US2503394A - Frequency control system - Google Patents

Description

Patented Apr. 11, 1950 UNITED STATES PATENT OFFICE ADJUSTABLE GRAVITY BALL PROJECTOR FOR GAME BOARDS" Alois'Lamprecht, Cheektowaga N. I I Application June 5, 1946, Serial No. 674.430

. 3 Claims. 1

This invention relates to games in which a ball is rolled in an aimed direction from one end of a game board toward the opposite end thereof, and more particularly to apparatus for projecting the ball on its course.

The principal object of the present invention is to provide such a ball projector which can be moved transversely of the game board and which can also be rotated about a substantially vertical axis so that the ball can be projected in an desired direction toward the target at the opposite end of the board.

Another object is to provide such a ball projector which is manually operable in a simple manner.

Another object is to provide such a. ball projector which retains the ball on the projector until the same is aimed as desired and then manually actuated to project the ball on its aimed course.

Another object is to provide such a ball projector which can be easily removed from the game board so that the complete unit may be shipped in a collapsed and compact condition.

Other objects are to provide such a ball projector which, is simple in construction, capable of standing up under abusive use, and inexpensive to manufacture.

In the accompanying drawings:

Fig. 1 is a top plan view of a portable bowling alley game incorporating my new ball projector.

Fig. 2 is a side elevational view thereof.

Fig. 3 is a bottom elevational view thereof.

Fig.4 is a fragmentary, vertical, longitudinal, partial sectionthrcugh my ball projector and associated parts. -The ball projector forming the subject of the present invention is shown in the drawings in conjunction with an indoor, portable bowling alley game having a fiat surface along which the ball rolls, although it will be understood that my ball projector can be employed in or adapted for any game-in which a ball is rolled in an aimed direction from one end of a game board toward the opposite end thereof.

The bowling alley game illustrated in the drawings comprises an elongated flat board I ll arranged at a slight incline-along its major axis and supported at its lower end by an end board ll andside boards 12-, the opposite higher I end being supported by a transverse leg l 3 The leg I3 is providedwith an opening through which an inclinedball return trough I4 is arranged and the opposite end of this trough is suitably mounted on the end board II. The top of the L 2 game board") on either side thereof is provided with a gutter I5 adjacent the end board H and side boards l2 which leads to a hole It, this hole being located immediately above the trough H. A plurality of ten pins 1-! are Shown as standing in the usual arrangement on the game board l0.

The ball projectingapparatus is arranged on the top of the game board Ill at the opposit end thereof and comprises a pair of transversely spaced support members 20 20 secured in any suitable manner-to the board and a transverse rod 2l-mounted on these members. A block 22 is slidably and rotatably mounted on the rod 2| and carries an upstanding pin or spindle 23. A ramp member 24 in the form ofa generally triangularly shaped block is relatively movably mounted on the pin 23. For this purpose, the ramp member 24 is provided with a bore 25 near its rear side which receives a portion of the pin 23 and is. adapted to slide and turn thereon, the bore 25 .at its lower end being enlarged, as indicated at 26, to receive a coil spring 21. The coil spring 2! surrounds the lower portion of the pin 23 and the lower. end of the spring bears against the top of the block 22 and the upper end of the spring bears againstthe shoulder in the ramp member 24 formed by the .bore 25 and counterbore 26. The top side of the ramp member which inclines downwardly from its high rear end to its front end is provided with a ball guide groove 28 which permits the ball to leave the ramp at about the level of the top surface of the game board l0 and moving in a direction approximately tangential thereto. However, the profile of the top side of the ramp member may be of an suitable contour. The rear end of the ball guide groove 28 adjacent the upper end of the bore 25 is slightly spherically recessed, as indicated at 29, so as to retain the ball 30, as shown in Fig. 2. To actuate the projectorthe ramp member 24 is manually pressed downwardly and to facilitate this a cleat or rib Si is secured to each side of the ramp member 24; H

