US2939991A - Electron velocity modulation tubes - Google Patents

Description

June 7,1960 A. H. w. BECK 2,939,991

ELECTRON VELOCITY MODULATION TUBES Filed June so, 1958 3 l/vrgRAcr qy 4 6 \I REG/ON 51 4 /NTEPACr/ON r 'wz'smv -fik 4 I 2 Inventor A .H .W- BECK Attorney United States Patent ELECTRON VELOCITY MODULATION TUBES Arnold Hugh William Beck, London, England, assignor to International Standard Electric Corporation, New York, N.Y., a corporation of Delaware Filed June 30, 1958, Ser. No. 745,688

Claims priority, application Great Britain Aug. 22, 1957 2 Claims. (Cl. 315-3) The present invention relates to high power electron velocity modulation tubes and is particularly concerned with electron collector arrangements therefor.

It is well known that the overall efliciency of a velocity modulation tube could be improved if the voltage of the electron collector electrode could be reduced below that of the cavity resonators, in the case of a klystron, or that of the slow wave structure, in the case of a travelling wave tube or backward wave oscillator or amplifier. It is nevertheless found that the collection efiiciency drops sharply when the collector voltages falls below some minimum value which is typiwlly 50 to 70% of the voltage of the resonator or slow wave structure, as the case may be. To take the case of the klystron, though what follows applies equally to other types of velocity modulation tube, the drop in efiiciency follows because the electron beam at exit from the output resonator is very strongly velocity modulated and, when confronted with a low D.C. collector voltage, many of the electrons have insutficient energy to overcome the retarding field of the collector region.

It is an object of the present invention to provide means for overcoming the above mentioned drop in collection efiiciency and hence to allow the electron collector electrode to be operated at low voltage, thereby providing an electron velocity modulation tube of increased overall efiiciency.

Fig. 3 is a fragmentary drawing of the output end of a klystron tube according to the invention.

In the drawings of Figs. 1 and 2, an electron gun cathode, from which the electron beam emanates, is indicated by the rectangle 1 and the electron collector electrode is shown at 2. The path of the electron beam is indicated by the dotted lines 3. Between the electron gun and the electron collector electrode the beam passes through a region of its path in which it is coupled, by electrode means not shown, to electromagnetic fields with which it interchanges energy. This region is indicated by the chain-lined rectangle 4. In the case of a klystron the region 4 comprises two or more resonant cavities separated by drift tubes, while in the case of a travelling wave tube the region comprises a helix or other form of slow wave structure linked with the electron beam path together with input and output waveguide or coaxial line feeders and the necessary impedance matching arrangements well known in the art. (The waveguides and, usually, the coaxial line feeders just mentioned, are not part of the tube itself, but are coupled to the slow. wave structure within the tube envelope by the said matching arrangements, part of which will, normally, be within the envelope enclosure.) Whatever be the type of tube, the electron beam within the interaction region, at least at the output cavity or the end of the slow wave structure, will be subject to high D.C. potentials and the electrons will have a corresponding high mean velocity The invention is based upon the realisation that if the according to the invention, by effectively suppressing the electron velocity modulation of the beam at the collector electrode; it may be carried out either by subjecting the beam to strong V.P. energy absorbtion after leaving the output resonator or slow wave structure by the expedient of passing the beam through a separate resonant elec- .1 trode means which may comprise a highly damped additional resonator,or the placing of 'an electron collector "at the end of a drift tube maintained at high D.C. po-

tential and of such length that the electron plasma waves on the beam are at a point of minimum A.C. velocity immediately in front of the collector. The electron collector electrode in this second case surrounds, as in conventional practice, a hollow space, but the entrance to this hollow space is gridded so as to provide a flat collecting field and help in the suppression of secondary emission.

The invention will be further described with reference to the accompanying drawings in which:

Fig. 1 shows, purely diagrammatically, a velocity modulation tube having an extra beam-demodulating cavity resonator according to the invention;

latter where the A.C. velocities are a minimum; and

penetrate through the interaction space.

along their path.

