US2939037A - Apparatus for suppression of multipactor - Google Patents

Description

1, 1960 R. L. JEPSEN 2,939,037

APPARATUS FOR SUPPRESSION OF MULTIPACTOR Filed Jan. 50, 1956 P05527- L dEPs EN I N V EN TOR.

United States Patent 2,939,037 APPARATUS FOR SUPPRESSION OF MULTIPACTOR Robert Jepsen, L'os Altos, Calif., assign-or to Varian Associates, San Carlos, Calif., a corporation of California Filed Jan. 30, 1956, Ser. No. 562,220 12 Claims. (Cl. 315-5.52)

Briefly, multipactor consists of a cloud of charged particles, such as, for example, electrons, oscillating back and forth between two surfaces under the influence of a radio frequency electromagnetic field. When the cloud strikes one of the surfaces, most of the initial particles are absorbed, but they are replaced by secondary particles knocked out of the impinged surface. The secondaries are driven back to the other surface by the field which has reversed itself by this time. For the multipactor process to build up, the secondary emission ratio of the surfaces must be at least greater than unity, the transit time of the particles between the surfaces must average one-half cycle of the radio frequency electromagnetic field, or some odd multiple thereof, and the RF. electromagnetic field must supply energy, since fast moving particles are replaced at each impact with the surfaces by slow moving secondaries which are accelerated by the field.

The process may be initiated by any charged particle that appears in the region at an appropriate time. The process builds up if the secondary ratio is high enough until a stable condition is reached where just as many extra secondary particles are produced as are lost from the cloud for various reasons.

The effects of multipactor on the operation of many high frequency devices are due primarily to the absorption of power, although reactive effects may produce some detuning. For example, in klystrons multipactor losses may be expressed as a power loss across the grid gap, and this power loss may vary very rapidly with voltage and hence also with power level in the tube. Thus, nonlinear output may be expected from klystron amplifiers in which multipactor occurs. In klystron oscillators much the same sort of phenomenon may occur, except here one is worried not about linearity, but about limitations on maximum output. As the oscillations build up, multipactor may start absorbing some of the power and limiting the voltage. Since this voltage is fed back to modulate the beam, the drive also tends to be limited; thus, further building up of oscillations may be impossible, and the output is less than would otherwise be expected.

From the theory as presented thus far it can be readily seen that if the secondary emission ratio can be reduced below unity the multipactor phenomenon cannot continue.

Heretofore, several schemes have been employed to reduce the secondary emission ratio of the mutually opposing surfaces. One such scheme comprised coating the mutually opposing surfaces with a coating of carbon, thereby reducing the secondary emission ratio of the surface. However, this procedure has generally been found iice unsatisfactory in many applications, primarily because the surfaces become contaminated subsequently, whence their secondary emission ratio rises. For instance, many electron tubes that have had the interior surfaces coated with carbon have been contaminated by emission conducive compounds, such as BaO, which had evaporated from the cathode and condensed on the coated surface. Another scheme employed in klystron tubes comprises beveling the free end portions of mutually opposing re-entrant drift tubes thereby varying the distance between mutually opposing surfaces and disturbing the transit time effects. This has been found helpful in reducing multipactor losses but has not eliminated the effects to the degree desired.

The present invention provides means for preventing multipactor by decreasing the probability of producing secondary particles which will have the same transit times as their parents and by recapturing secondary particles before they have an opportunity to enter into the oscillating cloud of charged particles.

The principal object of the present invention is to provide novel improved multipactor suppression methods and apparatus whereby charged particle electrical loading may be decreased in many electromagnetic devices thereby greatly enhancing their performance.

One feature of the present invention is the provision of a land and a groove in the surface of members which are likely to multipactor in use whereby the tendency for multipactor to arise under the influence of an alternating electromagnetic field is substantially reduced.

Another feature of the present invention is a novel improved multipactor suppression surface wherein a land and a groove are formed in the surface and wherein the thickness of the land at its free end portion is substantially less than the distance from the free end portion of the land to other opposed surfaces in use.

Another feature of the present invention is a novel improved multipactor suppression surface wherein a land and a groove are formed in the surface and wherein the width of the groove at the free end portion of the land is substantially greater than the width of the land at the free end portions thereof.

Another feature of the present invention is a novel improved multipactor suppression surface wherein a land and a groove are formed on the surface over which it is desired to prevent multipactor and wherein the land and the groove run substantially parallel to the alternating currents in the surface of the member thereby preventing perturbations of the current flow in use.

