US3201640A

US3201640A - Electron gun in the form of a multipactor

Info

Publication number: US3201640A
Application number: US178129A
Authority: US
Inventors: Philo T Farnsworth
Original assignee: Deutsche ITT Industries GmbH
Current assignee: TDK Micronas GmbH; International Telephone and Telegraph Corp
Priority date: 1962-03-07
Filing date: 1962-03-07
Publication date: 1965-08-17
Anticipated expiration: 1982-08-17
Also published as: GB1030708A; NL289897A; DE1464682A1; BE629274A

Description

Aug- 17, 1965 P. T. FARNswoRTH 3,201,640

ELECTRON GUN IN THE FORM OF A MULTIPACTOR Filed March 7, 1962 2 Sheets-Sheet 1 Aug- 17, 1955 P. T. FARNswoRTH 3,201,640

ELECTRON GUN IN THE FORM OF A MULTIPACTOR Filed March '7, 1962 2 Shee'lzs-Sheei'I 2 l l l I f f f United States Patent @ffice- 3,2%,549 Patented Aug. 17, 1965 3,201,646) ELECTRGN 'GUN EN THE FRM 0F A MULTEPACTGR Philo T. Farnsworth, Fort Wayne, Ind., assigner to Intel'- national Teiephone and Telegraph Corperation, Nutley, NJ., a corporation of Maryland )Filed Mar. 7, 1962, Ser. No. 178,129 7 Claims. (Cl. S15-12) The present invention relates to an electron gun, and more particularly to an electron gun for use in a cathode ray or similar tube and having high perveance and mutual conductance characteristics.

Conventional electron guns comprise a thermionic cathode as a source of electrons, a control electrode for varying beam current, and a set of focusing electrodes for focusing the beam. The perveance of such guns is dependent upon the electron-emitting capabilities of the cathode inasmuch as perveance is defined in the general case as beam cui-ent divided by the three-halves power of the laccelerating anode voltage. The mutual conductance of such guns is also dependent upon the electronemitting capabilities of the cathode, since it is defined as the change in beam current obtained by a given change in voltage on the control electrode. Thus perveance and mutual conductance have values which are limited by cathode eiiciency.

In the present inven-tion, high values of perveance and mutual conductance are obtained by providing an electron emitter of exceptionally high eiiiciency whereby an extremely high density beam is generated in the first instance for control by the control and accelerating electrodes. This electron emitter takes the form of an electron multiplier or multipactor wherein a high intensity electron space current is generated and controlled in such a manner as to form a high density electron beam of small cross-section.

It is therefore an object of this invention to provide an electron gun of unique construction which provides high values of perveance and mutual conductance.

It is another object of this invention to provide an electron gun for a cathode ray tube in which the electron emitter takes the form of an electron multiplier or multipactor wherein a portion of an intense electron space current generated in the multiplier is utilized and formed into a pencil-like electron beam of high density and small cross-section.

Other objects will become apparent as the description proceeds.

In the accomplishment of the present invention, there is provided an electron gun comprising a cathode having an extended active surface capable of emitting secondary electrons a-t a ratio greater than unity, an annular anode which is spaced from and disposed opposite said active surface, an electrode disposed on the side of said anode opposite said cathode and having an aperture therethrough, the area of the active surface being larger than the area of the aforementioned aperture, means including the cathode, the anode and the aforesaid electrode for converging electrons from the cathode active surface toward the aforesaid electrode in the vicinity of said aperture, means electrically connecting the cathode and electrode together, means for imposing an oscillating potential between the anode and the cathode and electrode of a frequency having a half period approximately equal to electron transit time (or an odd multiple of this frequency) between the cathode and electrode, the oscillating potential being of such magnitude as to impart to electrons a velocity suicient to cause secondary emission from said active surface at a greater than unity ratio and to cause electrons to penetrate through the aforesaid aperture, and a control element having an aperture therethrough in registry with the rst-mentioned aperture, the control element being disposed on the side of said electrode opposite said anode.

The above-mentioned and other features and objects of this invention and the manner of attaining them will become more apparent and the invention itself will be best understood by reference to the following description `of an embodiment of the invention taken in conjunction with the accompanying drawings, wherein:

FG. l is a longitudinal sectional illustration of one embodiment of this invention;

FIG. 2 is a similarillustration of another embodiment of this invention;

FIG. 3 is a geometric illustration of the electron optics and is used in explaining the operating principles thereof; and

FIG. 4 is a graph used in explaining the operation of the invention.

