US2376707A - Space discharge device - Google Patents

Description

May 22, 19 45. R. E. MCCOY SPACE DISCHARGE DEVICE Filed April 5 1941 3 Sheets-Sheet 1 ROBERT E. M CDY INVENTOR.

BY ATTORNEY.

May 22, 1945. R C 2,376,797 v SPACE DISCHARGE DEVICE Filed April 5, 1941 5 Sheets-Sheet 2 FUBEIQT E. MCEY INVENTOR.

BY 6 Z ATTORNEY.

y 1945- R. E. M coY 2,376,707

SPACE DISCHARGE DEVICE Filed April 5, 1941 s Sheets-Sheet s FDfiEET E. M CUY INVENTOR.

' 2. 0 5 ATTORNEY.

/.0 fig:

X TBHNS/T TIME 0 DE'FL EC T/O/V FREOUENC Y Patented May 22, 1945 UNITED STATES PATENT OFFICE 2,376,707 SPACE. DISCHARGE DEVICE. Robert E. McCoy, Portland, mg. Application April 5, 1941, Serial No. 387,029

14 Claims.

This invention relates to improvements in space-discharge devices; and more particularly, it relates to those in which the charged particles are caused to travel a considerable distance.

The principal object of my invention is to provide convenient and efficient means for the conversion of power from direct current to alternating current, or vice versa, which will function at very high frequencies. The invention is capable of even wider utility since it provides means for converting power from one frequency to another, or for the exchange of power among circuits operating with several different frequencies of alternating current and circuits operating with direct current. Moreover, it provides means whereby a small amount of power at one frequency can control a larger amount of power at the same frequency or another frequency related thereto; or a small amount of alternating current power can control a large amount of direct current power, or vice versa.

Since my invention depends largely upon the dynamic properties of an ion beam, it has a secondary object; namely, to improve the sensitivity and efficiency of the means whereby the motion of such a beam is controlled.

These and other objects will appear as my invention is more fully hereinafter described in the following specification illustrated in the accompanying drawings and finally pointed out in the appended claims.

In the drawings:

Figure 1 is a longitudinal, sectional elevation of a cathode-ray tube.

' Figure 2 is a sectional, detail view of an electron gun taken on the line 2-2 of Figure 1.

Figure 3 is a perspective view of an electron gun.

Figure 4 is a graph showing the variation of deflection sensitivity, and comparing the deflection produced by the two systems mentioned.

Figure 5 shows a number of patterns typical of those which would be traced on the screen of an oscilloscope whose deflection system had been replaced by that illustrated in Figure 1, if only one pair of deflection plates was energized.

Figure 6 is a graph showing the variation of the deflection sensitivity, as indicated by the major and minor axes of the elliptic patterns illustrated in Figure 5.

ings:

The preferred embodiment of my invention is shown in Figure 1. Its component parts are located in or near a cylindrical'glass tube 1 form- Referring now more particularly to the drawmg a discharge chamber. The whole consists of five overlapping zones, generally indicated as A, B, C, D, and E, whose functions are as follows:

Within the zone A, I provide an electron gun, generally indicated at 2, whose function is to discharge electrons throughout the length of the discharge chamber. Within the zone B, I provide means, generally indicated at 3, for deflect ing the electrons coming from the electron gun within the zone A. Within the zone C, I provide a deflection booster generally indicated at 4. Zone D is practically empty; but around the outside of the tube, I provide a simple armature -5. At the end of the tube in zone E, I provide a plurality of electrodes, generally indicated at 6, for collecting the electrons at the end of their passage through the discharge chamber. v

I shall now explain the operation of the device as a whole, treating each zone as a unit; then I shall describe the operation of each zone in detail. For the general explanation, fice to describe the results obtained by the elements of each zone, leaving the discussion of the means until later.

The electron gun 2 in zone A produces a beam of electrons which it shoots or projects along the axis of the tube into the zone B.

For convenience, the subsequent movements of the beam will be described by locating a typical electron in a system of polar co-ordinates. The co-ordinates used will be: the axial distanoe, measured from some arbitrary point, such as the end of zone B; the radial distance, measured from the beam to the nearest point on the axis; and the angle of the azimuth, which indicates the direction of the radial line in a plane perpendicular to the axis.

In zone B, transverse forces, alternating at a high frequency, deflect the beam. When the beam passes from zoneB to zone C, it has a small radial displacement. Successive electrons differ slightly in azimuth, since the deflection system is designed to make the azimuth rotate in synchronism with the alternation'of the deflecting forces.

In zone C, the radial displacement of the beam is increased. For this reason, zone C may be called a displacement magnifier, or a deflection booster. It makes possible the use of relatively small forces in the deflection zone B.

The beam leaves zone C, and travels through zone D at a substantially constant distance from the axisand close to the glass, along a path such as those indicated by the dotted lines.

Due to the rotation of the azimuth of the beam,

it will sufit is close to each armature conductor in turn. At one moment, it may be close to one conductor; a half-cycle (at deflection frequency) later, it will be close to an opposite conductor. By electromagnetic induction, the periodic changes in beam position generate alternating voltages in the armature conductors.

By the same process of electromagnetic induction, alternating currents in the armature conductors induce alternating electric fields within the tube. The beam, by its changes in position, commutates the alternating fields into the equivalent of a D. C. voltage in series with the beam.

Thus a transfer of power takes place from beam to armature circuits, or vice versa.

The beam then proceeds into zone E, where the electrons are collected by one of the electrodes, and conducted through the external D. C. circuits back to their starting point.

The electrical characteristics of my invention are very much like those of a more familiar device known variously as a synchronous converter, a rotary converter, or a double current generator.

To make the analogy complete, the mechanical type converter should be deprived of the usual stationary field magnet; it should have separate A. C. and D. C- windings on the armature; and the relative motion of commutator and brushes should be produced by a small two-pole synchronous motor.

In the mechanical verter:

Direct current flows from In the cathode-ray con verter:

Direct current flows from an external circuit, through one brush and commutator segment, then through the I) C. armature coils, to another commutator segment and brush; thence it returns to the external circuit.

