US2797322A - Magnetic induction electron accelerator - Google Patents

Description

June 25, 1957 Filed Aug. 1a, 1953 A. VON ARX 2,797,322

MAGNETIC moucnon- ELECTRON ACCELERATOR 3 Sheets-Sheet 1 ATTORNEYS lNVENTO R 3 June 25, 1957 A. VON ARX MAGNETIC INDUCTION ELECTRON ACCELERATOR Filed Aug. 18, 1953 3 Sheets-Sheet 2 INVENTOR June 25, 1957 A. VON ARX N MAGNETIC INDUCTION ELECTRON ACCELERATQR Filed Aug. 1 8, 1953 6 INVENTOR BY J MKM ATTORNEYS MAGNETIC INDUCTIUN ELECTRON ACCELERATOR Arnold von Arx, Baden, Switzerland, assignor to Aktiengesellschaft Brown, Boveri & Cie, Baden, Switzerland, a joint-stock company Application August 18, 1953, Serial No. 374,930

Claims priority, application Switzerland August 19, 1962 2; Claims. (Cl. 250-27) The present invention relates to electron accelerators and in particular to those of the magnetic induction type wherein electron streams are accelerated on a generally circular orbit within an evacuated tube under the combined influence of a magnetic field varying with time and which field has two components. One component of the field, known as the acceleration flux which passes through the tube inside of the orbit causes the electron stream to be accelerated. The other component, known as the control flux passes through the tube at the orbit and functions to maintain the electron stream on the orbit.

in an induction accelerator, the electrons are conducted out each time at the end of an acceleration period, for example, by an enlargement of the equilibrium circuit (expansion) out of the circular path passed through during the acceleration and, for example, for the production of X-ray radiation, they are directed to an anticathode. The expansion is efiected mostly by changing the relationship between the magnetic flux accelerating the electrons and the control flux retaining the electrons on the circular path. These two fluxes can be produced by a single exiter coil fed from alternating current mains. To change the relationship between the two fluxes a special coil (expansion coil) can be provided which changes at least one of the said fluxes when current flows through it. The feeding of the expansion coil takes place by way of brief current impulses which are switched in at the desired moment.

An advantageous circuit arrangement for the production of such expansion impulses is given in the United States Patent No. 2,654,838. According to the arrangement of this patent, a condenser is discharged each time at the desired moment over the expansion coil which is arranged so that it encompasses the acceleration flux. The recharging of the condenser is performed with the voltage induced by the acceleration flux in the expansion coil. This circuit arrangement is suitable for the production of expansion impulses which occur each time the electron stream reaches maximum energy. It can be used with special advantage in an accelerator in which the electrons are accelerated alternately in both directions of circulation (two-ray accelerator), because the impulses produced by the circuit also are of alternating direction.

It could be desired however to change the instant of the occurrance of the expansion impulse arbitrarily so that the electrons can be conducted out of the circular path also with a lesser energy than the maximum one. For the practical use of a two-ray accelerator it is particularly frequently advantageous if the electron energy of the one ray can be adjusted arbitrarily, while that of the second ray always remains close to the maximum value. This can be done in that the electrons of the first ray, by an expansion impulse occurring at the appropriate instant, are removed out of the circular path before they have reached their maximum energy.

The present invention relates to a device for production of expansion impulses for an induction accelerator workice ing in this way. In this, by means of a single regulation unit, both the instant corresponding to the desired electron energy and also the inherent amplitude of the expansion impulse are adjusted. This amplitude always has to be proportional to the control flux prevailing at the moment of the expansion. Therefore, if i designates the amplitude of the expansion impulse needed for the expansion at maximum electron energy, then the expansion impulse occurring at one arbitrarily chosen phase angle car must have the amplitude i=i sin car. If the acceleration of the first ray always starts at m=0, then, by shifting the phase of the expansion impulse between the values a =0 and a,=, any arbitrary electron energy can be attained between zero and its maximum value. The acceleration of the electrons of the second ray always starts at x=180, its removal from the circular path by an expansion impulse occurring at least approximately at a=270.

The device according to the invention contains an expansion coil which encompasses the acceleration flux of the induction accelerator and a condenser which is connected each time with the expansion coil over one of two thyratrons switched in parallel with opposite conducting direction. The device is characterized in that means exist which initiate the firing of the first thyratron each time with the arbritrarily adjustable phase 0: associated with the desired electron end energy of the first ray, wherein the conductivity of this thyratron remains unlocked up to at least the phase 180, and which cause the firing of the secondjhyratron at the phase (270+A) wherein sin A= /2[ /2.sin (a1+45)1].

