US2572414A - Magnetic induction accelerator - Google Patents

Description

R. WIDERDE MAGNETIC INDUCTION ACCELERATOR oer. 23, 1951 Filed Dec. 10, 1947 Patented Oct. 23, 1951 UNITED STATES PATENT OFFICE Application-December 10, 1947, Serial N0. 790,912 In Switzerland December 11, 1946 7 Claims.

This invention relates in general to devices for accelerating charged particles such as a stream of electrons to high velocity and hence high potential on a circular orbit and in particular to an improved arrangement for improving the stabilization of the electron stream at the time when the stream is first injected into the device, at the time that the stream is. removed after it has reached its final velocity, or both.

In the accompanying drawings, Fig. I is a view in central vertical section through an electron accelerator embodying the present invention; Fig. 2 is a diagrammatic view illustrating a conventional arrangement for shifting the electron orbit of an induction type accelerator; Fig. 3 is a diagrammatic view illustrating the approximate electron orbit and the orbitalshift effected by the arrangement of Fig. 2; Figs. 4-6 are explanatory curves relating to the type of orbit shift provided by the Fig. 2' arrangement; Figs. 7 and 7a are diagrammatic views illustrating the approximate electron orbitand type of orbital shift provided. by the present invention; Fig. 8 is: a diagrammatic view illustrating the arrangement of the auxiliary, orbit shifting coils included inthe Fig. 1- construction; Figs. 9 and1l0 are fragmentary viewsin vertical central section illustrating modified forms of the invention; and Fig. 1 1 is aschematic wiringdiagram of the control circuit for energizing the auxiliary coils in Fig. 1 bywhich the shift in the electron orbit is effected.

The invention is applicable to various types of electron accelerators but in this application is illustrated and described in its relation toan accelerator operating on the magnetic induction principle. The latter are now generally known as betatrons or ray transformers and a typical. construction is illustrated in the diametral section view of Fig. 1. The betatron is comprised of a magnetic field structure I!) made up from steel-laminations of appropriate contour to provide a pair of cylindrical poles Hl l separated by air gap l2 and located concentrically along axis a.a, and a pair of concentric annular poles l3l3' facing one another and separated by air gap l4. Yoke members complete the magnetic' circuit for a cyclically varying flux set up in the annular and cylindrical poles. Poles l|l and l3-l3 are surrounded by an annular winding preferably split into two coil sections Iii-16 connected in series for energization from a source of alternating current of suitable frequency as for example 100 cycles/sec. applied to terminals H. In accordance with the usual practice, a condenser 31 is also connected in shunt with the coils |6|-6 across terminals II.

An annular evacuated glass tube l8 rests in the air gap [4 between the poles |3-l3' and thereby surrounds the axial poles I I--l I Located within the tube is an electron emissive cathode which can be of the thermionic type such as illustrated by the coiled filament 20, the axis of which is placed parallel to axis a-a. The electron stream to be accelerated is of course produced at the cathode 20 and after it hasbeen accelerated to its final velocity along the circular orbit 7c which occurs when the current reaches its maximum value the stream can be caused to impinge upon a target anode 2| to produce X-rays, or the stream can be removed from the tube for other purposes. In the illustrated construction, cathode 20 is located radially inward from the circular orbit is and the target anode 2| radially outward of such orbit but these can be reversed in position, or both of them even located on the same side of the orbit k. Neither can, however, be located on the orbit.

By a control arrangement which is well known and for such reason has not been illustrated, the cathode 20 is energized periodically in timed relation with the alternating current wave applied to the coil sections I6, I6 to produce a stream of electrons at just about the instant that the wave passes through zero. The stream upon reaching the circular orbit is undergoes a constant acceleration under the influence of the magnetic field produced in the field structure [8- by'the applied alternating current. This field divides into essentially two operating components. One of these, the inducing field component acting axially through poles IlH may be Viewed as being responsible for the acceleration of the electron stream;. the other component commonly referred to as the control field which acts axially through the annular poles l3l3' produces a centripetal efiect upon the electron stream to ofiset the centrifugal forces of the electrons caused by their motion along the circularorbit k. Throughout the increasing acceleration of the electron stream, the centripetal force which likewise increases substantially matches the centrifugal forces of the electrons with the result that the stream isv confined to a circular orbit of constant radius. This is the orbit k.

