US2458652A

US2458652A - Electron discharge apparatus

Info

Publication number: US2458652A
Application number: US715900A
Authority: US
Inventors: Raymond W Sears
Original assignee: Bell Telephone Laboratories Inc
Current assignee: AT&T Corp
Priority date: 1946-12-13
Filing date: 1946-12-13
Publication date: 1949-01-11
Anticipated expiration: 1966-01-11

Description

Jan. 11, 1949. R. w. SEARS 2,458,652

I ELECTRON DISCHARGE APPARATUS Filed Dec. 13, 1946 4 Sheets-Sheet l //v l EN TOR R. W SEARS ATTORNEY Jan. 11, 1949. R. w. SEARS,

ELECTRON DISCHARGE APPARATUS Filed Dec. 13, 1946 4 Sheets-Sheet 2 U U DU U DU U U U U U D U UH U U U D U U U HEN v at lNl ENTOR V R14. SEARS ATTORNEY Jan. 11, 1949. R. w. SEARS ELECTRON DISCHARGE APPARATUS Filed Dec. 15, 1946 4 Sheets-Sheet 3 BEAM POSITION 4 M 4 Z 9 1 5 l m 4 CI f z 4 ,2 X

FIG.

/NVENTOR R. W SEARS ATTORNEY Patented Jan. 11 194g UNITED STAT-ES PATENT OFF-ICE 11 Claims.

1 This invention relates to electron discharge devices and moreparticularly to cathode ray devices of the type disclosed in the application Serial No. 715,999, filed December 13, 19%, of

.George Hechtand especially suitable for-use in signal translating systems wherein speech or other complex waves are sampled atsuccessive intervals and the samples are resolved into, code pulse groups eachcorresponding to .a respective sample amplitude.

Cathode ray devices of the type disclosed in the above-identified application comprise, ingeneral, a coding mask or electrode having a plurality of rows of apertures therein, the apertures being arranged in predetermined relation so that an electron beam swept across the several rows will .produce pulse. groups, the character .of each group being determined by the position of the sweep. The position of the sweep is determined, in turn, by the amplitude of the speech or other wave sample at the time of the sweep.

As disclosed in. the application, improved. performance for such a device may be realizedby providing an auxiliary electrode or. grid which functions to hold the beam in the prescribed code position during each sweep or. the beam .across the coding electrode or mask.

One object of this invention is to improvethe signal resolution'in cathode ray devices of the type above described.

Another object of this invention is to simplify the construction of such devices and of. systems including them.

A further object of this invention is to reduce the size of cathode ray devices for use. insignal coding systems.

Still another object of this invention is to'facilitate exactalignmentof the auxiliaryand coding electrodes.

A still further object of this invention is-to assure exact parallelism of the elements or wires of'the auxiliary electrode andmaintenance of such relation of the elements or wires during the operation of the device.

' Still another object of this invention is to obtain faster quantizing.

In accordance with one feature of this invention, the auxiliary electrodahereinafter referred to as the quantizing electrode or grid, comprises a plurality of elements or wires having a secondary electron emission coemcient greater 7 than unity. A collector electrode is provided in cooperative relation with the quantizing electrode a and in use of the device is connected to the'beam deflection system through a feedback circuit.

-'Ihe elements or wires are parallel and, closely .,adja cent, for example spaced a. distancecomparableto the diameter of the beam in the vicinityof the quantizing electrode, so that asubstantial current of instantaneousamplitude .determined by the position of the beamrelative to two adjacent elements or wires, i'lowsto the collector electrode. This current. is..a maximum whenv the beam is centeredupon any element or wire andis aminimum whenthebeam is centered upon the space between two adjacent. elements or .wires.

in accordance with another .feature of this invention, the quantizing electrode ,or grid is-con- .structed so that the elements .or wires are .main

tained continually under tension and in .exact parallel relation. In oneillustrative construcgroove-andis constantly urged toward thegroove by a resilient;member-mountedby the foundation or frame.

