US2984764A - Electron discharge devices of the magnetron type - Google Patents

Description

May 16, 1961 w. c. BROWN 2,984,764

ELECTRON DISCHARGE DEVICES OF THE MAGNETRON TYPE Filed Dec. 20, 1948 4 Sheets-Sheet 1 lNvENTOR WILLIAM C. BROWN TTORNEY W. C. BROWN May 16, 1961 ELECTRON DISCHARGE DEVICES OF THE MAGNETRON TYPE Filed Dec. 20, 1948 4 Sheets-Sheet 2 I TRANSMISSION FIG. 3 e

INVENTOR vl /l uAM C. BROWN TTORNEY y 6, 1961 w. c. BROWN 2,984,764

ELECTRON DISCHARGE DEVICES OF THE MAGNETRON TYPE Filed Dec. 20, 1948 4 Sheets-Sheet 3 30 A c 1 L U 55 |D E1 Ll L; LS C TRANggjlfilgSjigx LINE o o o SEC. SEC. SEC. 2 20 o-- o olNVENTOR ATRJRNEV W. C. BROWN May 16, 1961 ELECTRON DISCHARGE DEVICES OF THE MAGNETRON TYPE Filed Dec. 20, 1948 4 Sheets-Sheet 4 CHARACTEP/ST/C CURVES OF OPERATION wmmmwmmwmmwmwm 295% mi QN M85. him 5% E38? w RAT/0 0F FREQUENCY TO IT MODE mzoueucv INVEN WILLIAM C. B

TOR

R0 WN w ASTOR/v13) Patented May 16, 1961 ELECTRGN DISCHARGE DEVICES OF THE MAGNETRON TYPE C. Brown, Lincoln, Mass., assignor to Raytheon Company, a corporation of Delaware Filed Dec. 20, 1.948, Ser. No. 66,249

11 Claims. (Cl. 315-3969) William electron discharge devices known as magnetrons.

In magnetrons the maximum power which may be generated is dependent upon the amount of power which may be dissipated in the form of heat from the tips of the anode members hereafter referred to as vanes. Since there is a limit to the maximum dissipation density which may be used, approximately 1 megawatt per square centimeter (which will raise the surface of the vane to a melting point in approximately microseconds), the only satisfactory solution which can be obtained appears to be to increase the area of the vane tips Whereon the peak dissipation occurs.

Attempts to increase the size of the individual vanes resulted in the requirement for excessively large anode voltages, while increasing the number of vanes increases the change that the magnetron will operate at other than the desired frequencies.

These other frequencies are due to the magnetron oscillating in various modes, and the problem of separation of the frequency of the desired mode from that of the nearest undesired mode has been faced whenever it was desired to increase the power of the magnetron over its previous capabilities.

Accordingly, it is an object of this invention to produce a magnetron wherein the separation of the frequencies of the modes of operation is larger than was heretofore possible.

Another object of this invention is to produce a magnetron wherein the phase shift between successive anode members is less over the range of frequencies of operation of the tube than was heretofore possible.

Other and further objects of this invention will become apparent to persons skilled in the art as the description of the invention progresses reference being bad to the drawings wherein:

Fig. 1 is a vertical transverse sectional view of a magnetron built in accordance with this invention;

Fig. 2 is a partially cut-away view of the anode structure showing details of the strapping;

Figs. 3A, 3B, 3C, 3D and 3E show equivalent electrical networks for the anode structure and modifications thereof; and

Fig. 4 is a graph illustrating a comparison between operating characteristics of a magnetron built in conformance with this invention and one built from another design.

Referring now to Fig. 1, there is shown a magnetron having an anode structure 10 comprising a cylindrical ring 11 which is constructed of any good electrical conductor, for example, copper. Extending radially inward from the inner surface of the cylinder 11 are a series of equally spaced, flat, rectangular vanes 12 which, like the ring 11, may be made of copper. Each of these vanes 12 is joined to the cylinder 11 along one of the vanes edges, and the line of junction is parallel to the axis of the cylinder. Each vane 12 at this joining edge is about one-third the length of the cylinder 11 and is positioned approximately at the middle of the cylmders len h.