The bowlingalley game is played b placing the board assembly on a table or other suitable supporting surface and arranging the ten pins I! in the usualmanner on the board. The ball 30 is placed on its seat 29 and the other parts are in the condition shown in Fig. 2. The operator then by hand shifts the ramp member 24 from side to side transversely of the board Hi, this being permitted by the block 22 sliding on the rod 2|, and also swings the ramp member about the axis of the pin 23 until the projector has been aimed as desired. The, operator. then preferably placeshis April 11, 1950 R. R. LAW

FREQUENCY CONTROL SYSTEM 6 Sheets-Sheet 1 Filed Aug. 26, 1944 a mi MOF JJ 5W0 1060 dm E jnzd dw u 7532 \NVENTOR RUSSELL R. LAW

dokdl tw fiddma mam/340 95 1. dwE u 2 6250.50 m

ATTORNEY April 11, 1950 R. R. LAW 2,503,394

- FREQUENCY CONTROL SYSTEM Filed Aug. 26, 1944 6 Sheets-Sheet 4 OUTPUT Tb AMP No.1.

FEED BACK TO G210 a Flqs. as, a.

Fig. 7.

l NVENTOR RUSSELL R. LAW

ATTOR N EY April 11, 1950 R, R LAW FREQUENCY CONTROL SYSTEM 6 Sheets-Sheet 5 Filed Aug. 26, 1944 N UK a 0.2mm Z ma 0% 50E r m2.

I! 'IIIIIIII H 1 I 4 l/l/AIIII'IIII.

lo I

INVENTOR RUSSELL. R LAW 7K4 ATTORN EY A ril 11, 1950 R, LAW 2,503,394

FREQUENCY CONTROL SYSTEM Filed Aug. 26, 1944 6 SheetS Sheet 6 I l////// q I [III a mg I l I l /l/ L 0 IIIIIIIIIIIIIIIIII Lu I \NVENTOR RUSSELL R. LAW

ATTORNEY Patented Apr. 11, 1950 UNITED STATES PATENT OFFICE FREQUENCY CONTROL SYSTEM RussellR. Law, Princeton, N;J'.,.assignor'to'Radio Corporation of America; acorporation of Delaware Application August 26, 1944; SeriaINo. 551,296

6 Claims.

modulated carrier currents.

the'radiantenergy art. It is particularly useful for. relaying timing modulatedv oscillations from one point to another point. It: is: of especial interestwhereitisinecessaiy to handle wide modulation. bands without distortion. resulting.

Briefly, my invention: concerns a method and. meansv of amplifying and/or relaying timing modulated currents in which means the outputcurrent is generatedby a controllable frequency oscillator which. is loclred in synchronism. as to.

phase and frequency with timing modulated os cillations to be amplified, and thereby the oscillater is. caused. to generate: strong oscillations the.

timing of. which is modulated: in a mannercorresponding to the modulations on the incoming current.

The general object of myinventionisimproved relaying and/or amplificationxof timing modulated carrier currents.

An additional object of my invention is improved. relaying and/or amplification of high quency changers such as heterodyne converters or frequency multipliers or-frequenc dividers.

Inits broader aspect my inventionvincludesan oscillation generator of controllable frequency and a device wherein generated oscillations and the modulated oscillations to beamplified or. relayed'are subjected to a phase comparison proc-- ess. Th phase comparing device is arranged to provide an output current or. currents the in-- tensity of which represents. the phase relation of the generated oscillations and-modulated'oscillations, and thisoutput isusedto entrain-the oscillation generator.

Phase comparison devices-of the prior art are usually complicated in nature and use of the same does not permit attainment of a generalobject of my invention, i. e., a simple and efiicient timing modulated oscillation amplifier as free of tuned" circuits as possible.

An additional object of my invention then is.

improved current or voltage phase. comparison.