Another feature of the present invention is a novel improved multipactor suppression surface wherein the surface is permeated with a plurality of holes whereby the tendency for multipactor to arise over the surface under the influence of an impressed alternating electromagnetic field is substantially reduced.

Another feature of the present invention is a novel multipactor suppression surface configuration wherein a honey-combed type structure is formed on the surface of the member over which it is desired to suppress multipactor thereby effectively suppressing multipactor.

Another feature of the present invention is a novel multipactor suppression method and apparatus wherein a honeycombed type structure is secured at one side thereof to a thin sheet of flexible material which may be formed to match the contour of various shaped surfaces and adhered to said surfaces to prevent multipactor therefrom.

Still another feature of the present invention is a novel improved re-entrant drift tube configuration useful in re-entrant cavity resonators wherein the mutually opposing free end portions of said drift tubes are beveled at their outside surfaces and provided with a plurality of lands and grooves in said beveled surfaces whereby multipactor is substantially reduced between said opposed drift tube portions.

7 ing smooth surfaces under the influence of an alternating electromagnetic field and depicting the multipactor mechanism,

Fig. 2 is a fragmentary cross-sectional view of two mutually opposing surfaces showing the novel improved land and groove multipactor suppression surface configuration.

Fig. 3 is a fragmentary cross-sectional view of a novel triangular shaped land and groove multipactor suppression surface,

Fig. 4 is a longitudinal partial cross-sectional view of a re-entrant cavity resonator incorporating the novel improved multipactor suppression drift tube configuration of the present invention,

Fig. 5 is an enlarged view of a portion of the structure of Fig. 4 taken along line 5-5 in the direction of the arrows,

Fig. 6 is a perspective view of a novel hole permeated multipactor suppression surface configuration of the present invention, and

Fig. 7 is a perspective view of a novel honeycomb multipactor suppression surface configuration shown mounted upon a suitable base material.

Referring now to Fig. 1 there is depicted the mechanism of the multipactor phenomenon. A charged par ticle, for example, an electron e appearing in the region between two mutually opposing smooth surfaces 1 and 2 falls under the influence of an impressed alternating electromagnetic field E in the space S between the two surfaces 1 and 2. The electron e is accelerated toward'the surface Z'and impinges thereon knocking out secondary electrons. The number of secondaries on an average is proportional to the secondary emission ratio of the impinged surface 2. When the transit time for the electrons traveling between the surfaces approximates an odd integral number of half cycle periods of the impressed alternating electric field the multipactor phenomenon may develop.

Referring now to Fig. 2 there is depicted a cross-sectional view of a novel multipactor suppression surface configuration wherein mutually opposing members 3 and '4 have been grooved to provide lands 5 on the surfaces thereof. An electric field map has been drawn between opposing lands 5 to show that the electric field is substantially greater at the free end portions of the lands 5 than deeper within the groove portions.

The importance of the weak A.C. electric field in the grooved portions arises from the fact that emitted secondary particles, having some initial velocity when emitted from the surface, will be relatively little affected by the A.C. electric field and the probability is increased that they will be recaptured by an adjacent land 5 be fore the particle can enter into the cloud of charged particles.

Moreover, for those particles not recaptured the secondary transit time is quite likely to be different than its parent particle thereby inhibiting multipactor action. When the Width d of the land 5 at its free end is small as compared to the spacing S between surfaces theprobability of multipactor action is greatly reduced. When the height h of land 5 is made sufficiently great as compared to the width w of the groove substantially all particles emitted within the groove will be recaptured. When all groove-born secondary particles are recaptured then the effective secondary emission ratio of the multipactor suppression surface is reducedby a factor proportional to the ratio.of;the surface area-of the free endland portions and inversely proportional to the total original surface area, i.e.,

where S is the effective secondary emission ratio of the surface, S is the secondary emission ratio of the surface, A is the area of the free end portion of the lands, and A is the total planar surface area before the lands were provided. For instance, if the secondary emission ratio of the surface material is 10:1, then by reducing the free end portion area to of the original surface area the effective secondary emission ratio is reduced to 1 which is a suficient reduction to prevent multipactor from arising.

Referring now to Fig. 3 there is shown a second embodiment of the present invention wherein the multipactor suppression surface configuration comprises a triangular land 6 and groove portion. A field map has been drawn to indicate the portions of strong alternating electric field. It can be seen from the field map that the electric field is substantially greater at the free end portion of the triangular land 6. The probabilities are indeed quite small that secondary particles will have the same transit time between successive collisions with the opposing surfaces as did their parent particles. The reasons for the unequal transit times are many fold. If the parent particle collides in a region of weak field some of its fsecondaries may be recaptured by an adjacent land 6 Others of the secondary particles may proceed to areas of stronger electric field wherein the particles are more rapidly accelerated thereby decreasing the transit time as compared with the parents transit time. Moreover, the distance the secondary particles must travel before striking an opposed surface is made non-uniform thereby preventing the uniform transit time condition necessary for multipactor from arising.

