WO1992020088A1

WO1992020088A1 - Microwave tube with a frequency-tunable cavity resonator

Info

Publication number: WO1992020088A1
Application number: PCT/FR1992/000381
Authority: WO
Inventors: Henri Desmur; Armel Beunas
Original assignee: Thomson Tubes Electroniques
Priority date: 1991-05-03
Filing date: 1992-04-29
Publication date: 1992-11-12
Also published as: DE69202668T2; EP0537327A1; EP0537327B1; FR2676142A1; DE69202668D1

Abstract

A microwave tube having at least one electronically frequency-tunable cavity (3) defined by a conductive wall (10) and containing a conductive element (8) provided with at least one switching element (39) controlled by radiation (32)) from a light source outside the cavity (3). When the switching element (39) is open, the cavity (3) resonates at a first frequency, and when it is closed, said cavity (3) resonates at a second frequency. Applications: frequency-tunable magnetrons in particular.

Description

MICROWAVE TUBE WITH RESONANT CAVITY FREQUENCY TUNABLE.

The present invention relates to microwave tubes with frequency tunable resonant cavity. These tubes are in particular conventional or coaxial magnetrons or klystrons. * A resonant cavity is limited by conductive walls. It behaves analogously to a resonant circuit composed of inductors and capacitors. It is distinguished from it by the fact that it can resonate on a multiplicity of different frequencies. The distribution of these frequencies is fixed by the Helmholtz equation resolved taking into account the boundary conditions, imposed by the physical and geometric characteristics of the cavity. The solutions define the resonance modes of the cavity and each mode corresponds to a frequency. For each mode, the resonant frequency is inversely proportional to the dimensions of the cavity. A conventional magnetron generally includes:

- a central electron source cathode,

- a crown-shaped anode surrounding the cathode,

- a coaxial cavity surrounding the anode. Two transverse flanges close the magnetron and seal it. Each elementary cavity forms a resonant circuit. The elementary cavities are linked by the coupling between cavities and interaction space. The frequency of oscillation of the magnetron is a function of the resonant frequency of each elementary cavity. This frequency can be tuned using a tuning device.

A coaxial magnetron generally includes:

- a central electron source cathode,

- a crown-shaped anode surrounding the cathode, - a coaxial cavity surrounding the anode.

The anode has a plurality of elementary cavities and the coaxial cavity is coupled to the elementary cavities by slots located in the bottom of one elementary cavity in two.

The magnetron's oscillation frequency corresponds roughly to the resonance frequency of its coaxial cavity. This frequency can be tuned using a tuning device.

The tuning devices, used in magnetrons, are generally mechanical. They consist in introducing an element of the plunger or plate type in the elementary cavities (for conventional magnetrons) or in the coaxial cavity (for coaxial magnetrons).

This element modifies either the self or the equivalent capacity of the cavity. The movement of this element is either manual or motorized. In all cases, the frequency excursion speed is quite low and the tuning device is heavy and bulky. When the tuning device is motorized, the motor consumes a power which can be significant and it must be cooled.

Other tuning devices are electronic. They use varactors working in blocked mode, placed in a cavity. These are electronically variable capacities. They modify the reactance of the equivalent circuit of the cavity.

The disadvantage of varactors is their fragility. They are placed in a microwave electric field superimposed on a continuous field. Since they work in blocked mode, the voltage applied to their terminals must be lower than their avalanche voltage. This limitation is troublesome because it cannot be applied to a high power.

It is also possible to introduce into the cavity, limited by at least one conductive wall, at least one conductive element provided with at least one switch element controlled from the outside of the tube. When the switch element is open, the cavity resonates on a first frequency; when the switch element is closed, the cavity resonates on a second sequence. According to the invention, the switch element is controlled by radiation from an illumination source external to the cavity. A sealed window, transparent to radiation is mounted on the conductive wall opposite the switch element. The radiation can be transported by an optical fiber. The switch element may be a photosensitive diode.