Briefly, thisobject is'attaine'd' by provision of an improved beam tube wherein a beam is generated and propagated to two electron. multiplier chains inequaliamounts when the timing modulated oscillations which are applied to the de fleeting electrodes, and the generated oscillations which are applied to' a beamv intensity control g-rid are inwphase. The electron multiplier chains then have equal outputs. and these outputs are applied to the oscillation generator which. then operates: in synchronism. If. for some reason the oscillation generator runs ahead of the timing modulated oscillations to be amplified, thereis; a

phase displacement between the beam: intensity control'voltage: and the deflecting voltage, so that the beam is no longer propagated tothe two electron rnultiplier chains in equal amounts. The" outputs of these chains-are no longer balanced and* the unbalanced outputs change the frequenoy of the oscillation generator until syn'- chronism is reestablished.

The oscillation generator may take various forms. I prefer an oscillation generator of the entrained type such as disclosed, for example, in Hansell U. S. Patent #l,'787,979, dated January 6'; 1931; In this type of oscillationgenerator two'electron discharge systems are arranged to operate at different frequencies and are also err trained so. that they operateat a common frequency intermediate the diiferent frequencies. These oscillators are then, in accordance with my invention, controlled by the outputs of the two electron multiplier chains.

Oscillators of this type known in the prior art do not lend themselves well to application in-my improved system, and a further object of my invention is the provision of a novel'and improved oscillation generator using. the principle described briefly above. In accordance with my invention, this improved generator comprises. a pair of'. electron devices associated with and in resonant. chambers and arranged to oscillate at a common frequency which depends upon the emission electrode excitation. In my improved.

systemthe emission electrodes are differentially modulated'by, theoutputs of the electron multi-- plier chains.

Th manner in which the objects enumerated above and other objects which will appear here inafter are attained, will. now be described in greater detaiL.

In describing. my invention, reference willv be made to the attached drawings, wherein:

Fig. 1 shows by block diagram the essential features of an amplifier or communication system arranged in accordance with my invention.

Fig. 2 illustrates a preferred type of phase discriminator and dual electron multiplier used in the rectangle ll) of Fig. 1.

Figs. 3, 4, 5 and 6 show graphically the relation between current in the beam tube and in the two amplifier multipliers of Fig. 2;

Fig. 7 illustrates preferred embodiments of the controlled oscillator, one or the other of which may be included in the rectangle 16 of Fig. 1, while Figs. 8 and 8a show schematically and diagrammatically an electron beam phase discriminator and electron amplifier tube arranged in accordance with my invention.

Various attempts have been made by radio engineers to develop wide band ultra high frequency amplifier systems such as, for example, ultra high frequency relay systems for television and communication networks. In this work a number of problems are confronted which have not been worked out satisfactorily at the present time. For example, it is desirable to use ultra high frequencies because of their directional properties and because of their wide-band-carrying possibilities, and because of the room available in the ultra high frequency spectrum. However, amplification of ultra high frequencies is difficult at the present time due to tube structure, electron transit time in known tubes, capacity between electrodes, etc. Due to this inability to amplify wide band ultra high frequencies, in general it has been thought necessary to employ elaborate combinations of frequency converters, intermediate frequency amplifiers, and modulated output stages in amplifier systems such as the ultra high frequency amplifiers described above. These systems are unnecessarily complex, as will be seen by comparing the same with my improved system of the present application.

In Fig. 1, the received timing modulated signal is fed to a phase discriminator in [0, wherein it is compared as to phase with the oscillations generated in a controllable frequency oscillator IS. The phase difference is in a sense detected and predominant output is fed to one or the other of two amplifier or relay stages 12 or M, depending on whether the phase of the oscillations is ahead of or behind the phase of the incoming carrier. The two stages then cooperate with the oscillator to control the frequency of operation thereof. The amplifiers are designated as being included as separate units in rectangles l2 and M, although as will be seen hereinafter they may be included with the phase discriminator (rectangle in an improved electron discharge tube arrangement.