Referring to the drawings, and more particularly to FIG. l, an evacuated glass envelope 1, shown in part section, is of the conventional shape employed in cathode ray display tubes. In the neck portion 2 of the envelope is mounted an electron gun, this gun comprising, in general, a subassembly for generating and emitting electrons, a control electrode, and focusing and accelerating electrodes.

The electron-emitting subassembly is, in the general sense, an electron multiplier or multipactor and comprises a cathode or electrode 3 mounted in the left-hand end of the envelope neck 2, an anode ring 4 mounted opposite the cathode 3, and an end electrode 5 mounted on the side of the anode 4 opposite the cathode 3. All of these parts are symmetrically formed about the tube axis 6. Cathode 3 has an active surface 7 which is spherically concave inwardly toward the anode 4 and has its center of curvature on the axis 6. The cathode sould be fabricated of a material having a secondary emission ratio of greater than unity, beryllium copper being an example.

The anode 4 is formed of metal and is securely mounted in the neck 2 of the tube envelope. It is provided with a frusto-conically shaped opening 8 which is coaxial about the axis 6, the larger diameter end 9 facing the cathode 3 and the smaller diameter end 1t) facing the electrode 5.

The electrode 5 is a metallic disc securely mounted in the neck 2 and is provided with a centrally projecting convex portion 11 having a coaxial aperture 12 therethrough. This electrode S may be formed of a metal having a secondary emission greater than unity; however, as will appear from the following description, this emission characteristic is not absolutely necessary.

The remaining elements in the electron gun to the righthand side of the electrode 5 are conventional, the first one being a control electrode 13 secured in the neck 2 and having a coaxial limiting aperture 14 therethrough. Adjacent to this is mounted an accelerating anode structure 15 having a pair of limiting apertures i6.

To `the right-hand side of the anode 15 'may be mounted the usual additional electrodes utilized in electron guns of cathode ray tubes.

' The cathode 3 and electrode 5 are electrically connected together through a resonant tank circuit 13 having the center tap grounded as shown. A coil 19 constitutes a part of a radio frequency voltage input circuit and is coupled directly to the tank circuit 18 as shown.

A. power supply or battery 2@ has its positive terminal connected to the anode 4 and the negative terminal.

grounded as shown.

The control electrode 13 has the usual biasing potentiais applied thereto and also means for coupling an input signal. A biasing battery 2i has its positive terminal grounded as shown and a potentiometer 22 connected thereacross which is coupled to the control electrode 13. An accelerating anode battery or power supply 23 has its positive terminal connected to the anode and the negative terminal grounded.

In operation, an oscillatory electron space current is developed between the two end plates or cathode 3 and electrode 5 of the electron multiplier or multipactor section of the electron gun. A certain portion of this space current which occupies the conically shaped region between the two electrodes 3 and 5 penetrates through the aperture l2 and preferably is focused onto a cross-over point in the aperture 14 of the control electrode 13. From this point forward, the electrons are formed into a pencil-like beam by the apertures i4 and 16 in the usual manner. Inasrnuch as the formation and control of the beam after it leaves the control electrode 13 is conventional, only that portion of the structure to the left-hand side of the control electrode 13 needs to be described and elaborated further.

As already explained, operation of the multipactor section of the electron gun is based upon electrons oscillating back and forth between the cathode 3 and electrode 5 and releasing additional electrons by repeated impacts with the cathode 3. If the electrode 5 is fabricated of material having a secondary emission ratio greater than unity, impacts of electrons with this electrode will also result in the creation of secondary electrons.

The impacts are with the cathode 3 and in part with the electrode 5, and the multiplication occurs by secondary electron emission from either the cathode 3 or electrode 5 or both. A high frequency potential, which may be of the order of SO-megacycles, is applied between the cathode 3 and electrode 5, this potential being preferably relatively small as compared with the direct potential of the battery 2l) on the anode 4. Under the influence of this potential, electrons present in the tube strike at least the active surface 7 of the cathode 3, liberating secondary electrons, which are accelerated toward the opposite electrode 5 by the anode potential 20. If the potential of the latter is so related to the frequency applied to the cathode 3 by means of the input coil 19, that the released electrons traverse the space in time to be `accelerated by the oscillating potential on the opposite electrode S, a further impact and release of secondaries will occur, and if the product of the two secondary emission ratios of the two electrodes 3 and 5 is greater than one, a multiplication will take place which will increase until the number of electrons released at each impact is equal to the number collected by the anode 4, or until the process is stopped by changing the anode potential or removing the oscillating signal applied to the cathode 3 via the coil i9. Inasmuch as the electrode 5 is provided with a coaxial aperture 12, it is obvious that electrons approaching this aperture which would otherwise strike the electrode 5 will penetrate through the aperture and pass onward to the control electrode 13. However, in the discussion thus far, the electrode S can be considered as being solid for the purpose of determining what happens between this electrode .and the cathode 3.