The light source can be amplitude modulated and quickly vary the state of a switch element so as to frequency modulate a signal produced by the tube.

Preferably, the conductive element has an elongated shape and extends substantially along a current line which would be established in the conductive wall in the absence of a conductive element.

It can be envisaged that the cavity is an auxiliary cavity coupled to a main cavity.

The tube according to the invention can be for example of the family of magnetrons or klystrons but this is not limiting.

We will now describe in detail the invention, with reference to embodiments illustrated by the figures among which:

- Figure 1 shows an exploded view of a coaxial magnetron frequency tunable electronically;

- Figure 2 shows a cross section of a variant of a coaxial magnetron tunable in frequency electronically;

- Figure 3 shows an exploded view of a coaxial magnetron according to the invention;

- Figure 4 shows an exploded view of a second variant of a coaxial magnetron tunable in frequency electronically;

- Figure 5 shows an exploded view of a cavity of a klystron according to the invention.

Figure 1 shows an exploded view of a coaxial magnet ron controllable on frequency. It is built around a axis XX '. For the sake of simplification, the cathode is not drawn and the anode 1 has been shown partially: we see only its outer wall 2 provided with slots 7 for coupling.

A coaxial cavity 3 surrounds the anode 1. It is limited by a conductive wall 10 formed by the anode 1, an external lateral wall 4 conductive and transversely of two flanges 5 conductive. The external side wall 4 is cylindrical and coaxial with the anode 1. The coaxial cavity 3 can be tuned in frequency. The coaxial cavity 3 of a coaxial magnetron preferably operates in a TE .. _nι mode. The electric field lines are circular inside the cavity. Near the external side wall 4 and the anode 1, the electric field is minimum and the magnetic field is maximum. Microwave current flows in the external side wall 4. The current lines induced in the external side wall 4, by the TE -..... mode, are circular. The coaxial cavity 3 contains at least one conductive element 8, preferably arranged, along the current lines induced in the external side wall 4, by the TE mode ..... present in the coaxial cavity 3, in the absence of the conductive element. In Figure 1, there are shown two conductive elements 8, these are annular conductive tracks and concentric with the outer side wall 4. Each track is interrupted at a point and the two ends of the track are connected to the terminals of a source. S of DC bias voltage.

The two conductive elements 8 are carried by a dielectric support 6 which is cylindrical and coaxial with the external side wall 4. Each conductive element 8 is provided with at least one switch element 9 controlled from the outside of the high frequency tube. When the switch element 9 is open, the conductive element 8 behaves like an open circuit, no microwave current flows through it. Microwave current flows through the external side wall 4. The cavity resonates on a first frequency. When the switch element 9 is closed, the conductive element 8, connected to the source S of bias voltage, behaves like a short circuit. There is modification of the microwave current flowing in the external side wall 4 and appearance of microwave current in the conductive element 8. This modification of current simulates a displacement of the lateral wall 4 towards the conductive element 8. The conductive element 8 then transports part of the energy present in the cavity, this energy remaining substantially constant. The cavity resonates on a second frequency.

In Figure 1, there is shown several switch elements 9 on each conductive element 8. The switch elements 9 are regularly distributed along the conductive elements 8 and are mounted in series. This is just an example, other arrangements are quite possible. The conductive elements 8 are without contact with the conductive wall 10. The switch elements 9 are here diodes electrically controlled by the bias voltage. They are either in a blocked state or in a conducting state. The bias voltage source S is used to control the state of the diodes from outside the magnetron. With a bias voltage V, of suitable direction, the diodes are blocked, the conductive elements 8 behave like open circuits. The cavity resonates on the first frequency.

Two conductors pass tightly through the external side wall 4 and connect the source S of bias voltage to a conductive element 8. By applying a bias voltage in the opposite direction to the voltage V, the diodes 9 become conductive and there is modification of the microwave current flowing in the external lateral wall 4.