In this arrangement a portion of the signal generated in the controllable frequency oscillator I6, i. e., the output stage, is fed back to the phase discriminator wherein the phase of the outgoing signal is compared with that of the incoming signal and energizes predominantly one or the other of two amplifier chains, which in turn cooperate with the oscillator in It to raise or lower the frequency of the controllable frequency oscillator depending upon whether the phase of the outgoing signal lags or leads the phase of the incoming signal. In this manner I constrain the outgoing signal to keep very nearly in phase with the incoming signal, and the amplified outgoin current reproduces the timing modulations on the incoming signal.

A specific example of my novel phase discriminator and amplifier has been shown in Fig. 2. In the arrangement of Fig. 2, 20 indicates schematically a novel beam tube wherein the incoming oscillations are compared as to phase and control amplifiers excited in accordance with the results of said comparison. Only the essential elements of the tube are shown in this figure, and electrode source are omitted in the sake of simplicity. The tube 20 has an electron beam source K and beam deflecting electrodes 22 and 24 to which the incoming signal is applied. The tube also has two electron multiplier relays or amplifiers, designated #1 and #2, in the path of the beam. The amplifier outputs control the generator (rectangle 16, Fig. 1) which supplies the amplified timing modulated output. The tube 20 also has a beam intensity control electrode G to which generated oscillations are fed.

For the purpose of the present analysis, it will be assumed that the current or signal fed back to the control grid G serves only to switch the beam on and off, and that the incoming signal applied to the deflection plates 22 and 24 serves only to switch th beam from one collector-multiplier-arnplifier, say #1, to the other collectormultiplier-amplifier, say #2. Excitation on the control grid G then switches the beam on and off at a rate determined by the phase and frequency of the oscillations gene-rated in the generating means included in IE, while the incoming signal deflects the beam between amplifiers #1 and #2 at a rate and phase depending upon the frequency and phase of the incoming frequency modulated signal.

The operation of the system of Fig. 2 is illustrated in Figs. 3, 4, 5 and 6. With an incoming signal applied to the deflection plates 22 and 24, but with no feedback, i. e., in the absence of excitation from the oscillation generator in IS on the grid G, the beam current will be constant in amplitude, i. e., on continuously, and of a constant intensity. The unmodulated beam current is represented in line I of Fig. 3. The incoming signal acting through the plates 22 and 24 of the beam tube will switch the beam from one multiplier amplifier to the other in the time-phase relationship represented by lines 2 and 3 of Fig. 3. Note that now each electron amplifier is supplied with substantially equal portions of the beam current, so that except for the displacement in time between the pulses of beam current the amplifiers may be considered to have equal outputs.

Now assume that a signal, 1. e., carrier current, is fed back from the carrier current source in [6 to grid G to modulate the beam current as represented in Fig. 4, and further that the feedback energy bears such a frequency and phase relation to the incoming signal that the individual beam current pulses are equally divided between the two amplifier multipliers, as indicated in Fig. 4:. In this case again the average current to amplifier-multiplier #1 is equal to the average current to amplifier-multiplier #2. This condition may be called the median or operating condition at which the oscillation generator in i6 is in the carrier or no modulation condition.

If now the frequency of the incoming signal applied to unit I5 is increasing (under modulation), or if the signal generator at 6 should tend to fall behind, the beam current to amplifiermultiplier #1 will increase and the beam current to amplifier-multiplier #2 will decrease as shown in Fig. 5. Conversely, if the frequency of the incoming signal applied to ID is decreasing, or if accuses the signal generator in IS should tend to run ahead of its mean frequency, the current to amplifier-multiplier #1 will decrease and the beam current to amplifier-multiplier #2 will increase as shown. in Fig. 6'.

