US2591932A - Magnetron - Google Patents

Description

April 8, 1952 c. w. HANSELL MAGNETRON 2 SHEETS--SHEET l Filed Sept. 5, 1946 INVENTOR ,clz'l'and BY 7kg/Wb ATTORNEY C. W. HANSELL April s, 1952 MAGNETRON Filed sept. 5, 194e 2 SHEETS-SHEET 2 INVENTOR BY ATTORNEIY Patented Apr. 8, 1952 MAGNETRON Clarence W. Hansell, Port Jefferson', N. Y., as-

sgnor to Radio Corporation of America, a corporation of Delaware Applicationseptembcr 5, 1941i, Serial No. 634,994

13 claims. l

rlfhis invention relates to magnetron oscillation generators, and particularly to a method of, and

apparatus for, modulating the frequency of magnetrons.

An object of the present invention is to provide means for modulating the frequency of a magnetron oscillator.

A further object is to provide, a novel construction for a, magnetron oscillator which enables the magnetron to be modulated as to frequency by the application of modulation potentials to an electrode thereof.

From a broad aspect, the invention involves so constructing a continuously operatingmagnetron oscillator that it has two portions which tend to oscillate at different natural frequencies, and to so. couple these portions together as to force them to operate at a common frequency. Frequency modulation is effected by differentially modulating the contributions of these two portions of the magnetron to the common oscillation.

` An important feature of therpresent invention lies in the use of a tapered magnetic eld which is gradually tapered in strength in a direction parallel tov the axis and tov the cathode which lies along the axis. The arrangement is such that the intensity of the magnetic field is greater at one end of the magnetron than at the other, as a re-f vsult of which the natural frequency of operation of one portion of the magnetron is different from the other portion. 1

The preferred embodiment of the present invention comprises a magnetron havingV a multicavityV resonant anode structure withv a pair of cold cathodes arranged along the axis end-to-endA along the same straight line and spaced from each other at their adjacent ends.y These cold cath-v odes are water cooled and arranged to emit elli-.- ciently secondary electrons upon impact by primary electrons. A magnetic field is provided whose intensity is greater at one end of the magnetron than at the other, as a result of which, the, natural frequency of operation of one portion of the magnetron containing one of the .cathodes is different from the natural frequency of operation of the other portion of the magnetron containing the other cathode. Because of the com.. mon anode structure, both portions of the magnetron are closely coupled together and are forced to cooperate to produce a single oscillation frequency, and this frequency is intermediate the frequencies which the independent portions-of the magnetron tend to produce by themselves.v By applying differences of D. C. potentials (as by modulation) between thetwc cathodes, I am able to diifrentially vary the density of the rotating.

space charge around the cathode in such manner as to cause'one, portion or the other of the mag-e netron to predominate to a greater or less degree to thereby produce modulation in frequency.

According to another embodiment of the pres ent invention, the magnetron comprises two identical multi-cavity resonant anodes arrangedend-r to-end and insulated from each other, both resonant anodes enclosing a common evacuated space.v Extending through the centers of both resonant anodes, preferably along the axis of the' circuit. A tapered magnetic field isemployed to provide an intensity of magnetic field in one anode structure which is greater than the intensity of the magnetic field in the other anode structure. These anode structures are differentially modulated in potential in order to produce frequency modulation.

f' A more detailed description of the inventionV follows in conjunction with a drawing wherein:

Fig. 1 illustrates theY preferred embodiment of the invention, and l Fig; 2 illustrates another embodiment of thev invention.

Referring to Fig. 1 in more detail, there is shown a r magnetron ycomprising multi-pole or multi-cavity anode structure l0 surrounding@ pair of secondary emissive cold hollow cathodes C Il and l2 arrangedy in the'center alongtheaxis..

of the magnetron. Cathod'es Il and. I2 are ar ranged in the same straight line and spaced from each other as shown. These cathodes are pref-V erably oxide coated with secondary emissive material on their exterior surfaces within the anode structure. It should be noted that bothv cold cathodes are insulated from one another and from the anodeby means, ofglass seals i3, and. i4..

WaterY is used so that during operation; til@ cathode surface will be the coolest surfaceexposed to. the vacuum. This is done by making both cathodes hollow and introducing into each cathode a smaller hollow tube open at the end through which water or other suitable cooling fluid is introduced. The direction of Water flow is shown by the arrows. This cooling water is passed through the cathodes over two separate paths .in parallel in -order to prevent a difference in temperature .between the cathodes. The purpose of this cooling fluid is to causevolatile or and southJ semi-volatile activating materials present in the magnetron, such as caesium, to accumulate and remain on the surface of the cathode.

