US2736839A - Microwave oscillator - Google Patents

Description

1956 H. v. NEHER MICROWAVE OSCILLATOR Filed Nov. 26, 1945 FIG. 2

INVENTOR HENRY V. NEHER ATTORNEY nited States Patent MICROWAVE OSCILLATOR Henry V. Neher, Pasadena, Calif., assignor, by mesne assignments, to the United States of America as represented by the Secretary of the Navy Application November 26, 1945, Serial No. 630,947

Claims. (Cl. 315-42) This invention relates to transit time microwave oscillators.

Desirable characteristics for such oscillators may include the following: (1) low plate voltage, (2) low power for heating cathode, (3) simplicity of construction, (4) stability of operation and insensitivity to temperature changes, (5) wide range tunability, (6) constant power output over wide band of frequencies, and (7) good efiiciency. Wide range and constant output appear to be incompatible with high efficiency for low power tubes. However, the present invention'provides a tube in which the characteristics of low plate voltage, low cathode heater power, simplicity of construction and high eificiency can be obtained to a degree substantially equivalent to that achieved with conventional tubes operating in the long Wave length region. in particular, for example, a tube in accordance with this invention may operate in the 10 cm. region with a cathode supply of 1.5 volts at 0.25 amp.; a plate supply of 100 volts, plate current of 10 ma., and deliver 0.1 to 0.2 watt of power. Such a tube would have many applications.

It is therefore the object of this invention to provide a transit time microwave oscillator of simple, sturdy construction which will operate with low plate voltage, small cathode supply voltage and with high efficiency.

I Another object is to provide a transit time oscillator tube which is frequency tunable by either temporary or permanent deformation of the outer envelope.

Other features, characteristics and advantages of the invention will appear from the following detailed description when taken together with the drawings the figures of which are illustrative of the principles of this invention which should not be deemed to be limited except by the specification and the spirit of the appended claims.

Fig. l is a middle cross section showing the anode positions, Fig. 2 is an internal view also showing the cavity and anode positions therein, Fig. 3 is a diagram showing the relative positions of the various electrodes and Fig. 4 is a typical embodiment showing the location and relation of the essential elements of the tube.

Referring now to Fig. 1 there is shown a metal cylinder 1 having two anode vanes 3 and 5 mounted on the inner circumference at opposite ends of a diameter. If the spacing between the adjacent ends of the vanes 3 and 5 is, say 0.1 of the diameter of the cylinder 1 and if their thickness is equal to their separation, then this section of the cylinder 1 can be made to resonate in the fundamental mode, as shown, at a wave length about three times the diameter of the cylinder 1.

Fig. 2 also shows a cylinder 7 with the anode vanes 9 and 11 and the fundamental modes H and E lines.

Radiation is prevented from leaking out the ends of the cylinder 7 because at. this wave length the circular guide is beyond cutoff and has an attenuation of about 30 db for each diameter distance along the cylinder toward an end. V

Fig. 3 shows a cathode 13 in the form of a hairpin mounted midway between two anode vanes 15 and 17,

2,736,839 Patented Feb. 28, 1956 ice and between two pairs of grids, 19 and 21, and 23 and 25, mounted respectively between cathode 13 and anode vane 15 and between cathode 13 and anode vane 17. Let the grids 21 and 23 nearest the cathode 13 be kept at a small direct current potential with respect to the cathode. At one phase of the radio frequency cycle the electric field might be as shown in Fig. 1. It will be seen that a radio frequency voltage will be built up across the cathode-grid or input circuits which will be in phase with the voltage in the main anode circuits. The voltage gradient will be divided among the various gaps according to the relative reactances of the gaps, hence most of the voltage will be across the gap between anode and outer grid since the capacity across this gap will be normally only about one-fifth of the capacity across cathodegrids. Thus radio frequency voltage built up between the cathode and the first two grids supplies the power for driving the input circuit. In order to achieve a uniform field between cathode and grid 21 filamentary type may be used or a fiat solid cathode may be used though the latter is not suitable for battery operation. This will keep the phases of electrons passing through the grids the same.

The phase of the radio frequency voltage is such as first to eject current through grid 21, and then on the other half cycle through grid 23. These bunches of current must arrive in the regions between the second grids 19 and 25.respectively and their respective adjacent anodes 15 and 17 when the anodes are farthest negative with respect to their average potential, in order to deliver the most power into the anode circuit. It is evident then that there must be a phase difference of 11' radians (or 21r+1r or 4rr-I-1r etc.) between the radio frequency voltage across the input circuit and the arrival of the current in the output circuit. This can be obtained by adjusting the spacing between the respective pairs of first and second grids, that is, between 21 and 19 and between 23 and 25, and by properly choosing the potential of the outer grids 19 and 25.