'The triangular shaped land 6 and groove embodiment of the present invention is especially useful in many high frequency devices as it is particularly easy to produce. For example, it has been found that multipactor action between a cylindrical quarter-wave slideably operating non-contacting tuning plunger operating within a circular waveguide can be eliminated by threading the out,- side surface of the non-contacting tuning plunger. In this particular application, a thread groove of 20 angle and threads to the inch suppressed multipactor satisfactorily when the plunger-to-waveguide spacing was 0.010".

In many electromagnetic devices, the electrical char.- acteristics of the device are a function of the lengthof the current path associated with the device. In very high frequency devices, the current path is confined more and more to the outside surface of the electrical conducting member. For example, in a copper conductor carrying currents alternating at a frequency of one million times a second, the current density has dropped to 36.8% of the value'at the surface at a depth from the surface equal to 0.0026", Thus if the surface has been grooved to prevent multipactor and the grooves run at some angle to the current path then the currents will have to flow over thecontour of the lands thereby substantially increasing the length of the current path and thus altering theelectrical characteristics of the device. On the other hand,if the grooves are oriented with respect to the current paths such that the current flows parallel to the grooves then the length of the current path has been substantially unaltered, and correspondingly the electrical characteristics of the device have not been altered except as to remove the multipactor action.

An example of the groove and current path orientation feature-of the present invention is shown in Fig. 4 wherein there is depicted a typical re-entrant type cavity resonator 7.3 is corn monly found in klystron tubes. Prqt ruding fromppositeend walls Softhe cavityresonator 7 are two to-entrant portions 9 of a drift tube 11. The mutually opposing free end portions of the re-entrant portions 9 have had their outside surfaces beveled to inhibit multipactor action. In addition, the land and groove embodiments of the present invention have been provided in the beveled surfaces, running longitudinally thereof, to further suppress multipactor action. Where the longitudinally directed lands and grooves intersect the internal bore of the drift tube 9, at the free ends thereof, there are formed a plurality of longitudinally directed fingers. The peripheral thickness of these fingers is preferably less than the gap spacing between similar mutually opposed drift tube portions, as shown in Fig. 4. Also, as shown, the length of the individual drift tube fingers is preferably less than A the gap spacing to prevent excessive field pertubation and loss of thermal conductivity of the drift tubes 9. In use, circulating currents flow in the interior surfaces of the cavity resonator 7. In the free end portions of the re-entrant portions 9, the currents run longitudinally of the re-entrant portions. Accordingly, the grooves have been placed in the surface of the re-entrant portions 9 such that the current fiows substantially parallel to the grooves thereby leaving unaltered the eifective length of the current path and the attendant electrical characteristics of the cavity resonator.

Referring now to Fig. 6 there is shown a further embodiment of the present invention wherein the surface has been permeated with a plurality of holes to thereby reduce multipactor action in use.

In operation, the hole permeated surface configuration acts much like the land and groove surface configuration previously described. Provision of the holes substantially decreases the surface area of the member which is closest spaced from another member. In this manner the electric field distribution is made quite non-uniform there being provided areas of extremely high electric field surrounded by adjacent areas of lower electric field. A particle emitted from the surface with some initial transverse velocity will quite likely proceed to an area wherein the electric field is substantially different than the electric field seen by its parent thereby disturbing the uniform transit time effects necessary for multipactor action. Moreover, if the holes are relatively narrow and relatively deep the probability that an emitted secondary particle emitted from within the hole portion will be recaptured is quite high. Thus the previously described relationship elf applies if the distance between opposing surfaces is large as compared to the thickness d of the surface portions remaining between holes.

Referring now to Fig. 7 there is shown another embodiment of the present invention wherein a honeycomb surface 12 is provided thereby substantially preventing multipactor. It, can readily be seen that the honeycomb surface 12 is a limiting case of the hole permeated surface configuration, previously described. In operation, the action of the novel honeycomb surface 12 is identical with the hole permeated surface configuration, Fig. 6.