The relative motion of brushes and commutator an external circuit, causing electrons to move from the cathode, through the defied tion zone B, then through zone D, to the anode in zone E; thence the electrons return to the external circuit.

The changing azimuth of the deflection changes the path of the electrons through zone D. There are bars changes the path of the D. C. through the armature coils. There may be several dozen difierent commutator bars, and a corresponding number of paths through the D. C. armature.

The frequency with which the D. C. returns to the same path is the same as the frequency of A. C. in the small motor.

In either case, the direct current produces a magnetic field which, by its rotation relative to the A. C. armature conductors, generates alternating voltages therein.

Alternating currents in the A. C. armature coils of the mechanical device induce alternating voltages in the adjacent D. C. armature coils. The relative motion of brushes and commutator converts these induced voltages into a direct voltage, so far as the external D. C. circuit is concerned.

Alternating currents in the armature conductors of the electronic device induce alternating voltage gradients in the space adjacent. The electrons moving through zone D integrate these induced gradients into a direct voltage, which is effectively in series with the beam, so far as the external D. C. circuit is concerned.

Power may be transferred in either direction from beam (D. C.) to armature (A. C.) or from armature to beamdepending on the phase rela-' tionships in the A. C. circuits.

If the .electron gun in zone A were of conventional design, such as those used in cathode-ray oscilloscopes or in television picture tubes, the possible power output would be small (because of the small beam current).

trillions of electrons in motion at one time, and a cor responding number of diiferent paths through zone D.

The frequency with which any path is repeated is the same as the frequency supplied to the deflector sys tem.

Because it provides a beam of higher current with no larger cross-section, I prefer to use the gun structure illustrated in zone A, Figure 1, and shown in more detail in Figures 2 and 3.

With reference to the electron gun, the emitting surface of the cathode 1 is shown horizontal in Figures 2 and 3. Flanking it on either side are auxiliary electrodes 8 and 9. Above the cathode (and therefore not shown in Figure 1) is an accelerating anode I0. All four of these electrodes 1, 8, 9, and ID, are situated between the poles II and [2 of magnet l3.

A second anode I 4, containing a long, narrow aperture I5, is placed perpendicular to the emitting surface 1, near the ends of electrodes 9 and ID.

A third accelerating anode l6 faces the second.

Perpendicular to most of the electrodes already mentioned is a fourth anode I! which contains a small aperture l8.

Facing the edges of electrodes [4, l 6', and l I are the poles of a second magnet l9, oriented substantially at right angles to the direction of the first magnet l3.

With reference to Figure 1, the apparatus shown in Figures 2 and 3 is located within the tube I with the flat surface of electrode l1 perpendicular to the axis. The aperture I8 in electrode 11 is centered on the axis.

Forwardly of the fourth anode H, but still in zone A, is a focusing anode 20. This consists of a hollow conducting cylinder, containing two apertured discs, 2! and 22. The apertures are situated on the axis of the tube, which is also the axis of the cylindrical portion of the focusing anode.

Emitting surface 7 and auxiilary electrodes 8 and 9 are at the same potential. Anode I0 is at much higher potential.

The electrostatic field, due to this potential difference, tends to accelerate electrons upward. In addition, because of the shape and inclination chosen for the auxiliary electrodes 8 and 9, the electric field tends to draw toward the middle these electrons which are near the right or left edge of the cathode I. This side-thrust is relatively small, but it plays an important part in compressing the electron stream.

The first magnet I3 produces a magnetic field which is parallel to the surfaces of electrodes 1, 8, and 9, but perpendicular to the electrostatic field.

Electrons emitted by the cathode 1 are accelerated upward by the electric field; but as soon as they start to move, the magnetic field deflects their paths sidewise. The resultant paths are shown in Figure 2 for electrons starting near the edges of the cathode, and for an electron starting near the middle.

The important characteristic to be noted is the convergence of the various paths, as they approach the second anode 14. The thickness of the electron stream, as it passes through aperture 15, is very much less than the widthof the emitting surface 1. The improvement in the concentration of the beam depends upon the shape of electrodes 8, 9, and ID, the position of aperture l5 relative to the cathode 1, and the ratio of the electric field intensity. This ratio can be controlled readily by adjusting the potentials of the electrodes.

Upon emerging from the aperture I 5, the beam will be in the form of a thin horizontal stream, which is still wide in the direction normal to the paper in Fig. 2. Electrodes 14, I6, and I1, together force in its ownin single file, when 11th the second magnet is, compress the beam in his direction, and rotate it in a manner which :orresponds. closely to that employed by electrodes Hi, and i together with the first magnet P31.

As a result of the two-stage compressionof the seam, it emerges from the far side of aperture H3 in the form of a slender stream. Its cross section is still approximately the same shape as the surface of the cathode, but the cross-sectional area is perhaps a hundred-fold smaller.

Theoretically, if, the electrons left the cathode with no initial velocities, and if they did not repet one another because of their electrical charge. the cross section of the beam could be compressed to amillionth part of the cathode area, merely by refining the design of the system just described. However, because of the complications mentioned, there is a limit tothe improvement possible by precision in the design and. construction.

In. practice, the beam will emerge aperture N3 of electrode II with the various elec tron velocities directed almost along the tube axis, but. there will be a trace of radial velocity.

The focusing electrode is typical of the electron lens systems commonly employed in cathode-ray tubes. As the beam of electrons. passes through it, the electrostatic field (determined. by the shape and. potential of the electrode surfaces near the beam) changes the radial velocities of the electrons, causing them to approach the axis instead of diverging from it. The result is similar to the effect of a. convex lensv on a beam of light.

To some extent, the beam may be focused by adjusting the potentials of electrodes l4 and H (the second and fourth anodes, respectively).. If this adjustment proves sufficient, electrode 21] may be omitted.

For reasons that, will become evident later on. it is desirable that the focal length of the electron lens should be relatively long; in fact, the cross-over, where the electrons start to. diverge againafter crossing the axis, should be located a near the middle of zone D.