This current modification simulates a displacement of the external side wall 4 towards the conductive element 8 and therefore towards the dielectric support 6. This displacement simulation causes a variation in the resonance frequency of the coaxial cavity 3. The second resonance frequency, for the same mode, corresponds to that which a conventional coaxial cavity of smaller diameter would have. The diodes can operate in an avalanche regime to have maximum conductivity while driving. We are no longer limited in power as we were with varactors.

In FIG. 1, the dielectric support 6 is arranged, for the sake of clarity, quite distant from the external side wall 4. In practice, it will preferably be placed near said wall, so that it is in a zone with electric field as weak as possible and magnetic field as strong as possible.

The number of diodes, in series on a conductive element 8, is determined by the power level present in the cavity 3; we arrange for them to be subjected to a high frequency potential difference compatible with their characteristics. They will be chosen with an impedance as low as possible so that the conductive elements 8 are traversed by microwave currents as strong as possible so as to maximize the change in frequency.

In FIG. 1, a source S of bias voltage was used to control all the diodes of the same conductive element 8. With several sources S of bias voltage and several conductive elements, it is thus possible to modulate the oscillation frequency of the magnetron, according to the state of the diodes 9 of each conductive element 8. Several frequencies are then available. The magnetron thus produced is electronically agile. Instead of having, as shown, several conductive elements 8 on the same dielectric support 6, it could be envisaged to place a single conductive element per support and to have several supports, concentrically for example. We could also place several conductive elements per support and have several supports concentrically, to further multiply the number of available frequencies. The limitation of the number of frequencies comes essentially from the difficulty in making watertight passages to connect the conductive element to the source S, because the interior of the cavity is subjected to vacuum.

One could envisage that the diodes 9 of all the conducting elements 8 of the same support 6, are controlled by a single source S of bias voltage. This limits the number of watertight passages to be fitted in the external side wall 4, but it also reduces the number of frequencies available. In this case, the conductive elements 8 are connected to each other and to the same source S of bias voltage. This variant is shown in Figure 2.

To further reduce the number of watertight passages through the external side wall, another variant consists in connecting one end of a conductive element to a ground point of the cavity and the other end to a first terminal of the source S of bias voltage, a second terminal of the bias voltage source S being brought to ground. Instead of simulating the displacement of the external lateral wall of the coaxial cavity 3, it is possible to simulate the displacement of a flange of the coaxial cavity 3.

FIG. 2 represents, in cross section, a variant of a coaxial magnetron tunable in frequency electronically.

The difference with the magnetron of FIG. 1 is located at the level of the coaxial cavity 20 and there is no change at the level of the anode 21 and the cathode 22.

The coaxial cavity 20 is closed by two facing flanges, substantially transverse to the axis of the tube. In Figure 2, the flanges are not visible. The lines of currents induced in the flanges, by the TE _Q1 mode - present in the cavity, are directed according to concentric circles. Now the coaxial cavity 20 contains at least one conductive element 23, arranged spnsibly along a current line induced in one of the flanges transverse to the axis of the tube. In FIG. 2, two conductive elements 23 are shown, these are concentric conductive tracks carried by a dielectric support 24 in the form of a disc. The dielectric support 24 is placed near a flange and is substantially parallel to said flange.

The disc can receive one or more conductive elements 23 on one of its main faces or even on both, it suffices not to establish electrical contact between the conductive elements 23 and the flange. Each conductive element 23 is provided with at least one switch element 25.

As in FIG. 1, these are electrically controlled diodes. The two conductive elements 23 are cut at a point and their ends are linked in pairs. A common source of DC bias voltage S is connected at the cutoff. Arrangements are made for the cutoff to have sufficient capacity so as not to constitute a high impedance at the operating frequency of the tube.