In manufacturing, certain inherent advantages are to be found in the honeycomb surface 12 over previously described surfaces. For instance, at very high frequencies opposing members, in electromagnetic devices, are quite likely to have very small dimensions and, in addition, to have nonsymmetrical type surface contours. Great difficulty is likely to be encountered in trying to machine or otherwise provide grooves on such surfaces. However, the present honeycomb surface 12 provides a handy technique for providing the multipactor suppression surface configuration. The honeycomb structure itself is quite easily manufactured.

f Methods are well-known in the art for making ex- 'tremely fine, just barely macroscopic, honeycomb structures. One such method is fully described in U.S. Patent No. 2,619,438 of inventors S. F. Varian et a1. entitled, Method of Making a Grid Structure, issued November 25, 1952. Briefly, the method of making the honeycomb structure comprises the following steps: Individually coating a plurality of wires of a first metallic material with a relatively thin coat of a second metallic material; as sembling said coated wires in a sheath; drawing said assembly so that said coated wires are deformed into substantially hexagonal cross sections with the abutting surfaces of said coatings contacting one another and retaining their identities; cutting said drawn wires transversely to form a disc; removing, as by etching, the wires of first material from said discs and sintering the abutting surfaces among said coatings and said sheath so that they are coalesced. The honeycomb structure, once formed, may be adhered directly to the surface over which it is desired to prevent multipactor as by brazing or sintering.

If the surface, over which it is desired toprevent mul tipactor, possesses a-relatively small radius of curvature a slightly different technique may be utilized for securing the honeycomb structure to the surface. Using this technique the previously mentioned discs, before etching, are adhered as by brazing or sintering to a thin pliable sheet 13 of a suitable base material, as of, for example, copper. The honeycomb and attached base sheet 13 are then deformed as by pounding to match the contour of the surface over which it is desired to prevent multipactor. Thereafter, the first core material or wires may be removed from the honeycomb structure, as by etching, and the base sheet 13 may be sintered or brazed in place on the surface over which it is desired to prevent multipactor.

The order in which the previously described steps are performed is not too important. For example, it may be found that for a particular application or manufacturing situation it is more desirable to remove the core wire material as the last step in the operation. Moreover, it may be found that certain steps can be omitted altogether, for example, the honeycomb may be sintered or brazed to the surface of the member rather than to a pliable base sheet 13, thereby eliminating one step in the process.

In summary, the present disclosure in its broadest scope teaches methods and means for the suppression of multipactor action by the provision of irregularities in the surface of a member over which it is desired to prevent the formation of multipactor. Several examples of particular surface configurations, which are illustrative of the basic principle and which have been found particularly useful for various reasons, have been fully described herein. Finally, a technique was described wherein the surface irregularities may be oriented with respect to currents flowing in the member whereby the electrical characteristics of the particular device may be substantially unaltered.

It is readily apparent to those skilled in the art that an infinite number of surface configurations exist which incorporate the teachings of the present disclosure and which will satisfactorily prevent multipactor action. Since many changes could be made in the above construction of this invention and many widely apparently different embodiments of this invention could be made without departing from the scope thereof, it is intended that all matter contained in the above description or shown in the accompanying drawings shall be interpreted as illustrative and not in a limiting sense.

What is claimed is:

1. In an evacuated envelope containing two mutually spaced apart surface portions and having in use a high 7 mean b in ate than the us -be wee he re end portions'of adjacent land means for recapturing secgndary particles emitted from within the groove portions between land means, whereby in use multipactor action is effectively suppressed between the two spaeed apart surface portions.

'2. In an evacuated envelope containing two mutually opposed spaced-apartsurface portions having in use a high frequency alternating total electric field impressed therebetween of a magnitude and frequency tending to produce multipactor between the surface portions and in addition having high frequency electrical currents flowing in one of the spaced apart surfaces, land means formed in the current conducting surface portion, said land means running in a direction substantially parallel to the direction of high frequency current flow in the surface portion whereby in use multipactor action is effectively suppressed between the two spaced apart members and the current path length is not substantially altered.

3. In a high frequency evacuated apparatus employing re-entrant cavity resonators, a beveled end portion provided on the outside surface of a free end portion of a re-entrant portion of the resonator for reducing multipactor, and a land means provided on said beveled end portion to further inhibit multipactor, said land means oriented lengthwise of the cavity re-entrant portion whereby the circulating high frequency surface current path length is not substantially altered thereby minimizing resistive losses in the cavity resonator in use.

4. In an evacuated envelope containing two mutually spaced-apart member portions and having in use a high frequency alternating total electric field impressed therebetween of a magnitude and frequency tending to produce multipactor between the member portions, one member portion having a plurality of holes extending part way through the member whereby the plane area of the member portion closest spaced to the other member portion is substantially reduced and whereby particles emitted from said first member within the holes may be recaptured by the side and end walls of the holes thereby sup,- pressing multipactor action in use.