Inasmuch as transit time in the amplifiermultiplier stages and in the beam tube and capacitance effects tend to level. off the individual R.-F. pulses, after several stages of amplification the output current of. the amplifier-multipliers #1 and #2 will. be substantially stead-y and will be substantially proportional to the average. input current. The output currents of the amplifiermultipliers is used to entrain. or look in the oscillation generator l6 which is to send out or provide the amplified timing modulated wave.-

In accordance with my invention the steady output currents of the amplifiers #l and #2, the averagev values of which depend. upon the rela-- tion. of the phase of the currents generated in the. oscillation generator at It, and the incoming signals supplied. to the. deflection elements, which currents vary differentially in average intensity, are used to entrain or look in the said oscillation generator supplying the feedback current to the control grid This generator may be of any appropriate type and is preferably an improved and simplified generator utilizing the principle disclosed in Hansells U. S. Patent #1,?81979, filed March 23,. 1928,. and issued January 6, 1 931.

This oscillation generator may be said to comprise two generators, one of which is tuned to a frequency above the desired mean frequency, and the other of which is tuned to a frequency an equal amount below the desired mean frequency. In the present application this mean frequency is the mean frequency of the incoming signal. The said two oscillation generators are entrained to operate at the mean frequency. ,In the presence of modulation and assuming that the generator is synchronized, the frequency of the entrained: generator follows the frequency of the incoming. wave. In the event that the oscillation generator in It tends to run ahead of the incoming wave, the average intensities at the amplifier outputs change differentially to correct the frequency of the: oscillation generator in 16.

If desired, I may use a reactance. tube controlled oscillation generator of the type disclosed by Crosby, say, for example, in. his U. S. Patent- #2,250,095, dated July 22, 1941, or similar arrangements.

My system is particularly adapted to amplify or repeat ultra high frequency wave energy modulated by a wide signal spectrum such as, for example, used in television or facsimile. In Fig. 7 I have shown an improved oscillation generator which I prefer to use in my system. This improved. generator utilizes two novel tube structures arranged in and cooperating with enclosed spaces forming with the tubes and electrodes cavity resonators operating at different frequencies. The generators comprise a closure member 50 providing spaces wherein the cathodes 52 and 52, the control grid electrodes 56 and 56-, and an anode electrode 63 and 63' are enclosed. The cathodes may be carried by or integral with hollow tube leads 53 and 53 which pass through walls of the closure member. Cathode heating current may be supplied by leads 55 and 55 in the hollow tubes 53 and 53. The closure member 59 is shown as being of metal but may be of other material having an inner surface coated with conducting material. The

control grid electrodes 56 and 56' may be continuations or extensions of the walls of the closure member 50 The member 50- may have a section (in a plane perpendicular to the: paper) of. any appropriateshape such as square; rectam gular, orcircular, oval, etc.

In these cavity resonators, as is. well. known, the frequency of operation depends. primarily on the mode of operation, the dimensions and shape of the chamber, the L and C thereof, and.

to some extent. orr the energy losses in the system per cycle of generated. energy. The area in which the cathode 52 is located is dimensioned. and the mode of operation is such that the cath-.

ode 52 is in a resonant circuit operating at a frequency fl above the mean frequency atv which.

the two generators are entrained. Openings in the cavity introduce energy losses and the cham bars are as nearly completely closed as possible... Adjustable tuningmeans, however, are desirable.-

Plugs introduced in the cavities change their dimensions and are a practical means of adjJust= ing the frequency of operation. plugs may be adjustably introduced. in the space for tuning purposes. In my improved system the plunger members 51, 57" serve the multiple purposes of adjustable tuning means, chamber wall, and means for introducing the cathode heating and cathode excitation. leads into the cavities and also complete the high frequency tuned cathode circuits for the oscillators. Tuning of the cathode cavities is accomplished. by sliding plungers 5'1. and 5E. The members 5'1 and 51 have contact fingers which bear on the walls of the closure member 50 as the members i field. In the embodiment illustrated, movement of the members to reduce the cavity dimensions increases the frequency of operation of the oscillators.

In the manner described above, the cathode 52 is arranged to operate in a timed circuit resonant at a frequency f2, differing in the other direction from the mean frequency by an amount equal to the difference. between the mean frequency and the frequency fl.