'l he vanes are all the same size and extend inward toward the axis of the cylinder roughly one-third the distance thereto leaving a cylindrical space along the axis of the cylinder. This cylindrical space inside the opening formed by the series of vanes 12 is occupied by a cathode 13 which may be of any desired design. As illustrated, cathode 13 comprises a cylinder 14 whose length is slightly greater than the height of the vanes such that it extends somewhat beyond either end of the cylindrical opening defined by the vanes 12. Welded to the outside surface of this cylinder 14 is a wire mesh 15 which is of any good refractory conducting material such as tungsten, and this wire mesh is coated with an electron emissive substance 16, such as a mixture of the oxides of barium and strontium. The diameter of the cylinder 14 is such that the coated outer surface of the wire mesh 15 is spaced a small distance from the edges of the vanes, thus providing the correct electron interaction space in the magnetron. The ends of the cathode cylinder 14 are confined by a pair of flat plates 17 which are circular in shape and securely fastened to the ends of the cylinder 14. These fiat plates are of a slightly large diameter than the cylinder 14 but are smaller in diameter than the cylinder opening defined by the opening 12.

The cathode structure 13 is supported by being rigidly attached at the center of the outer surface of the lower disk 17 to a hollow rod 18 which is coaxial with the cylinder 14 and the anode ring 11. This rod 18 is rigidly attached to a metallic contact cap 19 which is fused to a glass insulating sleeve 25]. The sleeve 20 is attached to a metallic cylinder 21A which is fused into an opening in a circular end plate 21 which closes the lower end of the cylindrical anode member 11. Thus it may be seen that the cathode 13 is rigidly supported with respect to the anode structure. The hollow tubular member 18 contains therein a conducting rod 22 which is supported by being connected to the lower end of the tubular member 18 through an insulating head 23 which may be, for example, glass. The upper end of the rod 22 extends slightly into the interior of the cathode structure 13 and is electrically connected to one end of a heater coil 24. The other end of the heating coil 24 connects to the lower plate 17 and thence back through the tubular member 18 to the contact cap 19. A heater voltage is applied between the contact cap 19 and the portion of the rod 22 extending beyond the insulating head 23.. The coil 24 is itself coated by ceramic insulation 24A.

An annular ring 25 of magnetic steel is imbedded on the inner surface of the end plate 21 concentric with the axis of the anode ring 11. This annular ring is of approximately the same diameter as the cylindrical space formed by the vanes 12. An upper end plate 26 which closes the upper end of the cylinder ring 11 has imbedded therein a second annular ring of magnetic material 27 which is positioned above the first ring 25.

When a suitable magnetomotive force is: applied across the end plates 21 and 26 as, for example, by a permanent magnet whose poles are positioned one on each of end plates 21 and 26 respectively, the magnetic annular rings 25 and 27 will direct a magnetic flux across the interaction space between the cathode structure 13 and the anode members 12.

The upper end plate 26 contains a tube 28 which passes therethrough, and which is used to exhaust the magnetron and is then closed by the mass of glass 29.

Referring now to Fig. 2 which shows the details of the strapping of the vanes 12, there is illustrated an embodiment of the novel features of my invention. The vanes are all connected alternately by straps 30. This strapping consists of two pairs of cylindrical straps, one pair nate vanes such that adjacent vanes are connected to different straps.

The action of the straps 30 on the vanes is to stabilize the operation of the magnetron and increase the separation of the frequencies of the modes of operation of the magnetron.

In order to further stabilize the magnetron operation and further increase the separation of the frequencies of the modes of operation, the cavities defined by the vanes 12 are divided into groups whose operation will be later described. This grouping is accomplished by a series of auxiliary straps 31 which define and connect to the end points of a group of cavities. This is accomplished in this particular modification by choosing four points on each of the straps 30' which are 90 apart. This produces four points on each strap which may be divided into two pairs of points. The points of each pair are connected by a second strap whose length is equal to one wave length of the frequency of operation which is desired, for example, the 1r mode frequency.

As shown, each pair of straps 30 has the chosen points of division adjacent each other and corresponding division into pairs such that the auxiliary straps 3 1 are parallel to each other forming a parallel transmission line one wave length long. In the particular illustration shown, pairs of points on the top and bottom pairs of straps are alternately grouped such that alternate pairs of points on the upper and lower pairs of straps 30 are connected by the one wave length transmission line.

Referring now to Figs. 3A, 3B, 3C and 3D, there will be described an analysis of the operation of the strapping by straps Sit and auxiliary straps 31. Fig. 3A illustrates a segment of the series of vanes 12 including a portion of the straps St for purposes of illustration, the straps 30 being at one end only of the vanes 12.

Point A represents the point of connection of a first vane to a first of a pair of straps; point B represents a point on the second strap adjacent the same vane as that containing point A. Points C and D are points on the first and second straps respectively, point D being the point of connection of said second strap to a second vane adjacent the first vane, and point C being a point on the first strap adjacentthe second vane.