5. In an evacuated structure having two mutually spaced-apart surface portions and having in use a high frequency alternating total electric field impressed therebetween of a magnitude and frequency tending to produce' multipactor between the two surface portions, a honeycomb structure formed on one of the surface portions for disturbing the uniform secondary electron transit times and recapturing secondary electrons whereby multipactor action between the two surface portions may be substantially eliminated.

6. In an apparatus as claimed in claim wherein said honeycomb structure comprises a plurality of individual cells, and each cell having longitudinal dimensions over twice as great as their shortest transverse dimension.

7. A high frequency evacuated apparatus including, a cavity resonator, two axially aligned tubular drift tube members each carried at one end thereof from a side wall of said cavity resonator and the other free ends of said drift tube members extending into the interior of said cavity resonator forming a re-entrant portion thereof, the free end portions of said re-entrant drift tube members defining an interaction gap therebetween for coupling the high frequency electromagnetic fields of the cavity resonator to a beam of charged particles passable through said drift tube members and tending to multipactor across the gap in use, said re-entrant drift tube members being beveled at the free end portions thereof to form free end tubular portions tapered in radial thickmess, the thickness of said tubular drift tube members decreasing toward the free end thereof, said tubular reentrant drift tube portions being provided with longitudinal slots therein at the free ends thereof to form a plurality of mutually opposing longitudinally oriented 8 fingers of decreasing radial thickness toward the free ends thereof, whereby in use-multipactor action is elfectively suppressed between the opposing free end portions of said tubular drift tube members. a

8. The apparatus according to claim 7 wherein the peripheral thickness of said longitudinally extending tapered drift tube fingers is less than one-tenth the gap distance between corresponding portions of mutually opposing drift tube fingers.

9. A high frequency evacuated apparatus including, a cavity resonator, two axially aligned tubular drift tube members, each of said drift tube members carried at one end thereof from a side wall of said cavity resonator and the other free ends of said drift tube members extending into the interior of said cavity resonator forming a .re-entrant portion thereof, the free end portions of said re-entrant drift tube members defining an interaction gap therebetween for coupling the high frequency electromagnetic field of said cavity resonator to a beam of charged particles passable longitudinally through said drift tube members, said re-entrant drift tube members being beveled at the free end portions thereof to form free end tubular portions tapered in radial thickness, the radial thickness of said tubular drift tube members decreasing toward the free end thereof, said tubular reentrant drift tube portions being provided with longitudinally directed grooves in the beveled portions of said drift tubes to form a plurality of mutually opposing longitudinally directed lands and grooves, said grooves intersecting the internal'bore of said tubular drift tube members substantially at the free ends thereof to form a plurality of longitudinally directed fingers at the free .ends thereof, the peripheral thickness of said longitudinal extending fingers being less than one-tenth the gap distance between corresponding similar portions of mutually opposing drift tube fingers whereby in use multipactor action is effectively suppressed between the opposing free end portions of said tubular drift tube members.

10. The apparatus according to claim 9 wherein the length of said longitudinally directed fingers is less than one-fourth of the gap distance between the mutually opposed free end portions of said re-entr'ant drift tube members whereby the thermal conductivity of said re-entrant drift tube members remains substantially unimpaired and the desired electromagnetic field patterns within the gap are not excessively perturbed.

11. The apparatus according to claim 9 wherein said drift tube members are relatively thick walled having a maximum wall thickness greater than one-half the radius thereof whereby thermal energy is readily conducted therethrough to minimize overheating thereof in use.

12. The apparatus according to claim 10 wherein the length of, said fingers is greater than one and one-half times the peripheral thickness of said grooves between adjacent fingers whereby recapture of secondary emitted electrons is enhanced within the spaces between adjacent longitudinally extending fingers. 1

References Cited in the file of this patent UNITED STATES PATENTS 1,859,875 Kern May 24, 1932 1,952,493 Edelmann Mar. 27, 1934 2,325,865 Litton Aug. 3, 1943 2,460,141 McArthur Jan. 25, 1949 2,480,133 Hansen Aug. 30, 1949 2,499,977 Scott Mar. 7, 1950 2,515,997 Haeff July 18, 1950 2,530,703 Jonker NOV. 21, 1950 2,568,325 Diamond Q. Sept. 18, 1951 2,607,016 Kennebeck Aug. 12, 1952 2,619,438 Varian Nov. 25, 1952 2,680,209 Veronda June 1, 1954 2,687,490 Rich et al. Aug, 24, 1 954 2,697,800 Roberts Dec-21, 1954 2,741,718 Wang Apr. 10, 1956

Images