For shifting the equilibrium orbit such as to remove the. electron stream from orbit is after it has been accelerated in order to strike the target anode 2i it is already known to employ two electrostaticelectrodes located within the plane of 3 the circular orbit and between which a suitable voltage is applied at the proper instant. Such an arrangement is illustrated in my co-pending United States patent application, Serial No. 751,680, filed June 2, 1947, now U. S. Patent 2,533,859.

It has also been proposed in order to obtain such a translatory displacement to use a crosswound coil 25 as illustrated diagrammatically in Fig. 2 that creates opposite flux directions at the two points where the orbit diameter parallel to the direction of displacement R intersects the circular orbit is and to vary the current through such coil from its initial value to a higher or lower value. However, such a translatory displacement of the orbit entails considerable disadvantages as far as stabilization conditions are concerned, as will now be explained with refer-' ence to Fig. 3. Here orbit k which approximates a circle'represents the original position of the orbit and orbit k the same orbit after displacement. Orbit k intersects a dashed line B designating the desired direction of displacement in two points A and B. The stabilizing force on the electron stream produced by the magnetic control field component previously discussed functions in such a manner that any radiall inward or outward deviations of equal magnitude of the electrons from the equilibrium orbit give rise to equal stabilizing forces. Hence if such force X is plotted against the radius r of the equilibrium orbit, the conditions at the points A and B of Fig. 3 will be represented by the graph shown in Fig. 4 from which it will be seen that the deviations of orbit k from its prescribed radius 1'0 in the positive and negative sense, respectively, will give rise to oppositely directed but like varying stabilizing forces tending to force the electrons back" into theoriginal orbit.

Consider now the condition in which the orbit is displaced to orbit k. The stabilizing forces at the points C and D in which orbit kintersects the dashed line R are no longer the same as in Fig. 4; The forces at point C as regards departures from T are now as shown in the graph of Fig. 5 While those at point D vary acccording t0 'the graph of Fig. 6. This is so because the field of the coil 25 in Fig. 2 causes the relative position of the equilibrium orbit in the stabilizing field to be displaced so that Fig, 4 is transfigured into Fig. 5 as regards point C and into Fig. 6 as regards point D. In Fig. 5, the stabilizing force in the positive direction of r varies linearly with r for a wide change in 1', but in the negative direction, the force varies linearly throughout only a small change in 1', and then begins to reverse itself in a vectorial sense. In Fig. 6, conditions converse to those in Fig. 5 prevail.

Stabilizing force conditions of the character shown by the graphs of Figs. 5 and 6 are highly undesirable because the maximum stabilizing force available in one direction from To may be exceeded upon the occurrence of a relatively small deviation of the electrons from the circular orbit in that direction whereupon such electrons will no longer be restored to the radius 10.

In accordance with the present invention, the foregoing undesirable operating characteristics are eliminated and satisfactory stabilizing conditions obtained upon injection of the electron stream into the tube [8, upon its ejection following acceleration thereof, or both. Stated generally, the improved result is obtained by effecting -a change in the radius of the equilibrium orbit simultaneously with the displacement of its center such that the old and new orbits pass at least approximately through the same point at one place in tube l8.

From a consideration of Fig. 7 it will be easily appreciated that this results in an improvement over Fi 3. The old orbit in Fig. 7 is designated ICE. and the new i. e. displaced orbit by kb. For example, let it be assumed that to eject the electron stream from the orbit ks. upon which it has been accelerated to final velocity, the orbit is shifted to orbit kb. The target anode 2| can then be located in or adjacent to the point E while at the point F the stabilizing conditions are the same for both orbits he. and kb. On the other hand, if We assume the electron source to be located at point E, then in such case, orbit kb would represent the old orbit and orbit ka the new or displaced orbit. Accordingly, the dis-advantages explained with reference to Figs. 3, 5 and 6 will only occur to their full extent in such positions where it is unavoidable because of the positions of the electron source, of the target or both which, as will be directly understood, must necessarily be situated outside of the equilibrium orbit in which the electrons move during their period of acceleration.