The state of all the diodes can be controlled from outside the tube thanks to the source S of bias voltage. When the diodes are blocked, the cavity resonates on a first frequency. When all the diodes are conductive, there is simulation of the displacement of the flange towards the dielectric support

24 and therefore reduction of the height of the coaxial cavity 20. The cavity resonates on a second frequency.

The dielectric support 24 is preferably made of a material having dielectric losses as low as possible and good behavior in temperature. Alumina can be used. The invention proposes to use a switch element controlled by radiation from an illumination source external to the cavity. It can be a photosensitive diode. This construction is represented in FIG. 3. The main difference between this FIG. 3 and FIG. 1 is located at the level of the conducting elements and of the elements. switches. The other constituent elements of the magnetron bear the same references as in FIG. 1. The external side wall is no longer mechanically connected to a source of energy for controlling the switching elements. This figure shows two conductive elements 38; these are annular tracks carried by the dielectric support 6. The conductive elements 38 face the external side wall 4 of the coaxial cavity 3. They are in contact with the side wall 4. Each conductive element 38 comprises at least one switch element 39. The switch element 39 is in this case a photosensitive diode controlled by radiation 32 coming from an external light source (not shown). A single switch element 39 has been shown per conductive element, but this is only a nonlimiting example, there may be several.

To control each diode 39, a sealed window 31 transparent to the radiation used is placed in the external side wall 4, facing the diode 39. It can be made of sapphire, silicon, boron nitride or germanium, for example. The mounting of watertight windows 31 on the external side wall 4 is easier to achieve than the watertight passages of the wires connecting the electrically controlled diodes to the bias voltage source. There is less risk of leaks. The radiation 32, of wavelength and intensity adapted to the nature of the diode 39, can come from a laser modulated in amplitude by a Pockells cell or by a Sch artz modulator, for example. These examples are not limitative. The laser is outside the magnetron. It illuminates the diode 39 through the window 31. The radiation 32 could be transmitted near the window 31 by an optical fiber. In the absence of radiation, the diode is blocked. The cavity resonates on a first frequency. The external lateral wall 4 is traversed by microwave current but not the conductive element 38. In the presence of the radiation 32, the diode 39 becomes conductive and there is modification of the microwave current flowing in the external side wall 4 and appearance of microwave current in the conductive element 38. The cavity resonates on a second frequency. The current modification simulates the displacement of the external side wall 4 towards the conductive element 38. The diodes can be made conductive by infrared radiation having a wavelength substantially of 1.6 micron for example.

With the amplitude-modulated illumination source, it is possible to frequency-modulate the signal produced at the output of the magnetron by varying the conductive or blocked state of high-frequency diodes 39, for example a frequency close to the oscillation frequency of the magnetron. . This modulation amounts to widening the frequency band inside which the magnetron can produce energy.

It would be entirely possible to arrange the conductive elements 38 and the diodes 39 on the other main face of the cylindrical dielectric support 6. It would suffice to produce the dielectric support in a material allowing the control radiation of the diodes to pass; alumina and infrared radiation would be compatible, but this example is not limitative. The diodes 39 and the conductive elements 38 would face the anode 1.

Magnetrons using photosensitive diodes as switching elements are less fragile than those using electrically controlled diodes. We have freed ourselves from the wires connecting the polarization source to the conductors.

Another mode of construction of a tube tunable in frequency electronically is represented on figure 4. It is about a coaxial magnetron. Its coaxial cavity 3 is coupled to at least one auxiliary cavity 44. In FIG. 4, there is only one but one could place several of them all around the coaxial cavity 3. The auxiliary cavity 44 represented is a section of metallic rectangular waveguide closed by two transverse walls 49. The auxiliary cavity is limited by a wall 40 formed on the sides of the waveguide and the transverse walls.

The coupling between the coaxial cavity 3 and the auxiliary cavity 44 takes place through an orifice 43 passing through the external lateral wall 4 of the coaxial cavity 3 and one of the transverse walls 49 of the auxiliary cavity 44.