Fig. 38 represents the electric network which is the equivalent of the cavity defined by the vanes containing points A and D and the sections of the straps existing between the vanes. C represents the capacitance existing between the length of strap A to C and the length of strap B to D; L represents the inductance of the length of strap AC or BD; L represents the inductance of the cavity and C represents the capacitance of the cavity. Since each pair of vanes acts as a unit the circuit parameters may be represented by lumped constants. The capacitance between A and B is equal to one-half the strap capacitance AC and BD with the other half of the strap capacitance appearing between C and D. The cavity components are shown connected between points A and D since those are the points where the vanes of the cavity are directly connected to the straps.

This circuit may be now analyzed by standard network theory using Kirchoffs laws, and producing the following result:

where is the phase by which a wave travelling from points AB to CD is caused to lag, and w=Z1rF, where F .is the frequency of the wave in cycles per second.

for the -r ange, of operation desired herein, 7 the. circuit may be simplified somewhat, as shown in Fig. 3C, with small error. In this circuit, G has been lumped with C in the tank circuit of the cavity to form C and L, of one strap has been reduced to zero and replaced by an equal L in the opposite strap to form a T network. The error due to the lumping of the capacitance is small, while the transformation of the inductance causes no error, in conformance with network theory. Analysis of this network results in a formula of In Fig. 4 there is illustrated a plot of phase shift versus a function of frequency for two types of magnetrons. Curve 6 represents a magnetron of the type just analyzed, wherein the phase shift across 20 cavities is plotted on the abscissa, While the ratio of the frequency at any particular point divided by the frequency of the 1r mode of operation of the magnetron is plotted along the ordinate. In this plot the ordinate value 1 represents the 11' mode fre quency.

It is necessary, in order to reach the next higher mode of operation, for a wave travelling along the straps 30 to lag by one complete cycle, or 21r radians by the time it has travelled completely around the straps so that it will be in phase with the starting wave, to reinforce the same and build up a stabilizing oscillation in the cavities. From curve 6 it may be seen that for a 20 vane magnetron to produce a phase shift of 360, or 271- radians, the frequency of operation must be increased by more than 30 percent. This is represented by point E on the curve. However, if vanes are used as, for example, in the magnetron of the present invention, each 20 vanes need only produce a phase lag of one-fourth of a cycle or before producing conditions of oscillation for the next higher mode. As shown by point P on the curve 6, this condition exists at a frequency which is approximately 2% above the frequency of the 1r mode. This frequency is so close to the 11' mode that it will interfere therewith, thus reducing the efficiency of operation of the tube, if indeed operation is at all possible.

In order to reduce the phase shift occurring across each series of 20 cavities, the ends of each series are connected together by a pair of auxiliary straps 31, and the length of the straps is constructed to be approximately equal to A or one Wave length at the frequency of the 1r mode of operation. This is shown schematically in Fig. 3D wherein sections 1-20 are connected in series and have their end terminals connected in parallel with a transmission line representative of the auxiliary straps 31.

In accordance with network theory the overall phase shift of the sections is reduced to a point which is somewhat above the phase shift of the transmission line 31 alone. This is given by the formula:

where Opp is the overall phase shift of the sections in combination with the transmission line; s is the phase shift of the transmission line; B is the overall phase shift of the 20 sections; Z, is the characteristic impedance of the 20 sections, and Z is the characteristic impedance of the transmission line.

The curve 7 in Fig.' 5 shows the plot of this phase shift (app with respect to frequency. The sections of the curve 7 occurring at phase shift and 360 phase shift shown by dotted lines G and H represent points where the phase shift became unreal since the cosine of the angle became greater than 1.

If 4 sections are used, a 90 phase shift is required to produce the next higher mode of oscillation above the 1r mode. Curve 7 shows that a 90 phase shift requires a frequency change of approximately 6% of the 'n' mode frequency as shown by point I. This is roughly three' times that required without the auxiliary strap 5.

.It may beseen from the curves.6 and 7 that as the number of vanes is increased beyond 80, the amount by which the auxiliary straps multiply the mode separation increases. For example, with 360 vanes grouped in 18 groups of 20 vanes, and each of which is strapped by a transmission line 3 1, the mode separation would be roughly 1.55% while without grouping, using simply the double straps 30, the separation would be .1% thus producing over fifteen times as much separation with the grouping as without.