In the case of a klystron, when the beam crosses the gap in the output resonator the electron bunches excite the electromagnetic waves to which the resonator is tuned. The high frequency energy in the output resonator is obtained from the beam at the expense of the kinetic energy of the beam electrons; the electrons in the bunch are slowed down and the electrons reaching the gap in unfavourable phase are accelerated, so that energy is obtained by debunching of the beam. But, for optimum power extraction, this debunching is far from complete, and, in fact, on leaving the output gap many of the electrons will have velocities very far from the mean value, so that quite a large proportion would be reflected back along their path by the field of a low potential collector electrode placed, as is normally the case, immediately following the output resonator. As mentioned previously, it is normally found that the collector voltage cannot be reduced below some 50% to 70% of the output resonator voltage. The reflected electrons are either collected on the cavity walls at the full H.T. potential or In the first case they prevent the overall efliciency from being increased, while in the second they contribute to undesired instabilities and self-oscillation. Both effects must be eliminated. Similar considerations apply in the case of tubes where there is a continuous instead of a localised interchange of energy between beam and field. Thus, in the case of the backward Wave oscillator, in which the field energy is extracted near the electron gun end of the tube, although on the average the electrons will deliver up more energy to the field along the slow wave structure than they receive, they will leave the interaction region of their path with an appreciable velocity modulation.

In the arrangement of Fig. 1, after leaving the interaction region the beam passes along a further length of drift tube 5 and through a cavity resonator 6 all at high D.C. potential. In carrying out the invention the resonator 6 is caused to be very heavily loaded. If the parallel load resistance of the resonator 6 is made much lower than the load for optimum power extraction, the ratio frequency voltage developed is very low and the beam at ,exit is of practically uniform velocity. At the same time the Q of the resonator is very low while the energy absorbing gap in the resonator can be made long and the capacity across the lips of the gaps correspondingly low, giving flat tuning; thus the resonator. does not .have. to be tunable. The radio. frequency power in the. resonator ,6 canbe .dissipitated either in a built-in load .ormay be added to the useful output from the outputresonator proper. Afterpassing the resonator 6, the electrons are collected by electrode 2 which may be of conventional type and be located at any convenient distance beyond the output gap, but may now beoperated at or but little above the cathode potential." It follows-.from what has been saidabove about the characteristics of the resonator 6 that-the drift tubes may he made any con venient length.

i In'the' preferred arrangement of drift tube is used to. reduce the velocity modulation of the electrons.

aca am a before. The collector electrode 2 is joined to resonator 9 through insulators 11,.and the drift tube 7 is fixed in This. arrangement is based on' the space charge wave.

theory" of velocity modulation processes. From this theory it appears that the efiect of velocity modulation, of the beam. electrons is to set up on the beam electron plasma waves which form'along a drift tube, stationary waves of AC. current and AC. velocity, the current and Velocity waves being in phase quadrature. In the case. of a kly stron' the output gap is positioned at. an A.C. current antinode, which, since current andvelocity are in quadrature, is an A.C. velocity node of the plasrnawaves. If

the output gap could extract all the power from the beam, the A.C..current of the plasma waves would be reduced to zero. and the beam would be left with no velocity modulation at all; the electron collector electrode could then be placed at any subsequent position and'be maintained at cathode potential without repelling any electrons. In practice, however, the output gap not only does not demodulate the beam completely, but the high R.F. voltage developed across the lips of the gap re-' modulates the electron beam, and, from the point of of velocity modulation for the beam. An analogous argument can be developed for other typesof tubesuch as thetravelling Wave tube or backward wave tube. Here the end of the slow wave structure is to be regarded as the wall of resonator 9. The mouth of the collector electrode is closed by an electron permeable grid 13,

' which is separated by a short insulating gap 14 from the end of drift tube 7. The. distance between the output gap in resonator 9 and gap Mismade A 14. g p

The embodiment of Figs. 2 and 3 as so fa; described is based upon the simple single space charge wave theory. 'ln this the basic wave parameters are' determine d inter alia, by the fact that at the conducting surface 'of a drift tube the tangent-ial conrponents of electricv force must vanish. V Thus, if E, be the component. of electric force acting on the beam in the axial direction, andthe beam fills the tunnel formed by the drift tube, E- must vanish at the beam boundary. 1 But the consequent'radial variation of E across the beam is far from what must obtain in an ungrided hiystron gap, or indeed, in the: helix field of a travelling wave tube. Thelsituationissomewhat analogous to the excitation in the fundamental mode of a hollow waveguide by a probe; higher order modes view we are considering, must be regarded as a source must come into play. Butin'the hollow waveguide the higher, order'modes are-rapidly attenuated,and can be ignored for most purposes; the electronbeam, on the other 'hand, can and does transmit the higher order plasma waves, which should,- therefore, be takeniintoaccount throughoutthe length'of the velocity modulated beam.