In further explanation of the multiplication action, reference is made to FIG. 4, wherein the .anode to cathode voltage is represented by the graph. The straight line 24 represents the anode potential as .applied by the battery 20. The sine wave Z5 represents the oscillating potential applied between the electrodes 3 and 5 via the coil 19. As is obvious, the oscillating potential 25 as superimposed upon the constant potential 24 provides a varying potential between the anode 4 and each of the electrodes 3 and 5.

If it is assumed at the start that an electron initially leaves the electrode 5 and starts its transit toward the anode 4 and cathode 3 at the same instant as the starting point 26 of the oscillating potenttial 25, it is seen that this oscillating potential applied between the electrodes 3 and 5 accelerates the electron over and beyond that which is attributable to the steady-state potential 24 on anode 4. If it is assumed that the transit time of this electron between the electrodes 3 and 5 and attributable to the anode 24 potential is substantially equal to the time of a half period of the sine wave 25, the electron will strike the active suriace 7 of the cathode 3 with enough force so as to liberate secondary electrons. The secondary electrons then travel backwardly toward the anode 4 and electrode 5, reaching the electrode 5 at the end of the next half period, following which the electrons are again returned toward the cathode 3 to impact the same so as to liberate more secondaries. This action continues until the number of electrons multiplies to an equilibrium value as determined, in general by space charge at the cathode 3 and electrode S which prevents all but the faster secondary electrons from escaping therethrough. Thus, when the number of electrons collected by anode 4 equals the number of secondaries generated, the space current between the electrodes 3 and 5 becomes stabilized. It is thus seen that the space current in the chamber between the two electrodes 3 and 5 builds to an unusually high value in comparison to the current iowing to the lanode 4.

Obviously, it is necessary that the electrons make a repeated number of tri s between the two electrodes 3 and 5 before they are collected by anode 4. In order to accomplish this, it is necessary to guide the electrons along paths which will not intercept the anode 4, and in the illustrated embodiment, this guiding is provided by means of electron-optical forces set up between the electrodes 3 and S on the one hand and the anode 4 on the other hand. The action of the electron optics is best explained by reference to FIG. 3, wherein the

electrodes

3, 4 and 5 are geometrically illustrated. Point A is chosen as the starting point of an electron emitted from the cathode surface 7. This point A is chosen on the periphery of the cathode surface 7 for the reason that it represents an external ray and one for which the probability of striking anode 4 is greatest. This electron is accelerated along the radius AC of the active surface 7. A divergent lens at the opening 9 of the anode cone 8 refracts the electron away from the cone axis so that it leaves the cone along the line BD. This lens is divergent, because the electron moves from Ia region of high gradient into a region of lower gradient.

There is another divergent lens located at the small end 10 of the cone whereby the electron undergoes further refraction. However, the effect of this lens is small so that it may be regarded as negligible in the diagram of FIG. 3.

At D the electron is reflected back into the anode aperture 8 along a line DE, and at point E it is again bent away from the axis by the lens action and travels along a path EF.

The electron path EF is not quite along a radius. There is a small component of non-radial velocity directed toward the axis of the cone. The electron is reflected from the cathode 3 but retains the non-radial component of velocity.

The successive passages of the electron progressively approach the cone axis, with the electron acquiring an additional small amount of transverse velocity at each passage until the electron path crosses the axis. After the path has progressed to the opposite side of the axis, successive passages decrease rather than add to the tranverse component of velocity, and when the elect-ron path reaches the edge of the cathode 3 opposite to the point A at which it originated, there will again be no transverse component of velocity and the electron behaves just as though it had originated from that side.

However, the space charge which develops when the number of released electrons becomes very large drives the peripheral electrons, i.e., the electrons more remote from Ithe center of the cloud traversing the chamber, toward the anode, making their collection by the anode possible. Thus it is seen that by providing the electron optics as just explained in connection with iFIG. 3, the electron can be guided between the two electrodes 3 and 5 for a repeated number of trips before they eventually will be collected by the anode 4. This results in the multiplication of electron current which increases with each impact of an electron with the active surface 7 of the cathode 3. This multiplication process continues until the number of electrons collected -by the anode 4 equals the number of secondaries being emitted from the active surface 7.