, One constructional form of the invention is illustrated schematically in Fig. 8. Here the control poles I3, I 3 are seen to be surrounded by an auxiliary coil 30. This winding is energized through conventional controls, as exemplified by Fig. 11, at the proper instant when displacement of the electron orbit is desired and the turns of the winding are varied circumferentially of the orbit 76a in such manner that the strength of the magnetic field produced thereby and which is superimposed upon the main control field acting through the control poles varies along the circumference of the orbit ks. approximately in accordance with the equation where a is a positive or negative constant, and $0 the azimuth angle. For the construction of the coil shown in Fig. 8 the constant a is equal to unit. If a=+l the changed equilibrium orbit .701 will coincide with the initial orbit at one point such as the point F of Fig. 7a and its radius Will be enlarged and shifted as explained with respect to Fig. 7a, whilst with a=1 the radius of the orbit will be decreased as shown in Fig. '7, in an analogous manner. ,Fig. 1 illustrates the position of the turns of coils on the control poles l3-l3 assuming the vertical section there shown were to be taken along the line YY of Fig. 8.

As explained above, use can be made of conventional control circuits for energizing coils 30 at the proper instant, and one such control is shown in Fig. 11. Here the coils 39 are seen to be connected in series with a variable resistor 36 and a thyratron type of switching tube 33 to terminals 38 to Which a source of direct current designated by the conventional and symbols is applied. The voltage between the grid 33a, and cathode 331) the amplitude of which determines the instant at which the tube fires i. e. reduces the impedance across the cathode33b, anode330 to a nil value so that current can flow freely therethrough, is supplied from the secondary of transformer 34. The primary of this transformer is connected across the alternating current terminals I! through a phase shifting device 35. The circuit is so arranged that when the alternating current through coils I6I6' reaches its maximum value, at which time the electron stream has also reached its maximum energy level and must be removed, the voltage impressed upon the con trol grid 33a will have. likewise and simultaneously attained a value sufficient to fire the tube thus causing a flow of current through the coils 39.

The amplitude of the current in coils 30 can be adjusted by means of the Variable resistor 3E.

The phase-shifting device 35 is of course needed because the current wave in coils i6l6 lags substantially 9O electrical degrees behind the alternating voltage wave at terminals ll due to the cross-wound coil 25 of Fig. 2 can be used in con-- junction with one of the known betatron constructions in which provision is made for changing the diameter of the orbit at the desired instant such as for example at the time when the fully accelerated electron stream'is to be ejected from the orbit. In the latter arrangement, the desired result is obtained by varying the coil current so that oppositely directed fields are created at diametrically opposite points of the circular path while simultaneously altering the radius of the path. Two constructions for accomplishing the latter are illustrated respectively in Figs. 9 and 10.

In Fig. 9, the coils 25 are shown applied to the control poles l3l3 in the manner illustrated schematically in Fig. 2, it being assumed that the vertical section of Fig. 9 is taken in a plane coincident with the diameter R in Fig. 2. These poles are each seen to include a section 3! of reduced cross-section such that the control poles will begin to saturate towards the end of the accelerating period. As soon as pole saturation begins, the normal ratio between the control and inducing field which has been maintained up to that instant to confine the circling electron stream to an orbit of constant radius likewise begins to change. Specifically, the inducing field now begins to increase at a greater rate than does the control field with the result that the radius of the electron path begins to enlarge.

On the other hand when it is desired to decrease the radius of the orbit as already mentioned, this can be done as described in U. S. Patent No. 2,297,305, issued to D. W. Kerst, September 29, 1942, by dimensioning the central induction core so as to create a saturation of the central induction fiux and thus change the proportion between this flux and the control field flux in an opposite manner to that required for an increasing orbit radius.

Assuming that in Fig. 2 the electrons circle the orbit 7c, corresponding to the orbit ka in Fig. 7, in a counter clockwise direction and the current in coil 25 is changed from zero to such value that the field directions indicated in Fig. 2 obtain, then towards the end of the acceleration period, the electron stream will be shifted from orbit he. to an orbit corresponding to orbit kb of Fig. 7. Thusa coil of the type shown in Fig. 2 applied to saturable control poles brings about a result similar to that obtained with the coil 38 of Fig. 8 when used with control poles of uniform cross-sectional area as shown in Fig. 1. The control circuit for energizing winding 25 at the proper instant could of course be the same as that shown in Fig. 11 for energizing windings 30.

Fig. illustrates another practical arrangement. Here the cross-wound coil of Fig. 2 is applied to each of the control poles |3l3 in the 6 same manner as explained with respect to Fig.- 9. However, instead of employing a saturation e'fiect in the control pole to enlarge the electron orbit, provision is made for introducing a phase displacement between the control field and the inducing field. A short-circuited conductor ring 32 imbedded in the face of pole l I will serve this purpose as explained more in detail in U. S. Patent No. 2,297,305 issued to D. W. Kerst, September 29, 1942. Displacement in phase between the control and inducing fields is also discussed in Swiss Patent No. 254,937.