The basic principle employed to compress the beam in zone A is equallyzapplicable to space discharges of other types. The relative intensities of the electric and magnetic fields required dependon the ratio of charge to mass for the particles in the discharge. ciple of the beam compression is:

Charged particles enter the gion with velocities in the same general direction as the electric field which they encounter. Besides the electric field. the region contains a magnetic field which is substantially perpendicular to these initial velocities. At first. the forces on the moving particles due to the two fields are perpendicular, since an electric field exerts a direction, while a magnetic field acts at right angles to both its own direction and that of the moving particles. As the particles turn from their initial direction, the magnetic force turn too, so that it tends more and more to oppose the force of the electric field. In the end, the particles leave the compressor region with very little velocity in their initial direction, and with a much higher component of velocity perpendicular to that direction. Particleswhich enter the region side by side leave it substantially the direction which was at first sidewise becomes the principal direction of movement, and therefore forward.

This effect will reduce the transverse extent of the beam, as measured in the plane of the electric and magnetic forces acting on it, but will 'from the compressor re-.

To be precise, the prinnot directly affect its extent in the direction of the magnetic field. If the beam is thin enough in that direction, well and good; if not, a second compressor region, suitably orientated, can reduce. the extent of the beam in the other direction.

The same result may be achieved regardless of the source of the discharge or the distance traversed by the particles before they enter the compress'or region, although the structure used to provide the necessary fields may differ in detail from that shown in Figure 2- and. Figure 3, [16- I pending on the extent to which it can be combirred with adjacent structures.

Within the tube l-,. in zone: B, is a set of defiection plates, such as those used in the more common types of cathode ray tubes. Figure 1 shows one pair of plates, 23 and 2 perpendicular to the plane of the drawings, and one plate 25 of another pair parallel to the plane of the paper.

Outside the tube l, a little distance beyond the ends: of the deflection plates are two annular pole pieces, 26 and 21', concentric with the tube. Joining the outer peripheries of 26 and 21 are a number of magnetized bars, such as 28 and 29. Between the bars and the glass tube 1 are magnet coils, 30 and 3| co-axial with the tube.

Direct currents passing through the coils 3i! and 31, together with the residual magnetism of the bars 28 and Z9 and. the pole pieces 26 and 21, produce a magnetic field within the tube. In the vicinity of the deflection plates 23, 24, and 25 and its companion plate, this field is uniform and parallel to the axis.

Alternating voltages are supplied to each pair of deflection plate from a-suitable source, such as an oscillator disposed outside the tube l. The frequency used is very high-from megacycles to several billion cycles per second.

The phase relation betweenthe alternating voltages is so adjusted that the resultant electric field between the plate is substantially equivalent to a uniform electrostatic field revolving in synchronism with the alternating voltages.

Due to this revolving electric field, the electrons in zone B experience transverse forces. Like the field, the force exerted on each electron is uniform; but it changes in azimuth as the field rotates.

Under its influence, the electron gradually acquires a transverse component of. velocity in addition to it original velocity parallel to the axis.

The axial magnetic field does not affect the axial velocity, but it does. exert a transverse force upon the electron at right angles to the transverse velocity. This tends to rotate the direction of the transverse velocity at the gyromagnetic frequency, which is proportional to the magnetic flux. density (about. 2-8. megacycles per gauss). In general, the relation between magnetic field intensity and gyromagnetic frequency may be expressed as or B 2.1rfm/q path as it passes between the deflection plates 23, 24, and 25. When it passes beyond the influence of the deflecting forces, it will continue along a tangent of the spiral. Successive electrons will leave zone B at difierent azimuths, depending on theazimuth of the electric field.

The conventional system oikdefiection (without the axial magnetic field of critical intensity) would act in a somewhat similar manner, except that the spiral would return to the axi at intervals if allowed to continue. Due to this elfect, the deflection sensitivity of the conventional system is seriously reduced at high frequencies. In fact, at some frequencies, the conventional system will not produce any deflection.

Figure 4 is a graph showing the variation of deflection sensitivity, and comparing the deflection produced by the two systems mentioned.

' The abscissae of the graphs represent the product of the deflection frequency by the transit time; where the deflection frequency is defined by the alternating transverse force acting upon the electrons, while they are in the deflection zone B; and the transit time is defined as the time required for an electron of the beam to pass from one end of the deflection zone to the other. Since For example, if the transit time is one-billionth of a second, then abscissa, 1.0 represents a frequency of 1,000 megacycles per second, abscissa 2.0 represents a frequency of 2,000 megaoycles per second, and so on.

The ordinates of flection sensitivity; flection produced at the graphs represent the dethat is, the ratio of the dethe frequency in question to that which would be produced at a very low frequency, if the same voltage were applied to the deflection plates in each case.

The upper curve 32 in Figure 4 shows the deflection sensitivity of my system (with critical axial magnetic field). The lower curve 33 in Figure 4 shows the deflection sensitivity of the conventional system (without the axial field).

The mathematical process required to calculate the deflection sensitivity is rather involved, and will not be stated here. However, the advantage of my system is obvious from Figure 4.

There is still another the axial magnetic field specified. It permits the The deflection can be described most readily in terms of the pattern which the beam would trace on a plane perpendicular to the axis.

When both pairs of deflection plates are in action,

32 or as at 33, in Figure 4,

When only one pair of plates is active, with the assistance of critical axial magnetic field, the pattern will be elliptical. Typical patterns for this case are shown in Figure 5. Each of these patterns is drawn with its center at the abscissa indicating the frequency. For some of the patterns, the abscissae of their centers have been indicated by light vertical lines, which are continuations of the corresponding lines in Figures 4 and 6. I

The two curves in Figure 6 show the maximum and minimum radius of the elliptical pattern;

that is, they show the deflection sensitivity. Ti. scale of Figure 6 is the same as that of Figure Note that the average radius is just half of the obtained with both pairs of deflection plate working, and that the difierence between maxi mum and minimum radii is small at high fre quencies.

There are three possible arrangements for zonl B: The conventional system (with no axial mag. netic field) could be used; but then it would be necessary to increase the magnification of zone C, to compensate for the low sensitivity; and the velocity of the beam might have to be adjusted to avoid the condition of zero deflection sensitivity.