Up to Ithis point, it has been assumed in the general case that the electrode 5 is solid and that the aperture 12 iS non-existent. However, the aperture 12 is present such that it is necessary to examine what happens inside the chamber between the twoelectrodes 3 and 5 in order to obtain .the results as already described.

The electrons in the beams formed between the two diametrically opposite electrodes 3 and 5 have periods of oscillation and t-ransit time which are related to the D.C. anode voltage and the frequency of the radio -frequency voltage applied between the cathode 3 and the electrode 5, the frequency of this voltage being adjusted to correspond. The electrons leaving either the cathode 3 or electrode S at one instant of time absorb power from the field established by the radio frequency voltage andthe oneS leaving at another instant of time give up energy to the field. In the former instance, the electrons are said to be in-phase with .the applied voltage, whereas in the latter instance they are out-of-phase.

The electrons oscillate between the two electrodes 3 and 5 at approximately the same frequency as the radio frequency `and they make one trip between the electrodes in a half period of the radio frequency. The fact that the electrons are so accelerated insures that they strike the 'active surface .-7. The returning electrons also make the trip in slightly less than one-half cycle. The result is that the fastest electrons very rapidly get out of phase with the radio frequency voltage.

The process results in electrons being fed from the multiplication phase (in-phase) into the oppos-ite phase wherein an electron is decelerated rather than accelerated by the radio frequency. Electrons in this phase do not strike the electrodes 3 and 5 at all, but continue to oscillate between them, delivering energy to the external .circuit as they are slowed down by the radio frequency. This 'action is a phenomenon well known in the art of elect-ron multipliers of the multipactor type, typical multipactors being disclosed in my Patents Nos. 2,189,358, 2,107,782 and 2,071,515.

The electrons which are in-phase have .an oscillatory swing which carries for a distance greater than that between the two facing electrodes 3 and 5 such that .the electrons impact the active surface 7 with enough force t0 eject secondaries and to provide electron multiplacation; however, the out-of-phase electrons will be decelerated and will never quite reach the electrode surfaces.

The decelerated electrons do not -strike the electrodes 3 and 5 but continue to oscillate between them and in doing so form a space charge which results in potential minima in front of the facing surface, the position of which being indicated by the dashed

liners

27 and 28, respec-tively. These potential minima or surfaces `are established by that group of decelerated electrons which oscillate over a path length slightly shorter than the distance between the electrodes 3 yand 5, the two

surfaces

27 and 28 thereby acting as electron mirrors. Other groupsof ele-ctrons which do not have as great an oscillatory swing will be reflected inwardly of the

surfaces

27 and 28.

Thus, for those electrons having an oscillatory swing between the Itwo surfaces 27 and 28, the electrode 5 will appear as a solid plate. However, those electrons having a swing sufficient to impact the active surface 7 and also to penetrate the surface 28 will pass through the aperture 12 and will converge into `the aperture 14 of the control elect-rode 13. From this point forward, the electrons are collected into a pencil-like beam and are handled conventionally by the various accelerating, focusing and deflecting 4electrodes in a .cathode ray tube.

Since the space current build-up in the multipactor between the two electrodes 3 and 5 is intense, it is at once obvious that the electron beam which penetrates the aperture 12 yand converges into the opening 14 is of extremely high density. The electron gun, therefore, possesses large values of both perveance land mutual conductance which constitute one of the objects achieved by this invention.

A second and preferred embodiment of the invention is illustrated in FIG. 2 wherein like numerals indicate like parts. 1n this figure, the multipactor composed of the cathode 3, the anode 4 and the electrode 5 are encased in .a resonant cavity .as generally indicated by the numeral 29. The cavity 29 preferably is of cylindrical conguration and has two opp-osite end plates 3u and 31 which are spaced apart and parallel. The cavity 29 is of course made out of metal such that it is electrically conductive from one end to the other. The cathode 3 is conductively mounted securely in the end plate 30, and the electrode 5 is similarly mounted in the end plate 31. By this means, the Acathode 3 and the electrode 5 are electrically `connected together. They are otherwise spaced apart and disposed with respect to each other .and the anode 4 the same as already described in connection with the embodiment of FIG. 1.