The auxiliary cavity 44 contains at least one conductive element 48 provided with a switch element 45. The operation is the same as that which has been described in FIGS. 1 and 2.

In Figure 4, there is shown a set of conductive tracks 48 each provided with a diode 45; these tracks 48 are mounted in a bridge, their two ends are connected to the terminals of a source S of DC bias voltage intended to control the state of the diodes 45 via two separate tracks 46,47. The tracks 46, 47, 48 are carried by a dielectric support 42 disposed near a short side of the rectangular waveguide substantially parallel to this short side. They are without contact with the short sides of the guide. The tracks 48 provided with diodes 45 are transverse to the main axis of the rectangular waveguide. They extend substantially along the current lines induced in the short side of the guide by the mode present in the auxiliary cavity 44, in the absence of a conductive element. The tracks 46, 47 are isolated, on the continuous plane, from the long sides of the waveguide. On the other hand, sufficient capacity is provided at the two ends of the tracks 48 and the long sides of the waveguide and therefore between the tracks 46, 47 and the long sides of the waveguide. Consequently, the microwave current which will be established in the conductive elements 48, when the diodes are conductive, will be able to close through the conductive tracks and the sides of the guide.

The mode present in the auxiliary cavity 44 is excited by the mode present in the coaxial cavity 3. The mode present in the auxiliary cavity 44 can be, for example TE mode "- ...

When the diodes 45 are blocked, the auxiliary cavity 44 resonates on a first frequency. When the diodes 45 are conductive, the auxiliary cavity operates on a second frequency. There is simulation of the displacement of the short side of the guide towards the dielectric support 42. The auxiliary cavity 44 operates as if its long sides were reduced and therefore as if its electrical length was increased. The resonant frequency of the auxiliary cavity 44 is modified and this causes a variation in the reactance at the opening

43. There is then a modification of the resonance frequency of the coaxial cavity 3.

All the mounting variants described in FIGS. 1, 2, 3 can also be applied to the magnetron of FIG. 4. Instead of controlling all the diodes together using a single voltage source S, one could use several bias voltage sources each controlling one or more diodes. One could also consider placing several dielectric supports 48 in the auxiliary cavity 44, each supporting one or more conductive elements, provided with one or more switch elements controlled all together or independently. The electric switching diodes could be replaced by optical switching diodes. In this case, the tracks would be in electrical contact with the long sides of the waveguide, due to the absence of DC bias voltage.

Instead of being directed substantially along the current lines of a short side of the guide, the conductive elements could be directed substantially along the current lines of the transverse walls 49 of the auxiliary cavity.

The microwave tube according to the invention can also be a klystron. FIG. 5 thus represents a conductive cavity of a klystron according to the invention. It is a cylindrical cavity 50, of axis XX ', crossed by an electron beam (not shown) directed along the axis XX'. The beam is contained respectively in a sliding tube 51 at its entry into the cavity and in another sliding tube 51 at its exit. The two tubes 51 face each other in the cavity and are separated by a space. The cavity 50 is limited by a wall 59 formed of a cylindrical external side wall 52 and two transverse flanges 53.

The cavity contains at least one conductive element 54 provided with at least one switch element 55 controlled from the outside of the klystron. In Figure 5, there is shown a plurality of conductive elements 54 each provided with a single switch element 55. These conductive elements are conductive tracks carried by a dielectric support 56. The dielectric support 56 is coaxial with the external side wall 52 of the cavity 50. The conductive elements 54 are arranged substantially along the current lines induced in the external side wall 52, by the mode present in the cavity 50, in the absence of a conductive element. The current lines follow the generatrices of the cylindrical wall 52. Indeed, the cavities of the klystrons generally operate on the TNL mode ... -. The electric field is directed along the axis XX ', it is maximum near the axis XX' and more particularly between the ends of the sliding tubes 51. The current lines corresponding to this mode follow the sliding tubes 51, the flanges 53 and the external side wall 52 by forming a sort of open torus at the ends of the sliding tubes 51. When the switching elements 55 are open, there there is a microwave current flow in the outer side wall 52 and not in the conductive elements 54. The cavity resonates on a first frequency.