The anodes 63 and 63' are. mounted adjacent the respective grids 56 and. 5E. Resilient movable contact fingers cooperate with the anodes for tuning the anode circuits of the. oscillators, tuning being accomplished by adjusting the length ofthe anode circuits. The anodes of the two discharge devices are coupled in pushpull, and may be considered in a circuit tuned to a band of frequencies covering a range which at least includes f! and 12, but is preferably wider than the range fl to f2. Loops 68 and 68 in the anode portion of the cavity feed generated energy back to loops 69 and 69' in. the resonant cathode spaces to provide regeneration at fl in one tuned cathode resonator, and at f2 in the other tuned cathode resonator. The impedances, areas and reactances of the loops B9 and 69 are considered in arriving at the dimensions of the chambers re Separate tuning.

quired to tune the cathode circuits to the desired operating frequencies fl and f2.

The output circuit including a line 10 may be coupled by transformer means to the anodes f the generators. In a preferred embodiment, I use a line 12, the inner member of which is coupled by a loop ll into the resonant chamber. The transformer coupling is adjustable by means of a variable coupling 14. The coaxial line 12 is shorted at the outer end for voltage of the generated frequency by an adjustable coupling arrangement 16, at which point the line is of about zero high frequency potential. The end of the loop ll connected to the chamber 50 is also of low or zero high frequency potential. The loop ll extracts timing modulated generated oscillation from the cavity space and feeds the same over line 12 to the output line 10. The manner of arranging this type quarter wave line section transformer is well known in the art. The circult is equivalent to a parallel tuned auto-transformer, and the voltage may be stepped-up or down depending on the voltage characteristic of the line 72 and the point to which coupling 14 is adjusted. The line 12 might be of the order of \/2, or a multiple thereof. The coupling 14 may be adjusted to get a step-down transformer action, thus matching low impedance coaxial transmission lines. Preferably, the coupling 14 is adjusted to get the desired match. In arriving at the line length, the reactance of the loop ll coupling the chamber into the line, the impedance of the loop, the reactance at 14, etc., are considered.

A coaxial line 10' feeds the generated oscillations back to the control grid G of the electron beam tube. This feedback circuit is similar in principle and operation to the output circuit line 10, line 12, etc., and in the feedback circuit numerals similar to the numerals used in connection with the output circuit primed have been used The line 10 and the line 10 may be of any length provided that they are matched at both ends to the impedance into which they feed.

The final electron collecting target plates of the beam tube 20 are connected to the cathodes 52 and 52', respectively, over cathode resistance 15 and 15'. The adjacent terminals of the resistances I and 15 are connected to the positive terminal of a source of potential which supplies the potential to the final collecting electrodes in the electron multiplier amplifiers #1 and #2. At the median condition the cathodes 52 will operate at a D. C. positive potential with respect I, to D. C. ground (this means that the grids 56 will look negative D. C. to the cathodes 52, and the tubes will be partially biased off). On this D. C. potential is superposed the variations occasioned by modulation.

In operation, the cathode 52 circuit is detuned in such a manner that this tube alone tends to make the oscillator operate at a frequency fl, higher than the median operating frequency, and conversely, the cathode 52 circuit is detuned in such a manner that this tends to make the oscillator operate at a frequency f2, lower than the median operating frequency. When the two similar tubes are equally biased the combined system will operate at a frequency midway between fl and 12. This is a stable operating point, for should one of the tubes tend to assume control it will automatically bias itself off. As a tube tends to take more than its share of control, the average cathode current will increase, which 8 will increase the bias. With increased bias this tube will have less influence (i. e., its gm is lowered) and the other tube will assume its share of control. On the other hand, if the bias on the two tubes is unequal, the frequency of oscillation of the combined system will shift in such a direction as to favor the tube with the lower bias. The maximum frequency deviation will be realized when one of the tubes is biased to cutofi and the other is driven near zero bias. When one tube is at cutofi and the other near zero bias, the cutoff tube will have no control or effect, whereas the near-zero-bias tube will have more than its normal or median point influence. In this event, the frequency of oscillation of the combined system will be that of the near-zerobias tube alone.

