIG. 2). The U- shaped grid member 16 is disposed such that its open side faces the anode 11. Fine wires 22 of heat resistant material, such as tungsten, for example, are mounted on the U-shaped members 16 such that the wires 22 extend horizontally across the open sides thereof. Voltage is applied to these wires 22 so that they act as a control grid in the operation of the tetrode tube 10. The means for applying voltage to these wires (not shown) are known in the art. The anode 11 is centrally located within the tetrode tube and the individual cathodes l2 and the U-shaped grid members 16 are positioned concentric with and parallel to, the anode 1 1.

Between the anode l l and the control grid wires 22, there is positioned a circular array of vertical post members 26 made of copper, for example. These post members 26 serve as a second beam-confining means. Rectangular frames 28 made of copper, for example, fit into longitudinal slots at the sides of these post members 26, each pair of post members 26 supporting one frame 28. The post members 26 are adapted such that the supported frames 28 face the anode. Very fine wires 30 of heat-resistant material, such as tungsten, for example, extend horizontally between and have their ends attached to, the vertical members of the frames 28. These wires 30 serve as the screen grid in the operation of the tetrode tube 10. The wires 30 of the screen grid are in registry with and shaded from the individual cathodes 12 by, the wires 22 of the control grid. The electrons of each beam strike the anode 11 as a substantially continuous sheet having dimensions of height and width. The dimensions of the sheets are determined in part, by the geometries ofthe U-shaped grid members 16 and the post members 26, both of which serve to confine the beam; by the electric fields created by the screen and control grids; and by the cathode geometry.

The anode 11 is comprised of a plurality of pipes 32 of uniform dimensions and relatively thin walls, which pipes are disposed so as to intercept an electron beam 20 at an oblique angle. As used herein, a pipe is defined as an open-ended hollow member having a relatively thin, continuous wall of substantially uniform dimensions. There may be used, for example, pipes having an outside diameter of about 0.5 inch with a wall thickness of about mils. The pipes should be made of a heat conducting material, such as copper. Pipes of various cross-sectional geometric configurations may be used so long as the interception of an electron beam by the pipes is at an oblique angle. Generally, the anode pipes will intercept an electron beam at an oblique angle if the external electron-intercepting surfaces of the respective pipes are nonlinear in their transverse direction. Some examples of pipes having such nonlinear surfaces are those of circular or elliptical crosssectional configuration (FIGS. 1 and 2, and FIG. 3, respectively).

Alternatively, an anode may be comprised of pipes having external surfaces which are substantially linear in their transverse direction, so long as such pipes are disposed such that the electroncollecting portions of the pipe surfaces are oblique to the direction of the respective electron beams which they intercept. Pipes having linear external surfaces include those pipes whose cross-sectional configuration is substantially triangular (FIG. 4) or substantially rectangular (FIG. 5).

While the pipes (whether they are linear or nonlinear in their transverse direction) may be disposed such that an electron beam is obliquely intercepted by a single pipe, only, for reasons given below, it is preferred that such pipes be disposed such that an electron beam is obliquely intercepted by two adjacent pipes in substantially equal amounts.

Referring again to FIGS. 1 and 2, the pipes 32 of the anode 1 l are disposed adjacent to each other, it being preferred that each pipe 32 by contiguous with adjacent pipes so that substantially none of the electron beam 20 passes between the anode pipes 32. While FIGS. 1 and 2 illustrate the pipes 32 as being disposed in a circular array, other arrays (e.g., the linear array shown in FIG. 7) may be used. The pipes 32 have their lower ends 34 opening into a lower chamber 36 through which cooling fluid (not shown) can pass. Alternate ones 32a of the pipes are mounted so as to have their upper ends 37 opening into a first upper chamber 38 such that a cooling medium is able to enter this chamber 38 from a source (not shown) outside the tube and pass through the first upper chamber 38 and into the alternate ones 32a of the pipes opening therein. The other ones 32b of the pipes have their upper ends 39 opening into a second upper chamber 40. Coolant passes into the second upper chamber 40 from the other ones 32b of the pipes and thereafter is removed from the tube by means (not shown) known in the art. Hence, the course of the cooling medium, as shown by the arrows, is entry into the first upper chamber 38 from an external source (not shown) by means of a pipe 42 leading into that chamber 38. The coolant then passes down through the alternate ones 32a of the pipes to the lower chamber 36, through the lower chamber 36, and up through theother ones 32b of the pipes to the second upper chamber 40, from which the coolant is then withdrawn. Thereby, coolant is continuously being circulated through the pipes 32 so as to remove the heat generated by the interception of electrons on the external surfaces of the pipes 32. The pipe surfaces which extend between the upper and lower chambers 36 and 38, respectively, and are available for intercepting the electrons emitted from the cathodes, should be at least as long as the beam height so that the beam 20 impinges only upon the pipes 32.