The common anode structure is also provided with cooling water passages I through which water is passed to cool the anode.

A tapered magnetic field is provided by a pair of pole faces labeled N and S representing northf It should be noted that one of these pole faces, for example, N, as shown in the drawing, is considerably larger than the other pole face S, as a result of which the lines of force follow the general paths shown by the lines I6. These pole faces may form part of a permanent magnet or, if desired, may be connected together by a yoke, not shown, in turn surrounded by a suitable solenoid to which a constant direction current potential is supplied.

Space charge retaining shields I'I and I8 are provided at the ends of the cathodes to prevent diffusion or undesired excursion of the rotating space charge from the anode structure.

An inspection of the lines of force I6 shows that the magnetic field is tapered in the space between the cathode and the anodes and that the intensity of this magnetic field is greater at one end portion, for example, that portion containing the cathode I I, than at the other end portion containing cathode I2.

Output from the magnetron is provided by a coupling loop I9 located within the interior of the multi-cavity anode and which is coupled via a coaxial transmission line TL to a suitable antenna. Output line TL is provided in circuit therewith with an adjustable stub section S. The output coupling coil I9 and the stub S have such dimensions as to be resonant with the portion of the line TL therebetween to the center frequency of oscillation. The stub S is also a protective device for grounding the inner conductor of the line TL for high D. C. potentials in the event of undesired internal arcing within the magnetron. It should be noted that the outer conductor of line TL is grounded and insulated from the magnetron anode structure through a glass seal 20.

A variable Voltage rectifier 2| serves to provide high positive potentials relative to ground to the anode structure through lead 22. A coupling coil 23 is provided for tuning the input circuit of the magnetron to' a radio frequency which is preferably quite high but still below the magnetron output frequency. By way of example, the input circuit of the magnetron may be tuned to 10 megacvcles, while the output frequency of the magnetron may be 100 megacycles per second.

To aid in starting emission from the cold cathode magnetron, the coupling coil 23 is excited to produce a high voltage oscillation in the magnetron anode-to-cathode potential at say 10 megacycles. This coupling coil excitation is produced by means of a quenched gap 24 whichis equivalent to an old fashioned spark radio transmitter.

Arelay 25 is provided which operates when the anode-to-cathode magnetron D. C. current starts, as a result of the operation of which the inputs to the transformer T and the quenched spark gap 24 and coil 23 are effectively removed from the circuit.

Diiferential modulation of the two halves of the magnetron is achieved by means of a pair of modulator tubes 26 and 21 whose inputs are connected to a suitable source of modulation and whose outputs or anodes are connected through radio frequency choke coils 28 to the different cold cathodes via connections 29 and 30. It should be noted that there are two resistors R provided across the anodes of the modulator tubes 26 and 21 and that the rst grids of both of these tubes are connected to adjustable taps on the resistors R. Condensers 3l, 32 and 33 are radio frequency bypass Condensers.

The operation of the system of Fig. 1 may be better understood from the following explanation. Because of the tapered magnetic field, the angular velocity of rotation of the electron space charge is greater in the region of the strong magnetic field than it is in the region of the weaker magnetic field. Since the oscillation frequency is dependent to a large degree upon this angular' velocity of rotation of the space charge. it will be evident that the two halves of the magnetron each tend to provide oscillation at a different frequency. By differentially modulating the density of the rotating space charge in the two halves of the magnetron, I am able to obtain a modulation in the frequency of oscillation. This is accomplished by applying modulating currents between the two cold cathodes I I and I2, in order to force the rotating space charge back and forth in a direction substantially parallel to the magnetic field. The magnetic field does not oppose motion in this direction so that relatively little modulation input energy is required. Although it may be expected that the required modulation potentials applied between the cathodes will not be very large, it should be understood that due to the self-repulsion of the space charge which tends to make it spread out in the axial direction, the modulation potentials may be substantial. Because of the fact that a space charge rotating in a tapered magnetic field in a direcf tion more or less perpendicular to the magnetic lines of force tends to move off in a direction toward regions of decreasing magnetic field strength, I may in practice, apply an average biasing potential between the two cathodes II and I2 in order to overcome this tendency.

Controllable bias between the cathodes II and I2 may also be utilized as a means to provide automatic control of the average frequency of oscillation. Such bias may be provided by an unbalancing of the two resistors R in the anode circuits of the two modulator tubes 26 and 21 or by unbalancing the anode currents to the modulator tubes by different adjustments of the taps on resistors R connected to the first grids of these two modulator tubes.