In conclusion, I wish it to be understood that the principles underlying the present invention are not limited in application to electron accelerators operating wholly on the magnetic induction principle. The invention can be applied equally as well for'similarly changing the electron orbit in the well known synchrotron type of electron accelerator Where initial acceleration of the electron stream to an energy level of .the order of two million electron volts is effected through use of the magnetic induction principle and further electron acceleration is produced by high frequency electrostatic fields. In this case, the desired change in the orbit radiusiincrease or decrease) can be obtained by reducing or increasing the frequency of the accelerating voltage relative to its original value.

I claim:

l. The combination with a charged particle accelerator of the type wherein at least an initial acceleration of the particles is effected through magnetic induction, said accelerator including an annular tube within which the particles may follow an orbital path, meansproviding charged particles within said tube, a magnetic structure comprising inducing field pole means extending axially through said tube and a pair of annular control field poles arranged in confronting relation at and on opposite sides of said tube, and a magnetizing winding on said structure adapted to be energized by a cyclic, time varied current to produce a corresponding time varied magnetic field in said poles and which causes said particles to be accelerated around said tube along a path of substantially constant radius; of means including cross wound coil means disposed on said control poles and which are energized at a predetermined instant in the cycle of said time varied magnetic field for eifecting a change in radius of said path and simultaneously displacing the path center along a diameter thereof.

2. The combination with a charged particle accelerator of the type wherein at least an initial acceleration of the particles is efiected through magnetic induction, said accelerator including an annular tube within which the particles may follow an orbital path, means providing charged particles within said tube, a magnetic structure comprising inducing field pole means extending axially through said tube and a pair of annular control field poles arranged in confronting relation at and on opposite sides of said tube, and a magnetizing winding on said structure adapted to be energized by a cyclic, time varied current to produce a corresponding time varied magnetic field in said poles and which causes said particles to be accelerated around said tube along a path of substantially constant radius; of coil means disposed on said control poles and which when energized produces an auxiliary magnetic field varying in the circumferential direction of said path and having maximum and minimum values,

respectively at substantially diametrally opposite points on said path, and means energizing said coil means at a predetermined point in the cycle of said time varied magnetic field to thereby change the path radius and simultaneously displace the center of said path along a diameter thereof.

3. An accelerator for charged particles as defined in claim 2 wherein said auxiliary magnetic field varies circumferentially in accordance with the equation:

where Afi represents the varying field strength, Aflm and a constants, and go the polar angle.

4. The combination with a charged particle accelerator of the type wherein at least an initial acceleration of the particles is effected through magnetic induction, said accelerator including an annular tube within which the particles may follow an orbital path, means providing charged particles within said tube, a magnetic structure comprising inducing field pole means extending axially through said tube and a pair of annular control field poles arranged in confronting relation at and on opposite sides of said tube, a magnetizing winding on said structure adapted to be energized by a cyclic, time varied current to produce a corresponding time varied magnetic field in said poles and which causes said particles to be continuously accelerated around said tube along a path of substantially constant radius; of means operating in timed relation with said magnetic field for changing the radius of said path,

and means simultaneously displacing the center 3 of said path along a diameter thereof, said center displacing means including cross-wound coils on said control poles which when energized produce auxiliary magnetic fields of opposite directions at the two points where the path diameter coincident with the direction of its displacement intersects the path, and means operating in timed relation with said time varied magnetic field for energizing said auxiliary field producing coils.

5. An accelerator for charged particles as defined in claim 4 wherein the center of said path is displaced through a distance substantially equal to the change in radius whereby the initial and changed paths coincide at one point in said tube.

6. An accelerator for charged particles as defined in claim 4 wherein the means for changing the radius of said path are constituted by sections of reduced area in each of said control field poles and which are adapted to saturate at a predetermined point in the cycle of said time varied magnetic field.

'7. An accelerator for charged particles as defined in claim 4 wherein the radius of said path is changed by means effecting a phase displacement between the control and inducing field components of said time varied magnetic field.

ROLF WIDERGE.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 2,297,305 Kerst Sept. 29, 1942 2,394,070 Kerst Feb. 5, 1946

Images