In order to achieve a higher degree of power dissipation it is preferred that the anode pipes 32 be disposed such that each electron beam 20 (FIG. 2) is obliquely intercepted by the surfaces of two adjacent pipes 32 in substantially equal amounts an in the region of the juncture of these two adjacent pipes 32. This is done by disposing the pipes 32 such that the center line of each electron beam coincides with the juncture of the two adjacent pipes 32 intercepting that beam. In this way, the total anode surface area intercepting an electron beam 20 is considerably greater than the cross-sectional area of the beam 20. As a result, the power density experienced by a given unit of anode surface area is generally, and preferably always, less than the beam power density. Beam power density is defined as the beam power per unit cross-sectional area of the beam normal to the direction of its path.

Because of the uniform relative thinness of the anode pipe walls, the internal and external surfaces of the respective pipes are substantially coextensive. Hence, the increase in the external surface area of a pipe 32 intercepting the beam is accomplished by a corresponding increase in both the total internal surface area and the total effective surface area of that pipe. This brings about improved heat transfer from the anode and other benefits.

A specific analysis of this structure is given with reference to FIG. 6 where there are shown only two pipes 50 and 60 of an anode, each pipe 50 and 60 intercepting substantially onehalf of the beam 20. The external surface 58 of one pipe 50 is shown divided, for purposes of analysis, into equal incremental, or unit, areas 62, 64, and 66. The pipe 50 is of circular cross section so that the surface 58 is nonlinear, and the pipes are selected so that the width of the beam 20 is less than the diameter of a single pipe 50, 60. As a result, each one of these incremental areas- 62, 64 and 66 is disposed at an angle oblique to the direction of the electron beam 20. The incremental area 66 closest to the point of juncture 46 of the pipes 50 and 60 is least angularly disposed to the beam 20, (that is, the angle of interception of the beam 20 is smallest). The beam power experienced by the respective incremental areas 62, 64 and 66 of the pipe surface 58 increases from the point of juncture 46 toward the point 49 where the pipe 50 is impinged by the lateral extremity 46 of the beam 26. The greatest amount of beam power is experienced by that incremental area 62 located in the region of the point 49 impinged by the beam extremity 46 since the angle of interception of the beam is greatest in this region. However, because the beam width is less than a pipe diameter, no portion of the beam 20 is intercepted at an angle of 90 so that the average value of beam power over the incremental area 62 is less than the beam power density.

Where an electron beam is intercepted by two adjacent pipes which individually have a breadth less than one-half the beam width, part of the electron beam will be intercepted at an angle of approximately 90. Those pipe surface regions intercepting the beam at about 90 experience a beam power that is substantially equal to the beam power density. The breadth of a pipe is defined as the shortest distance between a line drawn through the point ofjuncture of that pipe and the adjacent pipe intercepting the same beam (all points on the line being equidistant from the centers of both such pipes), and that part of the pipe surface most closely disposed to the cathode. In the case ofa circular pipe, the breadth would be the pipe radius. For this reason, it is preferred that the pipe dimensions be such that the beam width be less than the sum of the respective breadths of the two pipes intercepting that beam.

The average value of beam power experienced by an incremental area on a pipe surface is calculable by the formula:

Da -D Sin D where Da is the average power density at the incremental area of the anode surface; I is the angle between the center line of the beam and the tangent to the anode pipe surface at the midpoint of the incremental area (i.e., the angle of incidence); and D is the average beam power density.

It may be seen from the above formula that the average value of the beam power experienced by an incremental area of pipe surface is equal to the power density of the beam only when I is 90. Hence, it is preferable that the average value of I be as small as possible and less than 90. In the case of cylindrical pipes, this may be accomplished by both restricting the beam width to less than the diameter ofa single pipe and distributing a beam substantially equally between two adjacent pipes. In the case of pipes having linear surfaces, this may be done by disposing the pipes obliquely to the beam direction; restricting the beam width to less than one-half of the breadth of that respective portion of a single pipe which (i.e., the portion) intercepts the beam; and causing the beam to be intercepted by two adjacent pipes in substantially equal amounts.