In accordance with a-f-urther. feature of; this invention, the coding and quantizing electrodes are, fabricated in a unitaryassembly' wherein the grid openings are positioned in accurate-alignment-; with certain of the openings in the coding electrode and are'maintaincd in such relation discharge device illustrative of one. embodiment of the invention, aportion of the enclosing-vessel being broken away to show the electrodes'--more clearly;

-Fig.-2 is a sectional view, taken along plane 2-2 "of-Fig.1, showing the form-and relation of the deflector plates;

Fig." 3 is a'sectional view, taken along plane 3-3 of Big. 1, showing the configuration and relation of the grid and collector electrodes;

vFig. 4 is aview in crosssection andto'an enlarged scale of the. grid electrode assembly;

3 coding electrode included in the device illustrated in Fig. 1;

Fig. 7 is a fragmentary sectional view to an enlarged scale taken along line 1-! of Fig. 6;

Fig. 8 is a circuit diagram showing one manner in which the device may be operated;

Fig. 9 is a diagram illustrating the relation of the grid openings and the apertures in the coding electrode in the device illustrated in Fig.

Fig. 10 is a graph showing the relationship of the current to the collector electrode and the beam position in the direction of the plane of and transverse to the grid wires; and

Figs. 11 to 11C are diagrams illustrating the relation of the signal and feedback potentials for various beam positions and for several relative amplitudes of the two potentials.

Referring now to the drawing, the electron discharge device shown in Fig. 1 comprises a highly evacuated cylindrical enclosing vessel formed in two parts IDA and 3B of vitreous material and having mounted therein, adjacent one end of the part lflA, an electrode system constituting an electron gun for producing a highly concentrated electron beam, for example a circular beam of the order of 0.008 to 0.012 inch in diameter. The electron gun may be of any one of a number of known constructions and comprises, generally,

a cathode H, a concentrating and control electablished by way of leading-in conductors [5 connected to terminal prongs IS on the base I! secured to the vessel ID. The electrode I4 is provided with one or more metallic fingers l8 which firmly engage a cylindrical conductive coatin [9 upon the inner wall of the part lllA enclosing vessel, which coating serves primarily as a shield. Electrical connection to this coating and, hence, to the electrode M may be established by way of a leading-in conductor (not shown) connected to one of the prongs IE or alternatively by way of a conductor (not shown) sealed in the side' wall of the vessel portion lOA.

Mounted opposite the electron gun and at right angles to each other are two pairs of parallel deflector plates a, 20b and 2 la, 2lb respectively,

the plate Zia having side flanges 22 at right angles to the body thereof, as shown clearly in Fig. 2. Leading-in conductors 23 connect the several deflector plates to respective terminal prongs l6.

Mounted within the part MB of the enclosing vessel and supported thereby is an electrode unit all of the electrodes in which are in axial alignment with the electron gun and the deflection system defined by the plates 20 and 2i. This unit comprises a collector electrode 24, a quantizing grid 25, a coding electrode 26 and a target or output electrode 2'! arranged in the order named, as illustrated in Fig. 1, and held in position relative to one another by four ceramic rods or tubes 28 to which they are firmly affixed, as by a suitable cement.

The target or output electrode 21 may be a circular metallic plate having a secondary emission coeflicient other than unity, for example of nickel or carbonized nickel, of slightly less diameter than the internal diameter of the portion IOB of the enclosing vessel and has affixed thereto, as by rivets 29, two or more spacers, for example, flexible wires 30 of tungsten, the ends of which engage the inner wall of the vessel part NIB. Electrical connection to the output electrode 21 may be established by way of a rigid leading-in conductor 3i connected thereto and to a cap terminal 32 sealed to the vessel part IUB.

The coding electrode 26, shown in detail in Figs. 6 and 7, also may be a circular metallic plate, for example of nickel or carbonized nickel, of a diameter slightly less than that of the internal cylindrical wall of the vessel part I03, and provided with apertures 33 into which the ceramic rods or tubes 28 are fitted, the plate being affixed to the rods or tubes as noted heretofore. The plate is provided with an elongated rectangular aperture I and a plurality of rows II to VII inclusive, of rectangular apertures, the rows being parallel to one another and to the aperture I and all the apertures in rows II to VII having their corresponding sides parallel. The aperture I and those in rows II to V inclusive, may be formed by punching alone; the apertures in rows VI and VII may be formed, as illustrated in Fig. 7, by first milling grooves or recesses 34 in the plate and then punching to produce the apertures.