It is to be clearly understood that the number of vanes in each group and the number of groups may be varied. Further, the groups can be strapped at both ends by the straps 30 and the auxiliary straps 31, or at either or both ends by the auxiliary straps 31 without straps 30, or at one end only by the straps 30 or a single strap 30 in conjunction with the auxiliary straps 31 at one or both ends.

It may be noted that the circuit analysis involved is very broad and may be applied to any network wherein a phase shift through said network varies with frequency. Moreover, other means than a simple one wave length transmission line may be used having equivalent compensatory eifects.

For example, as shown in Fig. 3E, a transmission line of a half wave length 7\/2 may be used and the lines of the transmission line crossed during the transmission to create the extra 180 phase shift.

This completes the description of the invention. However, applicant does not wish to be limited to the particular details described herein as other modifications will be apparent to those skilled in the art, which will fall within the spirit and scope of this invention. Accordingly applicant desires a broad interpretation of the appended claims commensurate with the scope of the invention within the art.

What is claimed is:

1. An electron discharge device, comprising an evacuated envelope containing a cathode and an anode structure, said anode structure comprising resonant cavities, adapted to be energized by discharges of said device, a transmission line connecting points on a first cavity to corresponding points on a second cavity, said second cavity being separated from said first cavity by a plurality of intervening cavities.

2. An electron discharge device, comprising an evacuated envelope containing a cathode and an anode structure, said anode structure comprising resonant cavities, adapted to be energized by discharges of said device, a transmission line connecting points on a first cavity to corresponding points on a second cavity, said second cavity being separated from said first cavity by a plurality of intervening cavities and the length of said transmission line being substantially equal to a multiple of a half wave length of the resonant frequency of said cavities.

3. An electron discharge device, comprising an evacuated envelope containing a cathode and an anode structure, said anode structure comprising resonant cavities, adapted to be energized by electron discharges in said device, and strapping connecting points on a first cavity to corresponding points on a second cavity, said second cavity being separated from said first cavity by a plurality of intervening cavities.

4. An electron discharge device comprising an evacuated envelope containing a cathode and an anode structure, said anode structure comprising resonant cavities, adapted to be energized by said discharge device, strapping connecting points on a first cavity to corresponding points on a second cavity, said second cavity being separated from said first cavity by a plurality of intervening cavities and the length of said strapping between said points being substantially equal to a multiple of a half wave length of the resonant frequency of said cavities.

5. An electron discharge device, comprising an evacuated envelope containing a cathode and an anode structure, said anode structure comprising a plurality of anode members defining resonant cavities, a set of straps connecting alternate anode members and further strapping connecting discrete points on each of said straps.

6. In an electron discharge device comprising an evacuated envelope containing a cathode and an anode structure, said anode structure comprising a plurality of anode members defining resonant cavities, a set of straps connecting alternate anode members, and a transmission line connecting discrete points on each of said straps and the length of said transmission line being substantially equal to a multiple of a half wave length of the resonant frequency of said cavities.

7. In an electron discharge device comprising an evacuated envelope containing a cathode and an anode structure, said anode structure comprising a plurality of anode members defining resonant cavities, and conductive means connecting points on a first of said cavities to corresponding points on a second of said cavities, said first cavity being separated from said second cavity by a plurality of intervening cavities.

8. In an electron discharge device comprising an evacuated envelope containing a cathode and an anode structure, said anode structure comprising a plurality of anode members defining resonant cavities, and a transmission line connecting points on a first of said cavities to corresponding points on a second of said cavities other than an adjacent cavity and the length of said transmission line being substantially equal to a multiple of a half wave length of the resonant frequency of said cavities.

9. In an electrical network, whose input to output phase shift varies with frequency, means for decreasing the rate of change of said phase shift with respect to frequency comprising a transmission line connecting said input and output whose length is a multiple of one half the wave length of the operating frequency of said network.

10. A first electrical network whose input to output phase shift varies with frequency, and means for decreasing the rate of change of said phase shift with respect to frequency comprising a second electrical network in parallel with said first electrical network, said second network having a phase shift which is a multple of a half wave length of the desired operating frequency of said network.

11. An electrical nework whose input. to output phase shift varies with frequency, and means for decreasing the rate of change of said phase shift with respect to frequency comprising a transmission line in parallel with said first electrical network, said transmission line having a phase shift which is a multiple of a half wave length of the desired operating frequency of said network.

References Cited in the file of this patent UNITED STATES PATENTS 2,196,272 Peterson Apr. 9, 1940 2,408,235 Spencer Sept. 24, 1946 2,414,085 Hartman Jan. 14, 1947 2,504,329 Heising Apr. 18, 1950

Images