The variation in beam power per unit, or incremental, area over the anode pipe surface 53 impinged by the beam 26 is attributable to the variation in the width of the respective beam portions '72, 74, 76, and hence, the quantity of beam power, impinging upon the several incremental areas 62, 64, and 66, respectively, of the anode pipe surface 56. The beam portion 76 impinging upon an incremental area 66 of the anode pipe surface in the region of the point ofjuncture 46 is narrower in width and, therefore, less in total power, than are the other beam portions 72 and 74 impinging upon the other incremental areas 62 and 64, respectively, which are more removed from the point ofjuncture 46. As an incremental area 62, 64 of anode pipe surface is increasingly removed from the point ofjuncture 46, the width of the respective beam portion 72, 74 and, therefore, the total power, impinging on that incremental area is greater, but still less than the beam power density.

This variation in the width of the respective beam portions 72, 74, 76 impinging upon the several incremental areas 62, 64, 66, respectively, is brought about by the variation in the angle of incidence 1 I ofthe several portions '72, 74, 76 of the electron beam 20 upon the several respective incremental areas 62, 64, 66 of the anode pipe surface 56. The variation in the angle of incidence is brought about by the nonlinearity of the anode surface. In the case of pipes having linear surfaces, the pipes must be disposed obliquely with respect to the beam direction in order to achieve these results.

The angle of incidence D of a portion 76 of the beam 20 on an incremental area 66 of an anode pipe surface 58 in the region of point 46 is smaller than the angle of incidence D,, 1 of other portions '72, '74 of the same beam 20 on those incremental areas 62, 64 more removed from the point 46. Hence, the smaller the angle of incidence between an incremental area of anode surface and that portion of an electron beam impinging thereon, the smaller will be the total power to which that incremental area is subjected.

Only a small number of incremental areas 62, 64, 66 on only one pipe 50 and a corresponding number of beam portions 72, 74, 76 constituting only one'half of a beam 20 have been illustrated in FIG. 6 for purposes of simplicity.

Utilization of the present invention to distribute an electron beam over an anode surface area significantly greater than the cross-sectional area of the beam also results in an increase in both the anode internal surface area in contact with the coolant and the total effective surface area of the anode. This is because of the relatively thin, substantially uniform pipe wall. As a result, there is not only a lower power density per unit area of the impinged anode surface, but a more rapid transfer of heat through the anode wall and away from the anode internal surface, as well. Further advantages to be gained by the use of the above anode structure include extended anode life and increased tube power-handling capacity.

As stated above, it is preferred than an electron beam be obliquely intercepted in substantially equal amounts by each of two adjacent pipes in the region of theirjuncture. However, where a beam has a width significantly less than the breadth of a single pipe, such a beam may be obliquely intercepted either by a single pipe, only, or by two adjacent pipes in unequal portions with satisfactory heat transfer from the anode. In these situations, however, best results are achieved where the interception of the beam is at an oblique angle and is limited to the general vicinity of the point ofjuncture: of two adjacent pipes so that the angle ofincidence is minimal.

It may be seen that this invention provides an increased effective surface area and an increased surface area for the interception of electrons, as well as an increase in the surface area in contact with a circulating coolant. As a result, there is, inter alia, a significant improvement in performance for electron discharge tubes embodying the present invention. Also, the invention provides these improved results with the utilization ofpipes of relatively simple design.

I claim:

II. An electron discharge device comprising:

a. a plurality of cathodes for emitting electrons;

b. means for focusing said electrons into a plurality of beams;

c. an anode comprising a plurality of contiguous pipes having substantially coextensive internal and external surfaces, said pipes being disposed in relation to said cathodes such that a portion of the external surface of each of only a pair of mutually adjacent ones of said pipes obliquely intercepts a portion of one of said electron beams; and

d. means for circulating a coolant through said plurality of pipes.

An electron discharge device comprising:

a cathode for emitting electrons;

. means for focusing said electrons into a beam;

an anode comprising a plurality of contiguous pipes having substantially coextensive internal and external surfaces, said pipes being disposed in relation to said cathode such that a portion of the external surface of each of only a pair of mutually adjacent ones of said pipes faces said cathode and obliquely intercepts a portion of said electron beam; and

d. means for circulating a coolant through said plurality of pipes.

o at

3. The electron discharge device defined in claim 2, wherein the juncture of said mutually adjacent pair of pipes coincides with the center line of said electron beam such that substantially one-half of said electron beam is intercepted by each of said external surface portions.