A second possibility is the use of a conventional deflection system capable of producing a circular deflection pattern by itself, and the use of critical axial magnetic field in addition. This is illusone form of electrostatic diiflculty in the operation of the cathode-ray converter. Furthermore, it is much easier to adjust a D. C." magnetic field than it is to adjust two A. C. circuits at frequencies such as 200 or .300 megacycles. The saving in cost, due to eliminatpair of deflection plates attendant A. C, circuits, will offset most of the cost of the magnet structure.

In zone C, the tube contains three electrodes which may be described generally as cup-shaped. The first one 35 is comparatively flat-more like a saucer than a cup; it has a small aperture 36 3! has a much lower potential, not much higher than the cathode. Electrode 39 has a higher potential than 31, and perhaps higher than electrode 35the exact value depends on the velocity desired for the electrons in zone D.

' in the axial component of velocity,

35 in electrode 35, they have scarcely left the axis. On this account, only a small aperture is needed to let them pass. Between electrodes 35 and 31, the axial velocity of the electrons drops gradually, unitl it reaches a minimum as the electrons pass through the aperture 38 in electrode 31. The minimum may be very low, even smaller than the radia1 component of velocity, if the potential or electrode 31 is made low enough.

While the axial velocity is retarded in the first part of zone C, the radial velocity suffers little change. The radial motion continues, and may carry the electrons halfway to the glass of tube l by the time they reach the aperture 38. For this reason, the aperture in electrode 31 must be fairly large to allow the beam to pass through.

Between the aperture .38 in electrode 31 and the end of electrode 39, the electric field gradually restores the axial velocity. While this change is taking place, the electrons continue their radial motion, until they almost reach the glass.

If the shape of the electrodes is properly selected, their potentials can be adjusted to make the electric field between electrodes 3'! and 39 reduce the radial velocity just before the electrons pass from zone 'C to zone 1). In any case, the axial velocity will be many times the radial velocity as the electrons leave zone C.

It is not .actually necessary to use three electrodes in zone C. The first electrode 35 serves mainly as an electrostatic shield between zone B and zone D. In addition, it serves to keep electrode 31 from reducing the axial velocity of the electrons in zone B. However, it could be eliminated at the cost of some added diificulty in adjusting the potentials of the other electrodes on either side of it.

Electrode 39 could also be dispensed with, under some circumstances. Its main function is to restore the axial velocity of the electrons, to the magnitude desired in zone D, while they are still close to electrode 31. If the space available between electrode 31 and zone B were sufficient, electrode 39 could be omitted. Then the electrons would regain their axial velocity due to the influence of an electrode in zone E which has a higher potential than would be chosen for electrode 39. The boundary of zone C would be rather indefinite-somewhere between electrode 3-! and zone ill-and zone C might overlap considerably into zone D.

If the beam consisted of positively charged particles, instead of electrons, the various electrodes would require potentials of polarity opposite to those mentioned above, but otherwise similarly related. The same principle applies to zone C in any case: reduce the axial velocity of the particles temporarily, to give the transverse component of Velocity time to produce the desired displacement within a, limited axial distance, then restore the velocity to a magnitude more suitable for subsequent use of the beam.

Zone C may be considered as a compound electron lens. From this viewpoint, the electrostatic field from electrode 35 to electrode 31 is equivalent to a concave lens, while the field from electrode 3-! to electrode 39 is equivalent to a convex lens. The first lens causes the beam to diverge more steeply from the axis, while the second lens has an opposite effect.

In general, if the path of the beam leaving the deflection zone B were projected backward, it would intersect the axis at a point near the middie of the deflection zone. This point of interarc of a helix, outside section would be substantially the same regardless of the degree of deflection. For convenience, it will be called hereinafter the deflection center. If, in a similar manner, a straight line were drawn backward along the direction of the beam as it left the first lens, this line would intersect the axis at a point farther from zone B, which may be called the first image point, because it is the location at which the first lens produces a virtual image of the deflection center. This point too is substantially independent of the degree of deflection.

The two electron lenses are so placed that the first focal point of the second lens coincides with the first image point of the first lens; thus the beam is directed parallel to the axis as it leaves the second lens in accord with the general rule that any ray passing through the focal point of a lens will emerge parallel to the axis.

In this application, the lens action aifects the beam as a whole, instead of the individual electrons within the beam, and it does not bring the electrons to a focus at the end of the tube as a conventional lens system would do. electrodes 35, 31, and 39 must be considerably larger and farther apart than the electrodesof any lens system designed to focus the beam in the usual sense of the word, by overcoming the divergent tendencies of the individual electrons.

In zone D is an armature, much "simpler in structure than most of those in common use, but the same in principle. It consists of a number of conductors-perhaps a score at most-such as iii and 41. Each of them is disposed along an the discharge chamber. Some of these conductors are connected to others at their extremities; and some are connected to A. C. circuits more or less remote from the armature, just as in the case of more commonly used armatures. In Figure l, the armature has ten main conductors, which are connected in pairs at the left and right ends alternately. One pair,

4t and M, are connected to the external load circuit, not shown; this is indicated by the horizontal extension of conductor 4|, which extends beyond the right end of the discharge chamber l; the corresponding extension of conductor 40 would be above the plane of the paper, and therefore does not appear in the drawings.

Within the tube l, zone D is almost empty. Its boundaries are somewhat uncertain, but may be considered to be close to electrode 39 of zone C, and close to electrode 42 in zone E This latter electrode is a disc, perpendicular to the axis of the tube.

Individual electrons in zone D travel substantially parallel to the axis of the discharge chamber l. Successive electrons difier slightly in azimuth, due to the design of the deflection system. As a result, the electrons in various parts of zone D at any instant are disposed along an arc of a helix. As time goes on, the electrons advance, and the helix rotates (one revolution for each cycle of deflection in zone B) The pitch of the helix may be computed from the speed of the electrons and the deflection frequency, since the two motions are related in the same maner as the linear motion of a nut and the rotation of a screw which drives it.

The helical form of the armature conductors is chosen to match that of the electron stream, so that the conductors closest to the beam at any given moment will be substantially parallel to it.