The invention will become more apparent from the following description of a preferred embodiment thereof and the accompanying drawings wherein:

Fig. 1 is a condensed schematic electrical diagram showing the general arrangement of the principal circuit components and their connection used to attain the objective;

Figs. 2 and 3 are graphs showing the variation with time of the voltage components involved as well as a plot of the expansion current pulses;

Fig. 4 is a graph showing the variation with time of the two components of the voltage applied to the grid of one of the thyratrons;

Fig. 5 is an electrical circuit diagram illustrating one suitable arrangement in accordance with the invention for obtaining the voltage pulses utilized to fire or ignite the thyratron tubes which control flow of current pulses to the expansion coil;

Fig. 6 is a vector diagram showing the phase relationship between the voltage components which make up the phase shifting device utilized in the invention; and

Fig. 7 is a plot of the voltage curve used to control ignition of one of the thyratron tubes.

With reference now to Fig. 1 which illustrates an application of the invention to a magnetic induction accelerator of conventional construction, the magnetic field structure thereof, indicated at 10 is made up of steel laminations of appropriate contour to produce a pair of cylindrical poles 11-11 separated by an air gap 12 and located concentrically along axis a-a, and a pair of concentric annular poles 13-13' facing one another and separated by air gap 14. Yoke members'15 complete the magnetic circuit for a cyclically varying flux set up in the cylindrical and annular poles. Poles 11-11 and 13-13' are surrounded by an annular winding which may as shown be split into two coil sections 16-16 connected in series for energization from a source of alternating current of suitable frequency as for example cycles/sec. applied to terminals 17.

An evacuated glass tube 18 rests in the air gap 14 between the poles 1313' and thereby surrounds the axial poles 11-11'. 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 produced at the cathode by energizing means actuated in timed relation with the time-varied alternating current input at terminals 17 to effect emission of an electron stream from the cathode at the begining of each half of the flux wave produced by such current. After the electron stream has been accelerated to its final velocity along the circular orbit k, which occurs when the magnetic flux approaches or reaches its peak value, it can be caused to impinge upon an anti-cathode 21 to produce X-rays, or the stream can be removed for other uses. Since an electron stream is injected at the beginning of each half of the magnetic flux cycle two electron streams will be produced for each complete flux cycle and these are accelerated in opposite directions around the orbit k in succession.

The componentsthus far described are conventional. This invention is directed to an improved arrangement for removing the electron streams of opposite accelerating directions from the orbit k utilizing a coil L which is applied to the central poles 1111 and thus surrounds only the acceleration component of the flux produced by the winding 16-16. Coil L is pulsed at a selected instant during the accelerating period of one electron stream and near the close of the accelerating period of the other electron stream and the effect of the pulsing is to upset the ratio between the accelerating and control flux components normally maintained during the accelerating phase to keep the electron stream moving around an orbit of substantially constant radius. That is, the flux produced by pulsing of coil L, adds to the acceleration component of the main flux and causes the orbit to expand with the result that the electron stream moves outward from the orbit on which it has been accelerating to thus strike the anti-cathode 21. The alternating current voltage induced in coil L by the accelerating flux is indicated by Ur... As is clear from Fig. 1 a condenser C is connected with the coil L through switching means comprising a pair of grid controlled gas discharge tubes V1V2 known as thyratrons arranged in parallel in back-to-front relation. Since these tubes will conduct current in one direction only, the baek-to-front connection enables one to discharge condenser C through coil L once in each half cycle in alternate directions by appropriate control over their grids as will be explained hereinafter in more detail.

With reference to Fig. 2, it is assumed that acceleration of the electron streams of opposite circulation directions begins each time that the phase a of the magnetic flux produced by main winding 1616 equals or 180", respectively. Hence the electron streams will reach their greatest energy level at phase a=90 or a=270, respectively. In accordance with the present invention arrangements are made to expand or remove the electron streams from their orbit in a selective manner such that expansion pulses applied to coil L for removal of one stream can be effected at any selected angle a1 between 0 and 90 while removal of the other stream can be eifected by an expansion pulse at an angle which differs from u=270 by only a small value A. As previously explained, these expansion pulses are produced by discharging condenser C through coil L first in one direction and then the other via thyratrons V1, V2 which are fired or ignited in alternation. The condenser discharge process exhibits the form of a weakly damped oscillation as a result of losses existing in the circuit which is interrupted after its first half period because the thyratron permits current flow in only one direction. Each discharge of condenser C thus produces a brief current impulse in the expansion coil L.