The number of apertures in rows II to VII are 2, 4, 8,16, 32 and 64 respectively. Although the operation of the device will be described in detail hereinafter, it may be noted here that the beam is deflected selectively in one direction, horizontally in Figs. 1 and 6, in the direction of and outside row VII by the deflector plates 20a and 20?) so that it is opposite either one of the apertures in row VII of the coding electrode or an area between two apertures or just beyond the last aperture, i. e. the rightmost aperture in Fig, 6.

The beam then is swept across the coding elec trode in the direction normal to the rows of apertures, i. e. vertically in Figs. 1 and 6. The apertures in the several rows are so arranged that for each position to which the beam is stepped it sweeps across a different combination of apertures and solid areas in the coding electrode whereby 128 different pulse or code groups may be produced at the target or output electrode 21.

In the particular construction illustrated, in

5 each of rows III to VII the apertures are equally spaced and, with the exception of the first aperture, i. e. the leftmost aperture in Fig. 6, are of the same width, i. e. the horizontal dimension thereof in Fig. 6. The width of the apertures in row VII is of the same order of magnitude as the beam diameter. The first aperture in each of rows II to VII is somewhat wider than the other apertures in that row for reasons which will appear presently. Advantageously, all of the apertures are of the same height, 1. e. the vertical dimension in Fig. 6. The leftmost edges of the first apertures in the several rows are in line with one another and the leftmost edge of the aperture I. Also each left side of each aperture in rows II to VI is aligned with a corresponding side of an aperture in each of the succeeding rows III to VII. Connection to the coding electrode is made by way of a leading-in conductor 39 connected to a cap terminal 49.

The quantizing electrode 25 comprises a circular metal plate 35, for example of nickel, of a diameter slightly smaller than that of the internal cylindrical wall of the vessel portion [0B and having aflixed thereto, as by rivets 36, resilient wire spacers 37, similar to the spacers 30, the ends of which engage the inner wall of the vessel portion I DE. The plate 35 is provided with a rectangular aperture 38 of somewhat larger dimensions than those of a rectangle suinciently large to encompass the aperture I and rows II gage-re to VII in the coding electrode 26. As illustrated in Figs. 3 and l, the plate 35 has thereon parallel rails or raised portions 4| adjacent the longer sides of the aperture 38 and one or both of these portions is provided with a longitudinally extending groove 42. Both of these portions 4| are provided also with parallel, transverse V- shaped grooves or slots 43 (see Fig. 5), each groove 53 in one portion 4! being an alignment with a corresponding groove in the other portion.

Seated in the grooves 43 and affixed, as by brazing, to the raised portions 4| adjacent the outer sides thereof are fine parallel grid wires 45 which, in a particularly advantageous construction, are of a material having a thermal coefiicient of expansion substantially equal to or greater than that of the plate 35 and having also a secondary electron emission coefiicient greater than unity. In a specific construction, the plate 35 may be of nickel and the wires 44 of the copper-nickel alloy known as- Monel, the latter having a secondary electron emission coefiicient of approximately 3.5. Aligned with one of the grooves 42 and bearing against all the wires 4 3 is a rigid rod or wire 45, for example of tungsten, which is forced partially into the groove by a spring strip 46, for example of beryllium-copper, coextensive with the rod or wire 45 and secured to the plate 35.

In the fabrication of the quantizing electrode, the grid wires 44 are stretched uniformly in place on the plate 35 and secured to the latter, after which the rod or wire 45 and spring 46 are mounted in place. The rod and spring serve to tension the grid wires and to maintain a tension therein during the evacuation treatment and operation of the device, so that the parallel relation of the grid wires is preserved. During the evacuation treatment of the device, the quantizing electrode is subject to substantial temperature changes and the grid wires may bow. However, because of the equality or difference in the thermal coefficients of the plate 35 and wires 44, the latter always are held under tension during operation of the device and, hence, the parallel relation thereof is maintained.