Strictly speaking, the situation is not quite so Physically,

simple as the foregoing description, but is scarcely possible to explain the precise form most suitable for the armature conductors without mathematics. Allowance should be made for the action of the armature as a group of transmission lines; for the effect of the currents in the end connections of the armature; and for the propagation time of the electro-magnetic field. Under some circumstances, it may be advisable to use different degrees of skew in alternate conductors of the armature, or a special curve near the ends of the conductors.

As viewed from the armature, the electron stream within the tube is indistinguishable from a current of electricity flowing in a helical conductor which revolves at high speed. Since the beam current is constant, it produces a magnetic field which is likewise constant, and which revolves with the current. The revolving magnetic flux cuts the stationary armature conductors, inducing alternating voltage in them. The fundamental frequency of the generated voltage, like the rotational speed of the beam, is the same as the deflection frequency used in zone B. However, the wave-form is nonsinusoidal; it may be analyzed quencies which are harmonics of the fundamental, including the fundamental frequency itself.

By a suitable choice of the sequence of the interconnections between the frequency devices. In practice, therefore, only the desired frequency is of much importance.

For example, with the ten armature conductors shownin Figure 1, connected together alternately at left and right ends, to. form a simple ten-pole wave-wound armature, the output voltage will harmonic of the deflection freq'uency,-the th harmonic, the th, etc. Assuming a beam current of 0.1 ampere, deflection freat 1500 megacycles per second, plus 2.59 volt at 4,500 megacycles per second, 0.434 volt at 7,500 megacycles per second, and smaller voltage comcurrent depends on two factors: the size of the .into sinusoidal components having fre-.

ode.

beam, which is proportional to the electrode dimensions; and the current density, which de pends on the shape of the electrodes and the current density at the emitting surface of the cath- According to Myers (page 496, Electron Optics, D. Van Nostrand Company, Inc., New York, 1939) the maximum current density attained in the prior art was 3.2 to 6.4 amperes per square centimeter. Assuming a current density of '5 amp/sq. cm., it may be calculated that a 0.100 ampere beam would require a diameter of 1.6 millimeters or approximately 1% inch. With an improved type of beam-forming device, such as that shown here (see Figures 2 and 3), much higher current densities might be obtained; and at 10 or 20 amperes per square centimeter, the maximum beam current might be over an ampere.

From the law of conservation of energy, when the armature delivers A. C, energy to an external circuit, a like amount of energy must be taken from the beam; and when the armature receives A. C. energy from an external source, a like amount of energy must be gained by the electrons in the beam. If no other influence acted on the electrons in zone D, the armature currents would retard or accelerate them, depending on whether the armature delivers or receives A. C. energy at its terminals.

For example, if electrodes 39 and 42 were both at a potential 5000 volts above the cathode, the electrons would enter zone D with a. velocity of about 4,200,000,000 centimeters per second. If then the armature currents acted to retard the beam by the equivalent of 1000 volts, the electrons would leave zone D with a velocity corresponding to that which only 4000 volts would produceabout 3,750,000,000 centimeters per second; and if the effect of the armature was reversed (by reversing the currents), would leave zone D at a velocity (about 4,700,000,000 centimeters per second) corresponding to a total voltage of 6000. In these two cases, the electrons would approach electrode 42 at velocities which would be expected if the armature had no effect but the potential of electrode 42 differed by 1000 volts from its actual potential.

In practice, it is desirable that the electrons travel at a substantially uniform velocity in zone D. The potential of electrode 42 is therefore adjusted to a value which makes the electrostatic armature currents. In the example above, when the armature currents retard the electrons, electrode 42 must be raised to a potential of 6000 volts to keep the electrons moving evenly; and when the armature currents act to accelerate the electrons, electrode 42 must be lowered to a potential of 4000 volts, instead of the 5000 originally asstuned.

The information thus derived is sufificient for the operation of my invention, but not for the design problems. Consider the eflect of armature currents in more detail.

the armature and the hypothetical conductor. Given the armature current,

. in each of the imaginary conductors which may be supposed to occupy successive short segments of its path.

In making this summation, take the voltage of each segment as of the moment the electron passes it, regardless of the voltage in that segment before or afterward. The limit of the total as the number of imaginary segments is indefinitely increased will be the true voltage induced for the electron by the armature currents. The effect of this voltage must be combined with the eifect of the D. C. electrostatic field in zone D to determine the net result.

Because the voltages induced in the imaginary conductors are alternating, that does not mean thatthe voltage affecting the electron is alternating; the electron moves from one place to another at high speed, so that it can (if the circumstances are favorably adjusted) enter a section of zone D while the induced voltage is momentarily "in one direction, and leave that section before the direction of the voltage reverses; and since successive sections of zone D may very well have induced voltages of opposite polarity, the next section that it enters may just at that moment be reversing its polarity to become as the first section was a moment earlier. By its motion relative to the alternating field, the electron commutates the induced voltage.

'If the armature currents all vary similarly, as in a single phase system, the induced voltages in various parts of zone D may be of opposite polarities, but they will all increase and decrease simultaneously. Their effect on the electrons will not be evenly distributed along the path of'the electron, and therefore cannot be entirely neutralized by the D. C. electrostatic field at every point. Consequently, the speed of the electrons will vary somewhat as they pass through zone D; however, the average velocity can be kept the same asthe initial velocity, by adjustment of the D. C. electrostatic field.

If the armature currents form a balanced polyphase system, the induced voltages will not differ greatly in amplitude, for points equally distant from the axis, and the phase angle of these voltages will vary gradually with changes in azimuth or axial position. As a result, it will be possible for electrons to pass through zone D without much change in velocity; the effect of the induced Voltages on the electron will be so nearly uniformly distributed that the D. C. electrostatic field can compensate almost perfectly for their influence on the speed.