When the switch elements 55 are closed, there is a modification of the microwave current flowing in the external side wall 52 and the flow of microwave current in the conductive elements 54. This results in a simulation displacement of the external side wall 52 towards the dielectric support 56 and consequently a modification of the resonance frequency of the cavity 50 of the klystron. The cavity resonates on a second frequency. The switching elements 55 shown are photosensitive diodes controlled by radiation 57 of appropriate wavelength and intensity. This radiation 57 comes from an illumination source which can be a laser. Only two radiations have been shown, but there is one for each switching element.

The device for controlling the photosensitive diodes 55 is identical to that described in FIG. 3. The external side wall 52 has sealed windows 58. Each radiation 57 passes through a window and illuminates one of the diodes 55. In FIG. 5, the elements conductors 54 and the diodes 55 are arranged towards the inside of the cavity 50. Each radiation 57, passes through a window 58 then the dielectric support 56 and illuminates a diode 55. The material of the dielectric support 56 is chosen so as to be transparent to the radiation. Each end of the conductive element 54 is in electrical contact with a flange 53 so that the microwave currents which circulate in the conductive elements when the diodes 55 conduct can close on themselves. The conductive tracks 54 can be produced by any known means, in particular by depositing a metallic paint on the dielectric wall.

The conductive elements are not necessarily tracks, they can be produced in the form of conductive wires, conductive strips or any other equivalent means.

The diodes, whether controlled by application of a polarization or of radiation, can work in an avalanche regime. The temperature has a lesser effect than in blocked mode. The frequency stability when the temperature varies is better than in tubes using varactors.

In the examples described, we have simulated the displacement of part of the wall of the cavity (side wall flange oxi) but we can of course combine the examples described to move both the side wall and the flanges.

Claims

1 - Microwave tube comprising at least one resonant cavity (3) tunable in frequency limited by a conductive wall (10), at least one conductive element (38) provided with at least one switch element (39), disposed in the cavity and intended to modify the resonant frequency of the cavity, characterized in that the switching element (39) is controlled by radiation (32) coming from an illumination source external to the cavity (3).

2 - Microwave tube according to claim 1, characterized in that a sealed window (31) transparent to radiation (32) is mounted on the conductive wall (10), opposite the switch element (39).

3 - Microwave tube according to claim 2, characterized in that the window is made of sapphire, silicon, boron nitride or germanium.

4 - Microwave tube according to one of claims 1 to 3, characterized in that the radiation (32) is transported by an optical fiber.

5 - Microwave tube according to one of claims 1 to 4, characterized in that the light source is a laser.

6 - Microwave tube according to one of claims 1 to 5, characterized in that the light source is amplitude modulated and quickly changes the state of a switching element so as to frequency modulate a signal produced by the tube.

7 - Microwave tube according to one of claims 5 or 6, characterized in that the light source is modulated by a Schwartz modulator or a Pockels cell.

8 - Microwave tube according to one of claims 1 or 7, characterized in that the conductive element is a photosensitive diode. 9 - Microwave tube according to one of claims 1 to 8, characterized in that the conductive element (38) and the switch element (39) are carried by the same dielectric support (6) placed near the conductive wall (10). 10 - Microwave tube according to claim 9, characterized in that, when the support (56) is arranged between the switch element (55) and the conductive wall (52), the support (56) is made of a material transparent to radiation. 11 - Microwave tube according to one of claims 1 to 10, characterized in that the cavity is an auxiliary cavity coupled by at least one orifice to a main cavity of the tube, the auxiliary cavity being external to the main cavity.

12 - Microwave tube according to one of claims 1 to 11, characterized in that it belongs to the family of klystrons or magnetrons.