For such a tube operating on 100 volt maximum, in the 10 cm. region, it is not easy to keep the total transit angle down to 1r radians. In fact there will be a phase shift of about 1r radians between cathode 13 and first grids 21 or 23. There must therefore be provided a transit angle of about 31r radians between the first grids 21 and 23 and their respective second grids 19 and 25 to bring the phase right, since about %radians will be used up in the anode circuit. This total transit angle is small enough to insure that no serious debunching will occur.

The spacing between anodes 15 and 17 and their respective adjacent grids 153 and 25 is not extremely critical. Though there is an optimum spacing the minimum of the curve is broad, and the resulting transit angle may be anywhere between 21 radians 4 and 11' radians without changing the efficiency apprecia v- The capacity between the various elements and its relation to the resistance due to the beam must be considered. The beam offers the highest loading, and therefore the minimum of resistance, in the region between the cathode and the first grids. Considered as lumped capacities the voltage across the gap will not be aifected appreciably if the resistance of the gap is greater than the capacitive reactance, or otherwise stated the Q must be larger than unity if these elements were in a resonant circuit. For reasonable values of spacing and size of grids this condition is fulfilled for both the input circuit and the anode circuit.

Another consideration which must be met in the construction of this oscillator is that since the first grids 21 and 23 must necessarily have radio frequency potential between them, therefore unless precautions are taken, not only will the radio frequency power be conducted out the leads to the base of the tube, but also this loading will cause a difference in radio frequency potential at different points of the grid, whereas with no center connection the potential would be constant or uniform due to the radio frequency field. To overcome this problem either of two solutions will suffice. As shown in Fig. 4 a loop 40 can be provided connecting the two grids 37 and 38 which will intercept just sufiicient magnetic field to set up the same potential where the loop is fastened to the grids as the grids would normally assume due to the capacity effects described above. This will level off the potentials again so that they will be uniform at all points along the grids. The direct current connection for supplying positive bias to these grids from lead 24 should be taken off as a center tap half way around the loop 40. The same provision of a correcting loop 41 must be made also in the case of the outer pair of grids 36 and 39. The optimum size of loop must be determined finally by test but the first solution may be had by assuming a reasonable distribution of the magnetic field and choosing the area of the loop to give the necessary voltage on the end of the grid determined from capacity considerations.

An alternative and perhaps simpler method of accomplishing the result is to make the connecting leads in length before shorting them.

If symmetry is preserved there will be little leakage at the leads to the base of the tube since the TEu mode here being dealth with has high attenuation along a concentric line. Only the asymmetrical component will be so propagated. Chokes could be included in the base of the tube if desired.

The provision of screen grids 36 and 39 serves two purposes. First, it provides a region between the grid and anode Where the transit time of the bunch of electrons after passing the first grids can be adjusted to arrive in the plate region in the proper phase, andsecondly it permits the electrons to enter the plate region with a relatively high energy, thus even though they stop at the plate they still have a comparatively small transit angle for a reasonable spacing between the outer grid and the plate.

The anode face should allow the electron beam to enter through the face without appreciable interception and be trapped so that secondary electrons are discouraged from emerging back into the anode-outer grid gap and thus absorbing energy. This can be accomplished either by placing a grid on the end of the anode or by mounting strip metal to give minimum interception to the beam. In either case the spacing between adjacent elements should be about the same as the spacing from anode to the outer grid. This will insure a good radio frequency field which is necessary to extract the energy from the beam.

Fig. 4 shows the complete tube arrangement, including the evacuated envelope within which the anode cylinder 32 with its two vanes 33 and 34 is inserted after assembly of the cathode 35 and the grids 36, 37, 38 and 39 with the anode structure, 33 and 34. The grids are indicated as equally spaced merely for drawing convenience, and the respective ,pairs are shown joined by the correcting loops 40 and 41. For low voltage tubes the cathode 35 and the grids 36, 37, 33 and 39, are held with mica disks 42 and 43 at each end, and these disks 42 and 43 in turn are held rigidly in the cylinder 32 forming the anode circuit. The anode cylinder 32 could be mounted on the stem of the tube and after complete assembly inserted into the metal shell 30 forming the vacuum envelope. If desired to insulate the anode 32 electrically from the shell 3%, any suitable insulating material such as a thin mica sheet 31 for example can be placed between them.