In zone E, there are several electrodes. Electrode 42 is a circular disc, concentric with the tube l, and perpendicular to the axis of the tube. Electrode 43 is a circular cylinder, co-axial with the discharge chamber 1. One end of it is near the edge of electrode 42, so that a narrow annular aperture is formed by the gap. The other end is closer to the end of the discharge chamber, and farther from zone D. Electrode 44 is roughly conical in shape, with the wide end toward zone D; it lies within electrode 43, but not in contact with that electrode. Electrode 45 is shaped more or less like a horn; it is symmetrical about the axis of the discharge chamber, like all the other electrodes in zone E; it encloses part of the lead 46 which connects electrode 42 to the external circuits; and it closes most of the opening in the small end of electrode 44.

In operation, electrode 42 is adjusted to a potential suiiicient to maintain substantially constant velocity among the electrons in zon D.

Electrode 43 is at a potential which does notd-ifl'er greatly from that of 42. Electrodes 4'4 and 45 are at considerably lower potentials, which are adjusted in accord with the voltage induced for the electrons by the armature currents in zone D.

There are two modes of operation. In cases when the electrode voltages need not be adjusted very often, electrode 44 is maintained at a potential above the cathode which is slightly higher than the voltage induced along the beam in zone D. As a result, the electrons passing between electrodes 42 and 43 are retarded by the electrostatic field between those electrodes and electrode 44; they strike it (44) with only a low velocity, corresponding to the slight excess of potential by which electrode 44 overcomes the retarding effect of the induced voltage to which the electrons were subjected in zone D. Electrode 45 is at a potential a little lower than that of electrode 44, and serves to repel any electrons emitted from the surface of that electrode, and to drive them back so that they will not reach electrode 42.

The other mode of operation, preferred when the conditions change rapidly, so that manual adjustment of the electrode potentials is diflicult or impossible, is as follows:

Electrodes 42 and 43 are maintained at high potentials, as before. Electrode 44 is maintained at a potential considerably higher than is necessary merely to allow the electrons to reach it. The electrons strike it with such force that additional electrons are emitted from its surface. These secondary electrons are attracted to electrode 45, which is at a potential slightly higher than electrode 44.

If the potential of electrode 44 is excessively high, the number of secondary electrons will be great, and a large current will flow between electrodes 44 and 45. If the potential drops too low, the number of secondary electrons will decline, reducing the current of electrode 45. This variation of current can be used to operate devices which will adjust the electrode potentials to more nearly the proper value. The control devices which could be used range elaborate systems of relays and motor driven switches for changing the D. C. supply voltage.

For many cases, electrode 43 may be connected to electrode 42 at any point where it is convenient to do so, and the connection does not obstruct the path of the beam. In other cases, it may prove convenient to keep electrode 43 at a potential slightly lower than electrode 42, in order to deflect the electrons slightly toward the axis as they pass through the aperture between 42 and 43. In still other cases, electrode 43 may be omitted entirely, on the grounds that it does not add enough to the control of the electron paths to justify its cost of manufacture. Then the diameter of electrode 44 would have to be increased.

It is highly probable that experienc will indicate some modifications of the shapes of electrodes 44 and 45, to make them more effective and limit the number of electrons which stray from their intended destination.

When the cathode-ray converter is used as an amplifier, the operating procedure will be about as follows: First, test the D. C. supply circuits, and adjust the electrode potentials to appropriate values. Then start the oscillator which controls the deflection, and adjust it to the desired frequency.

After the deflection system is set in operation, adjust the potentials of the electrodes in the defrom simple imped ances in the external circuits of the electrodes to flection booster, zone C, to make the beam pass between electrodes 42 and 43 in zone E without striking either. For'this stage of the procedure, electrodes 44 and 45 need not be more than 50 or 100 volt above the cathode potential; electrodes 42 and 43 should be at about the same potential as electrode 39. The adjustment will be easier if the face of electrode 42 has a coating of fluorescent material, to show where the beam strikes it; but when the adjustment is completed, such a coating will be of no particular use, since the beam will notcome in contact with it.

Next, raise the potentials of the electrodes in zone E to about the values at which they should operate, and connect the load to the armature. Then adjust the potentials more accurately. If the potential of electrode 44 is too low, the beam current'will flow to electrodes 42 and 43 instead,

but normallythe current to 42 and 43 should be almost negligible. If the potential of electrode 44 is too high, the electrons will strike it with more force than necessary, and waste energy. This condition may be detected by measuring the currents of electrodes 44 and 45; due to secondary emission, the net current of electrode 44 will decrease, and that of electrode 45 will increase, when their potentials are too high. The sum of the electrode currents in zone E will, of course, equal the beam current.

When all the adjustments are properly made, the cathode-ray converter will operate to convert D. C. power to A. C., with an efiiciency of 90 to 95 per cent. This is better than the average 60 cycle rotary converter, and surpasses the efficiency of any device previously known for the generation of radio frequency power.

For radio telegraphy, the output can be controlled by switching the beam on and off. Various methods have been devised for controlling the current according to the position of the telegraph key, for other types of amplifiers, and the same well known methods can be applied here.

For telephony, or television, amplitude modulation can be accomplished by modulating the beam current, This may be done by means of a control electrode, such as is used in a television picture tube, or by means external to the cathode-ray tube.

If the low frequency signal is used to control only the beam current, while the electrode potentials are kept constant, the efficiency of the converter will be reduced by 25 to 50 per cent, depending upon the degree of modulation employed.

If, on the other hand, the potentials of electrodes in zone E are varied simultaneously by an amount proportional to the variation of the current, the efficiency of the converter may remain at the same high figure (over 90 per cent) even when modulation is employed. In the latter case, however, there may be more power lost in the D, C. circuits outside the tube, and particularly in the apparatus used to control the potentials.

The modulation of output amplitude could be accomplished by modulating the amplitude of the A. C. supplied to the deflection system, or by modulating the magnification factor of the deflection booster.

The output of the armature varies with the radial displacement of the beam, and therefore depends on the deflecting force and the deflection sensitivity. To allow the beam passage, electrode 42 would have to be built like a sieve or a spiderweb, and some modification of electrodes 44 and 45 would also be advisable. Even after such tials of zone E need not change.

- some frequency, say f1.

changes in design, the efficiency with this method of modulation would be lower than with the method mentioned first, but the less efficient method might be more convenient in some cases.