Fig. 2 shows that a first discharge of the condenser C over the coil L occurs by ignition of the thyratron V1 at the phase a1. Immediately before this discharge, the voltage Uc at the condenser C has the value Ur. At the moment of the discharge, the coil L shows a voltage U2 produced by the acceleration flux. Since the half period of the free oscillation is passed practically free of loss, the voltage differences a and b are almost equally large. After the discharge, the voltage on the condenser remains at the value U3. Somewhat later, the voltage induced in the coil L also reaches the value U3. Since the thyratron V1 is still blocked according to the invention, the condenser is charged by the current supplied from the coil L to the negative peak value of the voltage UL. Since the thyratron does not permit a current flow in reverse direction, this voltage U5 remains on the condenser, even when the induced voltage UL approximates the zero value again. The thyratron V1 is blocked again during passage of the phase x=180.

The second discharge of the condenser C over the coil L by ignition of the thyratron V2 occurs during the phase u2=270+A. The voltage differences 0 and d are almost equally large; after this discharge the condenser again shows the voltage U1 and the described processes repeat. Therefore, in the coil L, the expansion impulses i occur which are shown on the lower axis of Fig. 2.

The size of A is chosen according to the invention so that the amplitude of the expansion impulse for the first ray or electron stream obtains just the right value proportional to the control flow at the moment of the expansion. The difference between the voltages U1 and U1. (oil) has to be proportional in this to the control flow, i. e. it must be proportional to sin a1. Fig. 3 shows a period of the processes already depicted in Fig. 2, drawn on a larger scale. It can be seen from this, that for the value a, which should be proportional to sin on, apart from the proportionality factors there applies:

a=l+2.sin A-cos a1=SiI1 on from which follows:

sin A== /2[ /E.sin (oc1+45)1] The expression for the second ray therefore does not occur exactly at 270 deg. but at a=270-l-A, and the expansion value for the second ray is proportional to the term (l-l-sin A). These two deviations from the ideal values are unimportant in practice, however: the reduction of the electron energy of the second ray as a result of expansion outside of its maximum amounts to approximate 2% at most, and the expansion fiux for the second ray diifers at most for approximate plus or minus 10% from its theoretical value if the phase of the expansion impulse is changed for the first ray between the values a1=0 and u1=90.

In the production of the voltages to be conducted to the control grids of the thyratrons V1, V2.

The grid voltage Uez for the thyratron V2 consists of the components )1 and f2. In this,

wherein the quantity f1 never becomes positive, however, but equals to zero at most.

Furthermore, it is true that:

becomes zero. This is the case only in the proximity of alpha equals 270, when f2=O and f1=0, i. e. at

therefore according to the invention at a,=270+A.

A device for the production of the grid voltage for the thyratron V2 is contained in the embodiment shownin Fig. 5. The component fl is obtained as a sum, firstly the amount 2.00s a from the secondary side (S1+S2) of the transformer T1, secondly the amount 1 which, for its part, is produced to the peak value by the rectification of the amount cos a from the secondary winding S2 in the rectifier G1 and charge of the condenser C1, and thirdly the amoun \/2 sin (a,+45), which is produced to the peak value by the rectification of the amount \/l.sin (a,+45).sin (t-Ot +'y) from the winding S1 of the transformer T2 in the rectifier G2 and charge of the condenser C2. To the primary winding P of the transformer T2 a voltage is conducted whose amplitude E is changed by arbitrary adjustment of the phase shifter P to the desired value of a, according to the function:

E=2.sin (m -P45). This phase shifter will be discussed further yet subsequently in more detail. The rectifier Ga, together with the resistance R1, prevents the occurrence of positive values of the component ii.

The component in is obtainedby a combination of the resistance R2 effecting a phase shift, and of the condenser C3, from the amount 2.cos on of the secondary side (s,+s,) of the transformer T1. The rectifier G4 prevents the occurrence of positive values of the component f2 forming on the resistance R3.

The phase shifter P, through Whose activation the phase a, of the first expansion impulse and thereby the desired electron energy is adjusted in the first ray, is produced by the voltage conducted to the primary side of the transformer T2 as the sum of a voltage E1 with fixed phase and a voltage E2 of the same amplitude with variable phase (Fig. 6). In order that the amplitude of the resulting voltage E follows the function given above, the phase angle of the voltage E2 has to be changed over a range of 180 if the phase a, of the expansion impulse is to be shifted in the range between 0 and 90. The phase shifter supplies a voltage of the form:

E:2.Sin (a +45).Si11 (Dt-Ot +'y) The grid voltage U61 for the thyratron V1 is UGi sin (oc,+45).(2.$i11 (otoc +'y)l) If 'y=30 is set, then this voltage, originating from negative values, reaches the value zero at a=a,, i. e. the ignition of the thyratron V1 takes place in the desired phase 0a,.

A device for the production of the grid voltage for the thyratron V1 is also contained in the embodiment shown in Fig. 5. The grid voltage appears as sum of the constant amount sin (a,+45), which is obtained from the winding s, of the transformer T2 by rectification in the rectifier G and charging of the condenser C4 to the peak value, and the amount which originates from the windings (s +s,) of the transformer T2.