' Thus, in order to obtain a closer approximation to prac? tice, instead of considering only asingle space charge wave, the higher 'orderspace charge wave theory con-- siders a Fourier-Bessel seriesof spacecharge waves which gives riseto a non uniformdistribution of -directed AC. velocities in the beam and explains, inter; alia, the experimental fact that in a solid beam there is a reversal in sign of AC. velocities across the beam section.

W'ne'n the higher order space charge wavejtheory is taken into account, the length of the drift tube 7 should be made greater than x 4 for the best compromise of velocitymodulation suppression at the collector electrode; the collector electrode would also. have to be somewhat above cathode potential. From this theory it would ap pear that the best position forthe electron collector electrode is where theAC. velocities. corresponding to the first and secondorder modesare-equal and opposite. In

the simple space'charge wave theory this drifttuhe 7 7 should be made of length K /,4 long, where-A is the: elc

'tron plasma wavelength, modified to=-take account of the'drift tube geometry, and the electron collector electrode should be placed at the end or. drift tube-7; the

.electron collectorclectrode is now'positioned atan A. C.

velocity null'of the space'chargewaves induced at the end of the interaction region. The electrons may thus be collected at low potential; as indicated in Fig; 2 where a D.C. source 8 is represented as connected between the cathode and drifttube '7 and collector 2 is connected near the cathode'end of source 8. The gap between collector 2 and'drift tube 7 may be. made asshort as allowed by A practical example of the invention of Fig. 2 as applied'to a klystron is shown in Fig. 3, which illustrates the basic construction of the output end of the tube. The 'outputresonator is indicated'at 9, the maindrift-tube at 10, and the additional drifttube and the collector elec-' 't-rode are identified by the same reference numerals as 7 example and notfas' a limitation o'nthe scope of the in- 7 '60 gt e n in d t c w en t em. The collector, as in conventional practice, isrnade hollow, but its mouth is V a practical construction the exact collector position for maximum collection efiiciency can be determined empirically.

While the principles of the invention havefbeen described above in connection with specific embodiments, and particular modifications thereof, it is to be clearly understood that thisv description is made only by way of vent-ion. Y V A.

Whaticlaimed is: I c j 1. An electron velocity modulationtube comprising electrode meausforprojeotingkan electron beam from a cathode along a given pathin'energy interactingrelationship with electromagnetic waves "in a given region of the path, means providing a high'direct potential operating a on said beam at the end of the said given region a' drif-t tube maintained at the-said high potential following ihe beam exit fromthe said region and of length exceeding one quarter electron plasma wavelength of the fi'rst'rriode of plasma waves by an'amourrt such'that'the: alternating current velocities of the first and seco'nd' modes of plasma waves on thesaid beam'are equaland opposite, an electron collector electrode at a potential near that ofthe said cathode, located'at the exit from the said drift tube, the said collector electrode providing a'liollow spaceior the collection of theelectrons and an electron permeable grid covering the end of the said hollow space to provide a substantially flat eq'uipotential surface, at collector potential. V

2. An'electron velocity modulation tube. comprising electrode means for projecting an electron beam. from a cathode along a given path. energy interacting'relaiiom.

5 6 ship with electromagnetic waves in a given region of the ing a spaced substantially unipotential surface immediate- P e ns pro ding a high direct potential operating on 1y adjacent the end of said drift tube. said beam at the end of said given region, a drift tube maintained at said high potential region following the References Cited in the file of this patent beam exit from said region and of a length substantially 5 UNITED STATES PATENTS equal to a one-quarter electron plasma wavelength of the first mode of plasma waves, an electron collector electrode 2,240,183 Hahn P 29, 1941 at a potential near that of said cathode located at the 2,312,723 Llwellyn Mali 1943 exit from said drift tube, said collector electrode present- 2,631,951 Wamecke J1me 1954

Images