The direct current base leads, 20 and 22 supply the cathode filament; grid lead 24 is center tapped to the correcting loop 40; while grid lead 26 is center tapped to correcting loop 41. Lead 28 supplies anode potential. The output lead for a low power tube of this type may consist of a concentric line such as shown, with inner conductor 46 and outer conductor 48 inserted through the top of the envelope 30 terminating in a loop 49 of proper size intercepting the magnetic field set up by the radio frequency currents. For high power tubes it would be desirable to have a window, not shown, on the top of the tube to open directly into a wave guide, not shown. The mode so generated if the tube fastened on to the end of the guide would be the usual TEu in a circular guide or the TEoi in a rectangular guide.

An additional feature of this invention is the fact that such a tube can be tuned, though over a limited range, say 10%, by squeezing the external envelope at the proper points, preferably say at points 59 and 52 at the region of the two anode vanes 33 and 34 respectively. Motion is thus transmitted through both cylinders 30 and 32 resulting in a closer anode to outer grid spacing. Otherwise if the squeeze is applied at points located at from the anode locations the anode to outer grid spacing would be increased. For many applications the metal cylinders could be deformed permanently so as to be set for a given wave length and this can be done after the tube has been sealed off.

What is claimed is:

l. A microwave transit time oscillator tube comprising, an open ended conductive cylinder having two inwardly projecting anode vanes of rectangular cross section diametrically opposed, with a separation between the adjacent ends thereof substantially equal to the thickness of said vanes, a cathode disposed midway between said anode vanes, a pair of grids disposed respectively between said cathode and each of said anode vanes, a second pair of grids each disposed respectively between one of said first named grids and the adjacent anode vane, and an evacuated envelope containing the aforesaid components.

2. A microwave transit time oscillator tube comprising, an open ended conductive cylinder having two inwardly projecting anode vanes of rectangular cross section diametrically opposed, with a separation between the adjacent ends thereof substantially equal to the thickness of said vanes, a cathode disposed midway between said anode vanes, a pair of grids disposed respectively between said cathode and each of said anode vanes, a second pair of grids each disposed respectively be tween one of said first named grids and the adjacent anode vane, and an evacuated envelope containing the aforesaid components, said envelope being adapted to be squeezed by means external to said envelope so as to alter the diameter of said cylinder in the region of said anode vanes whereby the spacing between said anode vanes and the grids adjacent thereto may also be changed for elfecting a tuning of the frequency of said oscillator.

3. A microwave transit time oscillator tube comprising, an open ended conductive cylinder having two diametrically opposed interior projecting anode vanes substantially centrally located, a cathode disposed substantially midway between the ends of said anode vanes, a pair of grids equispaced from said cathode on either side thereof, said grids being connected to each other by a coupling loop adapted for maintaining the end of the respective grid to which said loop is respectively connected at substantially the same potential as the other portions of said grid, a second pair of grids each being disposed respectively between one of said first main grids and its adjacent anode vane, said second named pair of grids being connected by a coupling loop substantially like that joining the first named pair of grids, and an evacuated envelope containing said components above described.

4. A microwave transit time oscillator tube comprising, an open ended conductive cylinder having two diametrically opposed interior projecting anode vanes substantially centrally located, adapted to oscillate in the fundamental mode, a cathode disposed substantially midway between the ends of said anode vanes, a pair of grids equispaced from said cathode on either side thereof, said grids being interconnected by a conductor of substantially one quarter wave length isolating the grids from each other for alternating currents induced therein, a second pair of grids each being disposed respectively between one of said first named grids and its adjacent anode vane, said second named grids being connected by a conductor substantially like that between said first named grids, and an evacuated envelope containing said above-described components.

5. A microwave transit time oscillator comprising, an open ended conductive cylinder provided with two interior projecting anode vanes diametrically disposed and adapted to oscillate in the fundamental mode at a wave length of approximately three times the diameter of said cylinder, a cathode at a predetermined potential disposed midway between the adjacent ends of said vanes, a pair of grids disposed equidistantly from said cathode on either side thereof at a predetermined distance therefrom whereby when said grids are provided with a positive bias of a predetermined magnitude relative to the potential of said cathode, electron bunches arrive at said grids and have a transit delay angle of approximately 1r radians, a second pair of grids respectively disposed between said'first pair of grids and their adjacent anode vanes at a predetermined spacing from said first named pair of grids, whereby when said second named grids are provided with a positive bias of a second predetermined magnitude and said first named grids are provided with a positive bias of said first predetermined magnitude, electron bunches arrive at said second named grids from said first named grids and have a transit delay angle of approximately 2 radians, said anode vanes being at a predetermined distance relative to said second named grids, whereby when said anode varies are connected to a potential source of a predetermined magnitude and said first named grids and said second named grids are provided with said first and second biases, respectively, the transit delay angle of bunches of electrons arriving from said second named grids to said anode vanes approximates '2- radians, and an evacuated envelope containing said components.

References Cited in the file of this patent UNITED STATES PATENTS

Images