The anode 4 is supported inside the resonant cavity 29 by means of an annular support 32 of insulating materlal. A wire 33 feeds through the end plate 3) and is soldered to the inside wall ofthe cavity 29 for the purpose of applying a` radio frequency voltage to the cavity 29.

. The multipactor portion of the elec-tron gun operates the same as that already described in connection with FIG. 1. The only difference in operation resides in the fact that this multipactor of FIG. 2 is excited by means of the oscillating field set -up inside the resonant cavity 29, the cavity itself being excited by loop 33. The intense beam emerges from the aperture 12 and is concentrated in the region of the aperture 14 of the control electrode 13, thereby producing a high density beam of small crosssect-ion already described.

While I have described above the principles of my invention in connection with specific apparatus, it is to =be cle-arly understood that .this description is made only by Way of example and not as a limitation to the scope of my invention.

What is claimed is:

1. Invan electron gun comprising an envelope, a pair of spaced-apart electrodes therein, a first of said electrodes being adapted to emit secondary electrons on impact by a primary electron, a second of said electrodes having an aperture through which electrons may flow, means for converging said electrons toward said second electrode in the vicinity of said aperture, said means including a concave surface on said first electrode, an intermediate electrode having a conically shaped opening tapering toward said aperture, an oscillatory circuit connected to said pair of electrodes, means for supplying alternating current to said oscillatory circuit to cause said electrons to oscillate between said pair of electrodes and to make repeated secondary electron-generating impacts with said first electrode and cause a portion of said electrons to penetrate said aperture.

2. In an electron gun, a cathode having an extended active surface capable of emitting secondary electrons at a ratio greater than unity, an electron permeable anode having a first aperture and disposed opposite said cathode, an electrode disposed on the side of said anode opposite said cathode and having a second aperture therethrough, the area of said active surface being larger than the area of said second aperature, means including said cathode, said anode and said electrode for converging electrons from said active surface toward said electrode in the vicinity of said second aperture and means applying a positive potential to said anode with respect 'to said cathode and said electrode for oscillating electrons between said cathode and said electrode, means for imposing an oscillating potential between said cathode and said electrode of a frequency having a half period approximately equal to electron transit time between said cathode and electrode, said oscillating potential being of such magnitude and frequency as to impart to electrons a velocity sufficient to cause secondary emission from said active surface at greater than unity ratio and to cause electrons to peneterate through said second aperture.

3. In an electron gun, a cathode having an extended active surface capable of emitting secondary electrons at a ratio greater than unity, an anode of annular shape positioned opposite said active surface, an electrode disposed on the side of said anode opposite said cathode and having an aperture therethrough, said active surface being concave facing said anode and having a center of curvature on the axis of said anode, said electrode being convex facing said anode with said aperture coaxial with said axis, said anode having a frusto-conically shaped second aperture wherein the larger diameter portion is adjacent to said active surface and the smaller diameter portion is adjacent to said electrode, said cathode surface and anode and electrode apertures having progressively diminishing areas; constituting an electron lens which converges electron flow from said active surface 'toward said electrode in the vicinity of said aperture, and a control element disposed on the side of said electrode opposite said anode, said control element having a beamforming aperture coaxial with said axis.

4. In an electron gun, a cathode having an extended active surface capable of emitting secondary electrons at a ratio greater than unity, an anode of annular shape positioned opposite said active surface, an electrode disposed on the side of said anode opposite said cathode and having an aperture therethrough, said active surface being spherically concave facing said anode and having a center of curvature on the axis of said anode, said electrode being convex facing said anode with said aperture coaxial with said axis, said anode having a frusto-conically shaped inner periphery wherein the larger diameter portion is adjacent to said active surface and the smaller diameter portion is adjacent to said electrode, said cathode, anode and electrode constituting an electron lens which converges electron flow from said active surface toward said electrode in the vicinity of said aperture, means for applying a constant positive potential to said anode with respect to said cathode and said electrode for oscillating electrons between the latter, and means for imposing an oscillatory potential between said cathode and electrode of a constant frequency having a half period approximately equal to electron transit time therebetween, said oscillating potential being of such magnitude and frequency as to impart to electrons a velocity sufficient to cause secondary emission from said active surface at a greater than unity ratio and to cause electrons to penetrate through said aperture.