In order to give the magnetron a nearly fiat band pass selective characteristic so that the circuits will present little opposition to the modulation of the frequency over the pass band, I contemplate coupling the load to the magnetron' through a coupled resonant circuit. This coupled resonant circuit comprises the stub S', the coupling coil I9 and that portion of the output line TL` between the stub S' and the coupling coil I9.

The high voltage rectifier 2| converts alternating current power from the three phase power supply to D. C. potential, and delivers this power to the cold cathode magnetron. The magnetron in turn converts the D. C. power to very high frequency power and delivers it to the output line TL.

The degree of maximum frequency modulation obtainable with the split cathode magnetron shown in Fig. 1, increases with increasing output loading Yand the'load'limitris d'ependent'upon the secondary remission ratio of the-cathode :and the details fof the design insofar as they affect :the electron bombardment of the cathode. The percentage frequency deviations required are not vvery vgreat even for frequency modulated television purposes, in the range of frequencies above 0-megacycles. For this reason, conversion efficiencies will still be high in ycomparison with what can properly be obtained in television broadcast transmitters operated in lower frequencies.

It is a characteristic of most magnetrons that the best operating voltage is on the order of rhalf or two thirds the theoretical cut-off voltage for the strength of magnetic eld used. Hitherto, it has-been found. that to get a hot cathode mag# netron to start, it has been necessary to raise the anode-to-cathode voltage up close to the cut-off value to getcscillations to start after which the voltage vhad to be reduced to half, or somewhat more thanh'alf the starting value in order to olo-` tain reasonably good power conversion eflicienoy.

In the magnetron of Fig. 1, the normal operating anode potential might be of the order of 12,000 volts, in which case the theoretical cut-ofi potential might be on the order oi 20,000 to 25,000 volts. For this assumed case, the spark excited coupling coil 23 may be made to deliver a peak high frequency potential on the order of 12,1300

to 20,000 Volts so that the sum of the D. C. and the peak high frequency potential delivered to the magnetron when they are adding, will go substantially above the theoretical cut-o potential. Under these conditions, once each cycle of the -high frequency potential developed inl coil 23 and lappearing between the common anode 'andthe two cathodes of the magnetron, the niagnetron anode-to-cathode potential will rpass downward through the cut-off potential at a very rapid rate. Experiments have shown that 'very 4large values of emission from the cold cathodes may thus be built up and the time taken to build up very large emission currents may be only a small fraction of a microsecond, being deter`m mined principally by electron transit time from the cathode out toward the anode and back to 'the cathode while the potential is falling rapidly.

The reason that there is a build-up of emission is that while the potential is falling rapidly, yany electron which leaves the cathode will circle out from it and return with an excess of energy because of the falling electric field. This eircess energy will resultin secondary emission produced. 'bythe returning electron and, so long as the sec onda-ry emission ratio is greater than unity (which is the case here), there will be a rapid multiplication of emission by repetition of the process. It has been found that while emission from cold cathodes may not grow to limiting values for every cycle or pulse of potential falling through the cut-off value, this emission does grow to limiting values often enough so that except for some initial hesitation, a cold cathode magnetron of suitable design can be constructed to buildup full emission and to start oscillation.

Apparently, there are a few initial electrons leaving even a cold cathode perhaps due to inn pact of gas ions, cosmic rays, photo-emission, condensation and reevaporation of volatile material or perhaps even thermionic emission. These. spontaneously emitted electrons, regardless of the cause, are very few but apparently enough to cause the growth of emission under the conditions described above within a few seconds or less. It is, of course, not vnecessary 'that .the normal .operating value :of emission be reached provided .that Vduring the growth process, enough rotating space .ch-arge is accumulated to cause the magnetron itself .tostart oscillation. As soon asithis happens, the emission will be maintained by secondary emission from the vcathodes pro-v duced by electron bombardment accompanying oscillation.

vAs soon as high emission 'and oscillation of the magnetron of the Vpresent invention'has started, themagnetron A. C. input impedance becomes so low as Yto constitute a substantial short circuit of coil vE13 so that'the 10 megacyclepotential variations between the common anode and the two cathodes become quite small, as a result of which, these variations .cannotcause quenching of the oscillations developed in the magnetron once the oscillations vhave started. The relay 25 is ,an additional precaution to remove thequenched gap 2t from 'thecircuit .It will lthus be seen that when the magnetron starts, .it also short circuits the starting circuit andthe starter lthereafter can have little eiiect. The power to start the system is then automatically removed by action 'of the relay 25, which has its control coil in series with the D. C. input circuit of the magnetron.

Although the modulation circuit of the invention is shown as comprising two modulator tubes 26 and 27, it should be understood that this arrangement is illustrative of anysuitable modulation circuit, and that if desired, a single modulator tube can besuitably arranged in the circuit to obtain the differential modulation of the tw portions of the magnetron.