If frequency modulation is employed, there is no variation of efficiency, since the output power remains fixed, The beam current and the poten- Any of the known methods of frequency modulation may be applied to the source supplying A. C. to the deflection zone B. Theoretically, the variation of frequency would cause a corresponding variation in the radial displacement of the beam, unless the magnetic field of zone B or the potential of electrode 31 in zone C were varied simultaneously. However, for the frequency deviation commonly used in frequency-modulation transmitters, the effect of frequency on deflection sensitivity is too small to need such compensation.

There is one case in which it would be important to modulate the potential of electrode 31, in order to modulate the magnification of the deflection booster; when the converter is used as a modulated oscillator. The deflection voltage, being derived from an armature whose output voltage varies with the beam current, will likewise vary; and to keep the beam at the normal radial displacement in zone D, the magnification of the defiection booster (zone C) must vary inversely with the amplitude of the deflection voltage, Both beam current and the potential of electrode 31 can be controlled by the same signal, through suitable auxiliary apparatus.

In any of the applications possible for my invention, the output frequency may be the same as the deflection frequency; but it may equally well be some multiple of the deflection frequency, if the armature is so designed. The structural difference between armatures designed to operate at various multiples of the fundamental frequency is similar to the difference between armatures designed for dynamos with various numbers of pairs of poles. Each armature may receive power from the beam, or deliver power to the beam, according to the nature of the external circuit to which it is connected, It does not matter whether the various armatures operate at the same or different harmonics of the deflection frequency. If the net transfer of power is from beam to armatures, a corresponding amount of power must be supplied by the'D. C. circuits of the beam, If the beam receives more power than it delivers, the potential of the collecting electrode in zone E can be reduced below that of the cathode 1; and the beam will persist in spite of this opposition. It will then deliver D. C. power to the external circuit.

The converter can be used as a mixer, such as the first detector of a superheterodyne. If desired, it can function as oscillator, or frequency multiplier, or both, to produce one of the voltages mixed, without impairing its operation as mixer.

The other functions have been explained above. The mixing function operates as follows:

The beam induces in the armature voltages of An external source conproduces currents of an- These currents induce a with components whose nected to the armature other frequency, say f2. voltage along the beam, frequencies are the sum or difference of various multiples of f1 and f2. By appropriate choice of armature design and beam velocity, most of the undesired results of the mixing can be eliminated, leaving the first difference frequency (fl-f2, or fz-f1) as the only important component. This voltage modulates the velocity of the beam, cause vng. a corresponding variation in-the impact of ;he electrons striking electrode 44; secondary emission from electrode 44 then convertsthe velocitymodulation into an alternating current of the same frequency, flowing between electrodes 44 and 45. This A. C. is superimposed on the D. C. normally flowing to said electrodes.

If f1 equals f2, the difference frequency will be zero. The induced voltage along the beam will then be A. C. at zero frequency, which is synonymous with D. C.

The operation as a mixer can be improved by using the output voltage (due to the A. C. flowing in a suitable impedance in the circuit connecting electrodes 44 and 45) to adjust the instantaneous potentials of electrodes 42 and 43, in order to keep the beam velocity nearly constant in zone D. This adjustment does not affeet the impact of the electrons on electrode 44, which is determined by the relation between the potential of electrode 64 and the voltage induced along the beam. Only a little power is required to control the potentials of the electrodes 42 and 43, since very little current flows to them.

Although I have referred to the space discharge mostly as a beam of electrons or a cathode-ray, this is not a necessary limitation. Protons, deutrons, positrons, or any other charged particles could be used in a space discharge, and would serve my purpose; but at the present state of the art, electrons are more readly available. If positively charged particles were used, the polarity of inter-electrode voltages would be the opposite of that used with electrons.

I have shown a case in which the discharge passes lengthwise through the discharg chamber, and in which a portion of the discharge path rotates periodically. This form is probably the most convenient, but in other embodiments of my invention the discharge might follow a more complicated path, and it might have reciprocating instead of rotary deflection.

For example, the discharge path might be a spiral, or a circular are, as in the Lawrence cyclotron; nevertheless, if one or more conductors can be placed near a portion of that path, and approximately parallel to it, any deflection applied to the discharge before it passes the conductors will shift the path and vary its distance from them; and this variation will generate voltage in the conductors by electromagnetic induction.

While I have shown a particular form of embodiment of my invention, I am aware that many minor changes therein will readily suggest themselves to others skilled in the art without departing from the spirit and scope of the invention; what I claim as new and desire to protect by Let ters Patent is:

l. A cathode-ray tube, an electron gun disposed within the tube for discharging and directing a beam of electrons, means for deflecting the beam, means for magnifying the displacement resulting from the deflection and neutralizing the effect of the deflection on the direction of the beam path, an armature wherein alternating voltages may be induced by periodic displacement of the beam due to deflection at a corresponding frequency, and a plurality of electrodes for collecting the electrons at the end of their passage through the tube.

2. A cathode-ray tube, an electron gu disposed within. the tube for discharging and directing a beam of electrons, a plurality of electrodes for collecting the electrons at the end of their passage through the tube, means interposed between. the electron gun and the electrodes for deflecting the electronsin arotary fashion and for boosting the deflection of the electrons in their passag from the gun to said electrodes, said deflecting means comprising-a plurality of diametrically opposed. deflection plates, said booster comprising a plurality of electrodes having central apertures of progressively increasing diameter, and an armature comprising a plurality of conductors adjacent to, the tube disposed generally parallel to the beam, and interconnected at their extremities, wherein alternating voltages may be induced by periodic displacement of the beam due to. deflection at a corresponding frequency.

3. A method of converting power from direct current to alternating current, comprising the steps of discharging electrons in a beam, periodically displacing thepath of the beam transversely in a substantially circular patternto produce a periodic variation in the distances between the beam and adjacent conductors disposed generally in the direction of the beam to induce alternating voltages in the conductors, allowingsaid voltages to cause alternating currents in' the conductors, then collecting the electrons at the end of their travel.