5. In an electron gun comprising an envelope, a cathode having an extended active surface capable of emitting secondary electrons at a ratio greater than unity, an anode of annular shape positioned opposite said active surface, an electrode disposed on the side of said anode opposite said cathode and having an aperture therethrough, said active surface being spherically concave facing said anode and having a center of curvature on the axis of said anode, said electrode being convex facing said anode with said aperture coaxial With Vsaid axis, said anode having a frustoconically shaped inner periphery wherein the larger diameter portion is adjacent to said active surface and the smaller diameter portion is adjacent to said electrode, said cathode, anode and electrode constituting an electron lens which converges electron ow from said active surface toward said electrode in the vicinity of said aperture, a resonant cavity having conductive Walls with opposite ends closed within the envelope, said cathode and said electrode being mounted in said ends, respectively, and electrically connected together, said anode being mounted inside said cavity and being electrically insulated therefrom, means for applying an oscillatory potential to said resonant cavity to cause said electrons to oscillate between said ends and to form a space charge, and means for applying a constant positive potential to said anode with respect to said cathode and electrode, said positive potential being of a magnitude which, in combination with said oscillatory poential, causes said electrons to oscillate over path lengthsv slightly greater than the distance between said cathode and electrode thereby producing secondary emission from said cathode and causing a portion of said electrons to penetrate through said aperture in beam form.

6. In an electron gun comprising an envelope, a cathode having an extended active surface capable of emitting secondary electrons at a ratio greater than unity, an anode of annular shape positioned opposite said active surface, an electrode disposed on the side of said anode opposite said cathode and having an aperture therethrough, said active surface being spherically concave facing said anode and having a center of curvature on the axis of said anode, said electrode being'convex facing said anode with said aperture coaxial with said axis, said anode having a frusto-conically shaped inner periphery wherein the larger diameter portion is adjacent to said active surface and the smaller diameter portion is adjacent to said electrode, said cathode, anode and electrode constituting an electron lens which converges electron flow from said active surface toward said electrode in the vicinity of said aperture, a resonant cavity having conductive walls of cylindrical shape enclosed within said envelope with opposite ends which are substantially flat and parallel, said cathode and electrode being mounted in said opposite ends and electrically connected to said cavity, said anode being mounted inside said cavity and electrically insulated therefrom, means for applying a constant oscillating potential to said resonant cavity to cause said electrons to oscillate between said ends and form a space charge, and means for applying a constant positive potential to said anode with respect to said cathode and electrode, said positive potential being of magnitude which, in combination with said oscillatory potential, causes said electrons to oscillate over path lengths slightly greater than the distance between Ysaid cathode and electrode thereby producing secondary emission from said cathode and causing a portion of said electrons to penetrate through said aperture in beam form.

'7. The device of claim 2 wherein said electrode includes a secondary emissive surface emitting further electrons.

References Cited bythe Examiner UNITED STATES PATENTS 2,240,713 5/41 Orthuber et al 315-12 2,416,303 2/47 Parker 315-12 2,817,033 12/57 Brewer 313-821 DAVID G. REDlNBAUGH, Primary Examiner.

ROY LAKE, Examiner.

Claims

1. IN AN ELECTRON GUN COMPRISING AN ENVELOPE, A PAIR OF SPACED-APART ELECTRODES THEREIN, A FIRST OF SAID ELECTRODES BEING ADAPTED TO EMIT SECONDARY ELECTRONS ON IMPACT BY A PRIMARY ELECTRON, A SECOND OF SAID ELECTRODES HAVING AN APERTURE THROUGH WHICH ELECTRONS MAY FLOW, MEANS FOR CONVERGING SAID ELECTRONS TOWARD SAID SECOND ELECTRODE IN THE VICINITY OF SAID APERTURE, SAID MEANS INCLUDING A CONCAVE SURFACE ON SAID FIRST ELECTRODE, AN INTERMEDIATE ELECTRODE HAVING A CONICALLY SHAPED OPENING TAPERING TOWARD SAID APERTURE, AN OSCILLATORY CIRCUIT CONNECTED TO SAID PAIR OF ELECTRODES, MEANS FOR SUPPLYING ALTERNATING CURRENT TO SAID OSCILLATORY CIRCUIT TO CAUSE SAID ELECTRONS TO OSCILLATE BETWEEN SAID PAIR OF ELECTRODES AND TO MAKE REPEATED SECONDARY ELECTRON-GENERATING IMPACTS WITH SAID FIRST ELECTRODE AND CAUSE A PORTION OF SAID ELECTRONS TO PENETRATE SAID APERTURE.