The multi-cavity anode structure is now well known in theart ,and `may be composed v.of aplurality of inwardly projecting anode portionssuitably spaced from one another. Reference is herein made to my U.. S. Patent #2,217,745, granted October 1,5, 1940, and to U. S. Patent #2,053,342 granted December 8, `19.36, to Samuel for a magnetron showinga suitable typeofmulticavity anode structure.

Fig. 2 illustrates another embodiment of the in vention wherein a single' water cooled secondary emissive cathode il extends through the centers of a pair of identical multi-cavity resonant anodes d2 and i3 arranged end to end and insulated from one another as shown. The tapered magnetic field is produced by the large pole face N and the small pole face S arranged at opposite ends or" the magnetron in order to produce vlines of iiux vsubstantially .parallel to the cathode-and transverse to the 'anode-to-cathode path. but

whose intensity of field is greater within the anode structure d3 than within the anode structure d2. Glass ring 44 serves to insulate the two resonant anode structure 42 and 43 from one another. Individual output loops-45 and 46 extend within the cavity Vanode structures and are joined externally `of thevmagnetron. These loops are extensions of inner conductors of coaxial lines 4l and 08 whose outer vconductors are insulated from .the anode structures. A suitable adjustable stub S is provided in the common transmission line TL extending to the antenna. The anodes 'are differentially modulated by potential from 'source which, if desired,'may be similar in arrangement to the modulated tubes 25 and 21 of Fig. 1. The anodes are maintained vat a lpositive poten'- tial relative to the .cathode by means of the D. C. power supply 2 l.. Thezcathodes are provided with end shields 49 for retaining the space charge within the magnetron.

What is claimed is:

. 1. An electron discharge device comprising an evacuated envelope containing therein cathode and anode electrodes spaced from each other in a direction transverse to the axis of said device, one of said electrodes being divided into two parts arranged end-'to-end along the axis of said device but spaced from each other, and magnetic pole pieces of different size at the respective ends of said device positioned along the axis of said device for producing a tapered magnetic ield in a direction substantially parallel to said axis but whose intensity in the region of one of said two parts within said envelope is greater than in the region of the other of said two parts within said envelope.

2. An electron discharge device comprising an evacuated envelope containing therein cathode and anode electrodes spaced from each other in a direction transverse to the axis of said device, one of said electrodes being divided into two parts arranged end-to-end along the axis of said device but spaced from each other, and magnetic pole pieces of different size at the respective ends of said device positioned along the axis of said device for producing a tapered magnetic eld in a direction substantially parallel to said axis but whose intensity in the region of one of said two parts within said envelope is greater than in the region of the other of said two parts within said envelope, and means coupled to said two parts for applying modulation thereto in opposite sense.

3. An electron discharge device comprising an evacuated envelope containing therein spaced cathode and anode electrodes, the space between said anode and cathode electrodes being free of any electron emission control element, one of said electrodes being divided into two substantially identical parts arranged end-to-end and spaced from each other, means adjacent said envelope for producing a constant but tapered magnetic field in a direction substantially parallel to said cathode but whose intensity in the region of one of said two parts within said envelope is greater than in the region of the other of said two parts within said envelope.

4. An electron discharge device comprising an evacuated envelope containing therein spaced cathode and anode electrodes, the space between said anode and cathode electrodes being free of any electron emission control element, one of said electrodesbeing divided into two substantially identical parts arranged end-to-end and spaced from each other, said device including means for producing within said envelope a constant but tapered magnetic ield in a direction substantially parallel to said cathode but whose intensity in the region of one of said two parts within said envelope is greater than in the region of the other of said two parts within said envelope, and circuit means located externally of said envelope and coupled to both of said parts for producing a potential difference therebetween in accordance with modulation. s

5.v An electron discharge device comprising a hollow cold cathode and a surrounding anode structure, means for supplying cooling huid to the interior` of said cathode, said cathode having its outer surface activated to be secondary emissive at a ratio greater than unity upon bombardment by primary electrons, said .cathode being divided into two spaced parts arranged end-toend in the same straight line, said cooling means communicating with the interior of both parts of said hollow cathode, said anode structure being common to both parts of said cathode, and means producing a tapered magnetic iield acting substantially transversely to the anode to cathode path and whose intensity is greater in the region surrounding one part of the cathode than in the region surrounding the other part of the cathode.