4. A-method of converting power from direct current to alternating current, comprising the steps of discharging electrically charged particles in a beam, periodically displacing the path of the beam transversely in a pattern forming a closed curve to produce aperiodic variation in the distances between the beam and adjacent conductors disposed generally in the direction of the beam to induce alternating voltages in the conductors, allowing said voltage to cause alternating currents in-the conductors, then collecting'the particles at-the end of their travel.

5. A device for producing a beam of electrically charged particles, comprising a source of electrically charged particles, a group of electrodes near this source to produce anelectrostatic field, one of 'saidelectrodes having an eperture to provide egress for theparticles, and a magnetic field disposedperpendicular to the electrostatic field, said magnetic field extending throughout the major part of the region in which the electrostatic fleld acts upon the particles.

6. A method of'generating alternating voltages comprising the steps of producing a space discharge, and controlling the path ofthe discharge in such fashion as totrace out a closed curve to cause a periodic variation in" the distance between the discharge and adjacent conducting means disposed generally in the direction of the discharge, thereby inducing alternating voltage in the conducting means;

"I; A method of generating voltage in series with a space discharge, comprising the following steps: producing a'space discharge in such fashion as to traceout a closed curve, controlling the path ofthe discharge to cause a periodic variation in the distance between-said discharge and adjacent conducting means: disposedgenerally in the direction of the discharge, and causing to flow in the conducting means alternating current whosefrequency differs from the frequency of the desired voltage by the frequency of the voltage which the discharge induces in the conducting means.

8. A cathode-ray tube, an electron gun disposed within the tube for discharging and directing a beam of electrons, a plurality of electrodes for collecting th electrons at the end of their passage through the tube, means interposed between the electron gun and the electrodes for deflecting the electrons in such fashion as to trace out a closed curved and for boosting the deflection of the electrons in their passage from the gun to said electrodes, said deflecting means comprising a plurality of diametrically opposed plates having terminal connections for supplying the plates with fluctuating potentials for beam deflection purposes, and said booster comprising a plurality of electrodes having central apertures of progressively increasing diameter, and having terminal connections for maintaining them at desirable potentials, and an armature comprising a plurality of conductors adjacent to the tube disposed generally in the direction of the beam, wherein alternating voltages may be induced by a periodic displacement of the beam due to deflection at a corresponding frequency.

9. A device for reducing the cross-sectional area of a beam of electrically charged particles, comprising means for producing an electrostatic field, means for producing a magnetic field disposed perpendicular to the electrostatic field, said means for producing an electrostatic field being provided with apertures to allow the beam to enter the field in one direction substantially perpendicular to the magnetic field, and to leave in another such direction substantially perpendicular to the first, and said magnetic field extending throughout the major part of the region traversed by the particles between the apertures.

10. In a space discharge device requiring a periodic deflection of the discharge, a method of counteracting the adverse effect of transit time upon deflection sensitivity, and of producing a circular or elliptical deflection only slightly affected by the deflection frequency, which method is as follows: producing at least one transverse field in a region traversed by the discharge, whereby the particles composing said discharge are accelerated transversely in the direction of the force exerted on them by the transverse field; producing within the same region a longitudinal magnetic field substantially parallel to the av erage direction of the discharge within that region whereby the discharge is further deflected, the particles being accelerated transversely at right angles to their transverse velocity, so that the direction of their transverse velocity is gradually rotated; varying said transverse field at the frequency desired for deflection of the discharge; and adjusting the intensity of said longitudinal magnetic field to substantially the critical value at which the gyromagnetic frequency is equal to the frequency of variation of the transverse field, by which adjustment the rotation of the transverse velocity of the particles is made synchronous with the alternations of the transverse field, and successive half-cycles of the transverse field are made additive in respect to the total transverse momentum of the particles, even though opposite in direction, and the transverse momentum of the particles increases continually as they progress through the region under the joint influence of the aforementioned fields.

11. In a space discharge device of the cathode ray type, means for deflecting the beam of charged particles comprising the following: a plurality of electrodes adjacent to the path of the beam means for varying the potentials of the electrode; at a high frequency, to produce transverse elec' trostatic forces acting on the discharge; a plurality of permanent magnets disposed generall: parallel to the path of the beam; annular pol pieces of magnetic material at the ends of tht permanent magnets, to direct the magnetic fielc axially through the region between the aforesaid electrodes; coil means coaxial with the pole pieces; means for producing a direct curreni through the coil means, and means for adjusting the intensity of the current, to adjust the intensity of the magnetic field within the discharge chamber.

12. A cathode-ray tube, an electron gun disposed within the tube for discharging and directing a beam of electrons, a plurality of electrodes for collecting the electrons at the end of their passage through the tube, means interposed between the electron gun and the electrodes for deflecting the electrons in such fashion as to trace a closed curve on the collecting electrod and boosting the deflection of the electrons in their passage from the gun to said electrodes, said deflecting means comprising a plurality of diametrically opposed deflection plates together with means for producing a magnetic field along the beam, said booster comprising a plurality of electrodes having central apertures of progressively increasing diameter, and an armature comprising a plurality of conductors adjacent to the tube wherein alternating voltages may be induced by periodic displacement of the beam due to deflection the main conductors of said armature being disposed generally parallel to the beam in such manner that the distance from the beam to the nearest conductor will be substantially uniform throughout the length of the armature.

13. In a space discharge device employing a beam periodically displaced transversely in a rotary fashion and inductively coupled to adjacent conducting means, the arrangement of said conducting means comprising a plurality of longitudinal conductors disposed generally in the direction of the beam and interconnected at their extremities to form an armature, characterized in that the longitudinal conductors are skewed with respect to the axis of rotation of the beam by substantially the amount necessary to make the distance from the beam to the nearest conductor uniform throughout the length of that conductor.

14. An armature as in the preceding claim, characterized in that the longitudinal conductors are skewed with respect to the axis of rotation of the beam by an amount substantially equal, on the average, to that skew which would make the distance from the beam to the nearest conductor uniform throughout the length of that conductor, and further characterized in that the maximum and minimum skew of the conductors differ from the average by substantially the amount which would be required for that average according to the relation stated above, if the beam traveled with the velocity of electromagnetic waves.

ROBERT E. McCOY.

Images