This controllable frequency oscillator is made to follow or reproduce the incoming signal in the following manner. Suppose the frequency of the incoming signal is increasing, or that the frequency of the signal generated by the oscillator tends to fall behind the incoming signal as illustrated in Fig. 5. The number of electrons arriving at the final collector electrode of amplifier #1 increases and the number of electrons to the final collector electrode of amplifier #2 decreases. These changes in current change differentially the biases on the oscillator tubes so as to increase the effectiveness of the oscillator tube, which tends to operate at fl and drive the entrained circuit at a frequency above the median operating frequency and to decrease the effectiveness of the oscillator tube, which tends to operate at f2 and drive the entrained circuit at a frequency below the median operating frequency. In a like manner it may be demonstrated that should the frequency of the incoming signal be decreasing, or should the signal generator tend to get ahead of the incoming signal, the oscillator tube tending to drive the generator circuit at a lower frequency will be favored at the expense of the oscillator tube tending to drive the circuit at the higher frequency. Over a limited range these changes in relative effectiveness will be proportional to the changes in bias, that is, the forces tending to restore in-phaseness are proportional to phase shift. These restoring forces will be balanced against forces exhibited by the oscillator which has a predisposition to operate at the median frequency as well as inertia which tends to make it resist changes in frequency. Inasmuch as all of these forces are proportional to displacement for small displacements, the controllable frequency oscillator may be expected to follow the incoming signal provided there is sufficient amplifier gain, and provided the rate-of-change-of-frequency and the frequency deviation does not exceed limiting values imposed by the characteristics of the tubes and circuit.

Many arrangements may be provided for supplying to the cathodes 52 and 52' the differential operating potentials required in the operation described above. In Figs. 8 and So, I have shown a preferred embodiment of my phase discriminator electron amplifier tube. In this tube, which will be described in greater detail hereinafter, the cathode K may operate at say minus 500 volts direct current. The electron multipliers #1 and #2 each comprise five amplifier and multiplier stages. The several stages are connected in series to ground by resistances R, which may be of the order of 1000 ohms each. The target electrode AC is integral with or connected to a metallic member P between the two amplifier chains. The member P serves to electrically separate the amplifiers, and also to supply potential to the cascaded multiplier amplifier stages connected by resistances R. A lead from P through the tube envelope is connected to a negative point A on one of the direct current potential supply potentiometers, say PR. The collector electrode CE in amplifier-multiplier #1 is coupled to the cathode 52, While the collector electrode CE in multiplier-amplifier #22 is coupled to the cathode 52. As stated hereinbefore, the cathode 52 is coupled to a point on a positive source by a resistance l5, while the cathode 52' is connected to this same point by a resistance 15. The resistances i5 and 55' may each be of the order of 1000 ohms. In the preferred embodiment the adjacent ends of i5 and 75 may be supplied with a voltage of the order of +200 volts direct current. Then in the presence of carrier frequency the biasing potentials at the cathode ends of resistance I5 and 75' are equal and are of the order of 100 volts bias, so that the oscillators are equally effective and are entrained to operate at the median frequency.

When the current through the electron amplifier-multiplier #1 increases, the bias at the oathode end of 15 becomes less positive or more negative. This is the same as making the grid of this oscillator more positive or less negative, and this entrained oscillator is more effective. At the same time the current through the electron multiplier-amplifier #2 decreases, and the potential at the cathode end of resistance 15' becomes more positive or less negative. This has the effect of making the grid of this oscillator more negative and less effective in setting the frequency of the operation of the entrained oscillator which now swings toward the frequency fl.

When the current through the electron multiplier-amplifier #2 increases, and that through 1 decreases, the reverse operation takes place. In this manner, as described hereinbefore, the entrained oscillators are caused to repeat exactly the frequency modulated signal supplied as input to the cathode ray tube discriminator and electron multiplier-amplifier 20, designated generally in Fig. 2, and in considerable detail in Figs. 8 and 8a.