6. An electron discharge device comprising a hollow cold cathode and a surrounding anode structure, means for supplying cooling fluid to the interior of said cathode, said cathode having its outer surface activated tobe secondary emissive at a ratio greater than unity upon bombardment by primary electrons, said cathode being divided into two spaced hollow parts arranged end-to-end in the same straight line, said cooling means communicating with the interior of both parts of said hollow cathode, said anode structure being common to both parts of said cathode, and means producing a tapered magnetic field acting substantially transversely to the anode to cathode path and whose intensity is greater in the region surrounding one part of the cathode than in the region surrounding the other part of the cathode, means connected to said anode for maintaining said anode at a positive D. C. potential relative to said cathode, and means coupled to both parts of said cathode for producing a potential difference therebetween in accordance with a signal.

'7. An electron discharge device comprising an evacuated envelope containing therein a cold cathode and a surrounding multiple-cavity anode structure spaced from each other in a direction transverse to the axis of said device, said cathode having its cuter surface coated with secondary emissive material, said anode structure being divided into two spaced parts arranged end-to-end along the axis of said device, said cathode being common to both parts of said anode structure, magnetic pole pieces of different size at the respective ends of said device for producing a unidirectional tapered magnetic eld acting substantially transversely to the anode to cathode path and whose intensity is greater in the region of one part of said anode structure than in the other part. Y Y

8. An electron discharge device comprising an evacuated envelope containing therein a cold cathode and a surrounding multiple-cavity anode structure spaced from each other in a direction transverse to the axis of said device, said cathode having its outer surface coated with secondary emissive material, said anode structure being divided into two spaced parts arranged end-toend along the axis of said device, said cathode being common to both parts of said anode structure, magnetic pole pieces of different size at the respective ends of said device for producing a unidirectional tapered magnetic iield acting substantially transversely to the anode to cathode path and whose intensity is greater in the region of one part of said anode structure than in the other part, space charge retaining shields at the ends of said cathode, means for diierential- 1y varying the relative potentials of said two anode parts in accordance with modulation, and means for separately abstracting output energy from both anode parts and for combining this abstracted energy.

9. An electron discharge device comprising a hollow cold cathode and a surrounding anode structure, said cathode having its outer surface activated to be secondary emissive at a ratio greater than unity upon bombardment by pri- 9 mary electrons, said cathode being divided into two spaced hollow parts arranged end-to-end in the same straight line, space charge retaining shields for the ends of said cathode parts, said anode structure being common to both parts of said cathode, and means producing a tapered magnetic eld acting substantially transversely to the anode to cathode path and whose intensity is greater in the region surrounding one part of the cathode than in the region surrounding the other part of the cathode, and pipes extending into the interior of both parts of said cathode from opposite ends thereof for supplying cooling fluid to said cathode parts.

10. A magnetron device having an anode and a cathode located within said anode and extending along the axis of said device, and including means for producing within said envelope a tapered magnetic eld acting substantially transversely to the anode to cathode path and whose intensity at one end along the axis is greater than the intensity at the other end, one of said electrodes being divided into two substantially identical spaced parts arranged end-to-end, and means for cooling both said parts during operation of said magnetron, said last means including conduits for introducing a cooling uid into each of said spaced parts.

11. In an electron discharge device having concentrically arranged anode and cathode electrodes, one of which is composed of two similar parts arranged end-to-end and spaced from each other, the method of operation which includes applying a unidirectional tapering field acting substantially transversely to the anode to cathode path and whose eld intensity is greater in the region of one of said parts than in the region of the other part, and applying a potential difference between said two parts which varies in accordance with variations of modulation.

12. A device in accordance with claim 8, characterized in this, that said cathode is hollow, and including means communicating with the interior of said cathode for supplying cooling iluid thereto.

13. An electron discharge device comprising a hollow cathode and a surrounding anode resonator structure, said cathode being divided into two parts arranged end-to-end but spaced from each other, at least one of said parts of said cathode having its outer surface activated to be secondary emissive at a ratio greater than unity upon bombardment by primary electrons, means communicating with the interior of said onepart for supplying cooling liquid thereto and means communicating with said anode resonator structure for supplying cooling liquid thereto.

CLARENCE W. I-IANSELL.

REFERENCES CTED The following reierences are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 1,558,120 Simpson Oct. 20, 1925 2,114,114 Roberts Apr. 12, 1938 2,298,673 Brown et al Oct. 13, 1942 2,329,780 Zalesak Sept. 21, 1943 2,412,372 Usselman Dec. 10, 1946 2,414,085 Hartman Jan. 14, 1947 2,421,636 McArthur et al June 3, 1947 2,429,755 Hallmark Oct. 28, 1947 2,442,786 Somers June 8, 1948 2,444,435 Fisk July 6, 1948

Images