4. The electron discharge device defined in claim 3, wherein the breadth of each of said external surface portions is greater than one-half of the width of said electron beam.

5. The electron discharge device defined in claim 2, wherein each one of said mutually adjacent pair of anode pipes is hollow and has a relatively thin continuous wall of substantially uniform dimension.

6. The electron discharge device defined in claim 2, wherein each of said external surface portions is nonlinear in the transverse direction.

7. The electron discharge device defined in claim 6, wherein each one of said mutually adjacent pair of pipes is of circular cross section.

8. The electron discharge device defined in claim 6, wherein each one of said mutually adjacent pair of pipes is of elliptical cross section.

9. The electron discharge device defined in claim 2, wherein each one of said mutually adjacent pair of pipes is of substantially triangular cross section.

10 The electron discharge device defined in claim 2, wherein each one of said mutually adjacent pair of pipes is of substantially rectangular cross section.

11. The electron discharge device defined in claim 2, wherein said plurality of pipes are disposed in a circular array.

12. The electron discharge device defined in claim 2, wherein said plurality of pipes are disposed in a linear array.

13. An improved electron discharge device of the type having:

a. a cathode for emitting electrons;

b. means for focusing said electrons into a beam;

0. an anode for collecting said electrons; and

d. means for circulating a coolant through said anode;

wherein the improvement comprises:

said anode including a plurality of contiguous pipes, each of only a pair of mutually adjacent ones of said pipes having an electron-collecting surface portion obliquely intercepting said electron beam.

14. The electron discharge device defined in claim 13, wherein each one of said mutually adjacent pair of pipes has smooth and substantially coextensive internal and external surfaces.

Claims (14)

1. An electron discharge device comprising: a. a plurality of cathodes for emitting electrons; b. means for focusing said electrons into a plurality of beams; c. an anode comprising a plurality of contiguous pipes having substantially coextensive internal and external surfaces, said pipes being disposed in relation to said cathodes such that a portion of the external surface of each of only a pair of mutually adjacent ones of said pipes obliquely intercepts a portion of one of said electron beams; and d. means for circulating a coolant through said plurality of pipes.

2. An electron discharge device comprising: a. a cathode for emitting electrons; b. means for focusing said electrons into a beam; c. an anode comprising a plurality of contiguous pipes having substantially coextensive internal and external surfaces, said pipes being disposed in relation to said cathode such that a portion of the external surface of each of only a pair of mutually adjacent ones of said pipes faces said cathode and obliquely intercepts a portion of said electron beam; and d. means for circulating a coolant through said plurality of pipes.

3. The electron discharge device defined in claim 2, wherein the juncture of said mutually adjacent pair of pipes coincides with the center line of said electron beam such that substantially one-half of said electron beam is intercepted by each of said external surface portions.

4. The electron discharge device defined in claim 3, wherein the breadth of each of said external surface portions is greater than one-half of the width of said electron beam.

5. The electron discharge device defined in claim 2, wherein each one of said mutually adjacent pair of anode pipes is hollow and has a relatively thin continuous wall of substantially uniform dimension.

6. The electron discharge device defined in claim 2, wherein each of said external surface portions is nonlinear in the transverse direction.

7. The electron discharge device defined in claim 6, wherein each one of said mutually adjacent pair of pipes is of circular cross section.

8. The electron discharge device defined in claim 6, wherein each one of said mutually adjacent pair of pipes is of elliptical cross section.

9. The electron discharge device defined in claim 2, wherein each one of said mutually adjacent pair of pipes is of substantially triangular cross section.

10. The electron discharge device defined in claim 2, wherein each one of said mutually adjacent pair of pipes is of substantially rectangular cross section.

11. The electron discharge device defined in claim 2, wherein said plurality of pipes are disposed in a circular array.

12. The electron discharge device defined in claim 2, wherein said plurality of pipes are disposed in a linear array.

13. An improved electron discharge device of the type having: a. a cathode for emitting electrOns; b. means for focusing said electrons into a beam; c. an anode for collecting said electrons; and d. means for circulating a coolant through said anode; wherein the improvement comprises: said anode including a plurality of contiguous pipes, each of only a pair of mutually adjacent ones of said pipes having an electron-collecting surface portion obliquely intercepting said electron beam.

14. The electron discharge device defined in claim 13, wherein each one of said mutually adjacent pair of pipes has smooth and substantially coextensive internal and external surfaces.

Images