In Figs. 8 and 8a, I have shown somewhat completely an embodiment of my improved beam tube. This tube comprises an envelope including a cathode K, a heater element HE therefor, and an electrode intensity control grid G. Since it is desired to provide a beam of electrons in the form of a sheet in a plane perpendicular to the drawing sheet, the cathode K and its heater element HE are elongated in this direction as shown in Fig. 8c. The cathode may run at say -500 volts direct current. The control grid G may be a metallic member elongated in the same dimension as the cathode, and the grid may operate at an adjustable potential more negative than the cathode. The limiting apertures or anodes AI and A2 are also elongated in the same dimension and operated at adjustable potentials which are positive with respect to the grid potential. Potentiometer resistances 90, to which the cathode and grid are adjustably tapped, supplies the desired direct current potentials. In practice, the electrodes described hereinbefore, including the two limiting aperture electrodes, are insulated from the metallic tube envelope. The apertures Al and A2 are spaced along the tube axis as desired, and the apertures are drawn in as indicated to concentrate and bunch the beam in a narrow concentrated sheet perpendicular to the plane of the drawing. These electrodes Al and A2 may also be supplied with direct current potential by adjustable taps on apotentiometer resistance 92. The aperture electrodes Al and A2 are positive relative to the cathode and grid, and electrode A2 may be more positive than electrode Al, although it willbe understood that here, as in conventional beam tubes, the sign of the electrode potentials is a matter of choice, and all that is necessary is that the relative potentials are as required. For instance, shifting of the ground connection (here shown as positive) may change the sign of certain of the direct current potentials.

Since the electron beam is concentrated in a flat sheet of electrodes of considerable extent in a plane perpendicular to the plane of the drawings, this beam in the absence of applied input and feedback strikes the initial collecting electrode AC at a point such that the electron supplies to the #1 and #2 amplifier-multipliers are evenly distributed between these amplifiers.

The electron discharge tube also includes a pair of the electron beam deflecting electrodes 22 and 2 to which the incoming signal is applied. These deflecting electrodes are connected to a potentiometer resistance PR connected across a direct current potential source with the positive end thereof grounded. This adjustable direct current supply is a means for focusing and centering the beam, as desired, on the electrode AC. PB, 00 and 02 may represent a single potentiometer being shown separate in order to simplify the connections.

The electron multipliers #1 and #2 may ine clude the desired number of stages arranged as shown and separated in the tube structure by a metallic partition P fastened to or integral with the metallic tube envelope. Each amplifier stage starting with the initial stage is connected with the following stage by a resistance B. so that each stage operates at a somewhat higher direct current potential than the preceding stage, because the final stage is grounded as is the positive terminal of the direct current source in the embodiment illustrated. When a potential of the order of 200 volts direct current is supplied to the electrode AC and a potential of about +200 volts is supplied to the adjacent terminals of the resistances l5 and 15' (Fig. '7), and thence to the collector electrodes CE and CE, and resistances i5 and '85 are of the order of 1000 ohms, the resistances B may be of the order of 1000 ohms. The multiplieramplifiers #1 and #2 may each be substantially as described more in detail in Fig. 6 of a paper by V. K. Zworykin and J. A. Rajchman entitled, The electrostatic electron multiplier, I. R. vol. 27, No. 9, September 1939.

The feedback from line 10' of Fig. 7 to the control grid G may be through a coaxial line and an adjustable coupling transformer substantially similar to the output line l0, l2 coupling and transformer described hereinbefore. A similar double line arrangement illustrated diagrammatically in Fig. 8 may be used as a transformer means introducing the incoming signal to the deflection electrodes. These line arrangements also provide a means of applying the adjustable direct current potentials to the grid G and the centering potential to the deflecting electrodes.

The electron beam tube operates as a current phase detector, since the relative intensities of the amplifier outputs depends on the relative phases of the incoming wave applied to the de-

Images