The course of this grid voltage is shown in Fig. 7 on the same abscissa scale as Figs. 3 and 4. It can be seen from this that the ignition of the thyratron V1 takes place in the phase a, and that the grid voltage of this thyratron subsequently, during a third period, does not become negative. The thyratron therefore can ignite again when the voltage UL reaches the value V2 (Fig. 2) and it extinguishes finally only when, after charging of the condenser C (Fig. l) to the negative peak value of the voltage Ur. in the phase 180, its anode voltage becomes negative.

In conclusion while the described device for the production of expansion impulses has been applied to a magnetic induction accelerator it can also be used on other electron accelerators, for example synchrotons. The expansion coil encompasses in this, instead of the acceleration flow lacking in these apparatuses, at least a part of the control flux.

- I claim:

1. The combination with an electron accelerator of the type comprising an annular tube into which streams of electrons are injected for acceleration on an orbit established within the tube, a magnetic structure including a central core portion extending through the central opening within said tube and annular control pole portions confronting one another at said tube, a main winding on said magnetic structure surrounding said control poles, and a source of alternating voltage connected across said main winding to establish an alternating current wave therein producing a cyclically varying magnetic field of wave form comprising a first accelerating component through said central core efiecting acceleration of said electrons and a second control component through said control poles confining the electron streams during the accelerating period to a path of travel along an orbit of substantially fixed radius, successive electron streams being accelerated in opposite directions around said orbit and being introduced respectively for acceleration at phase a=0 and 180 wherein or. is the instantaneous phase of the magnetic field wave; of means for producing ex pansion pulses for effecting a removal of said electron streams from said orbit following their acceleration, said means comprising, an expansion coil surrounding only said central core portion and which is inductively coupled with said main Winding so as to cause an alternating voltage to be induced therein, a condenser, first and second grid controlled gaseous discharge valves connected in parallel in back-to-front relation, means connecting said expansion coil, condenser and paralleled valves in series, and means producing grid voltages for effecting ignition of said valves in alternation in timed relation with the phase of said magnetic field wave to thereby discharge said condenser alternately in opposite directions through said expansion coil and change the ratio between said accelerating and control components of said magnetic r field thereby to remove said streams of electrons from said orbit, said means for producing the grid control voltage for said first valve associated with removal of the electron stream introduced at phase or equal to 0 comprising a phase shifting device fed from the same source as that feeding said main winding, a first transformer, a secondary winding of which is connected to the grid of said first valve and the primary winding of which is connected to said phase shifting device and supplied therefrom with a voltage E equal to 2.sin (a,+45), said phase shifting device being adjustable to effect ignition of said first valve at a phase a, between 0 and and said means for producing the grid control voltage for said second valve associated with removal of the electron stream introduced at phase at equal to eifecting ignition of said second valve at a phase at, equal to 270+A wherein sin A=%.[\/2.sin (a l45)l] comprising a second transformer, the primary winding of said second transformer being fed from the same source as that feeding said main winding, a first rectifier connected with a secondary winding of said first transformer and a first condenser charged by said first rectifier to a first voltage proportioned to the amplitude of said voltage E, a second rectifier connected with one half of the secondary winding on said second transformer, a second condenser charged by said second rectifier to a constant second voltage and a first resistor, circuit means connecting the entire secondary winding of said second transformer and said first and second condensers and resistor in series to 75 establish a first component of grid control voltage for said second valve constituted by the sum of thevoltage on said condensers and the voltage across the secondary of said second transformer, a .third rectifier connected across the series circuit formed by said resistor, said first and second condensers and the secondary of said second transformer to prevent said first component of grid control voltage from reaching positive valves, a second resistance and a third condenser connected across the secondary of said second transformer, a third resistance connected across said third condenser to thus form on said third resistance a second component of the grid control voltage for said second valve, and a fourth rectifier connected between said third condenser and third resistance to prevent. the said second component of grid control voltage from reaching positive values, and circuit means arranging said first and second components of grid control voltage in series between the grid and cathode elements of said second valve.

2. Induction type electron accelerator apparatus as defined inrclaim 1 wherein the grid element of said first valve is directly connected to one side of the secondary of said first transformer, the cathode element of said firstvalve is connected to the other side of the secondary of said first transformer through a condenser, and said condenser is bridged by a rectifier connected to a tap on said secondary.

References Cited in the file of this patent UNITED STATES PATENTS 2,480,169 Westendorp Aug. 30, 1949 2,535,710 Westendorp Dec. 26, 1950 2,538,718 Wideroe Jan. 16, 1951 2,654,838 Wideroe Oct. 6, 1953 2,754,419 Wideroe July 10, 1956

Images