The quantizing electrode is mounted parallel to the coding electrode 26 by the rods or tubes 28 so that the aperture 38 is opposite the coding apertures in the electrode 26 and the grid wires 4d are exactly parallel to the vertical edges (in Fig. 6) of the apertures I and in the rows II to VII. The grid wires 44 are spaced so that there is a grid opening opposite each of the smaller apertures in row VII in the coding electrode. In a specific construction, the apertures mentioned may be 0.010 inch wide, spaced 0.024 inch center to center, and there may be 129 grid wires 0.004 inch in diameter and spaced 0.0116 inch center to center, The leftmost aperture (in Fig. 6) of row VII may be 0.054 inch wide. One of the grid wires is positioned opposite the left-hand (in Fig. 6) edge of this aperture. An additional pair of grid wires 44 is positioned just beyond the rightmost aperture in row VII.

The quantizing grid is aligned with the coding electrode, i. e. the vertical axes (in Figs. 3 and 6) of the two are in the same plane, Hence, the central grid openings are aligned, or essentially so, with the central apertures in row VII of the coding electrode. However, as indicated above, the center-to-center spacing of the apertures in row VIII is greater than twice the center-to-oenter spacing of the grid wires 44 so that the apertures to either side of the center in row VII are offset 6. relativeto the corresponding grid opening and the ofisje't increases as the distance fromthe center of" theiow' increases. The reason for this construction will'be understood from the following considerations with reference to Fig. 9.

Operation-of the device requires that each aperture in row VII in the coding electrode have a corresponding grid opening associated therewith. Consider now an electron beam B which is deflected in the direction indicated by the arrow in 9 about the pivot point X in line witha central aperture a in row VII in the coding electrode 26. It will be noted that for this apertur'e' iii the grid opening between wires 44A is centrally aligned with the aperture. However, if the beamfis deflected to pass through an apert'ure'z to one side of the central aperture a, because or the distance between the grid and the plate 25; it is necessary that the corresponding grid wires 442 be offset relative to the aperture 2 in order that the beam will pass through and be centered relative'to the grid opening corresponding to the aperture 2. That is to say, each grid opening must be aligned with the corresponding' aperture 'whenviewed from the point X; It willbe noted that the amount of offset necess'ary'increases as the distance between openings a ands increases. The ofiset between grid wire openings an'd apertures in the coding electrode Zifwill be determined inany particular case by the distance y between the grid 44 and the coding plate 26 and the distance between the point X and the grid plane, and, of course, can be calculated from the geometric relations involved. Theparticu'lar dimensions for grid opening and apertures'set forth hereinabove have been found satisfactory for a spacing y of 0.142 inch between the grid 44 and the plate 26 and a distance of 4.126 inches from the point X and the grid plane. The point X is located, in an actual system, be tween the deflector plates 20a and 20?), as is known in the art.

The collector electrode 24 is rectangular as shown" in Fig. 3, of dimensions slightly greater thanthe corresponding dimensions of the apertui'e 3:8 andis" centrally aligned with the apertime 38. It may be formed of a suitable metal, such asnickel, and is mounted from the rods or tubes 28 by metalbrackets 48.

Electrical connection to the quantizing and. collector electrodes may be made by way of leading-in conductors 49 connected to cap terminals 50 afix'ed to the vessel portion 103.

In the fabrication of the device, the electron gun and the deflector plates are mounted in the vessel portion IOA and the collector, quantizing, coding and output electrodes are mounted in the vessel portion MB. The two assemblies are joined by fusing the two vessel portions together and in such relation that the grid wires 44 are accurately parallel to the deflector plates 20a and 20b and the vertical (in Fig. 3.) axis of the grid is accurately aligned with the median plane between these deflector plates.

One'ma'nne'r inwhich the device shown in Figs. 1 to 7 and described hereinabove may be operated for pulse'code modulation is illustrated in Fig. 8. 'Thdfocus'sing and accelerating electrode i4 and theq'iiantizing electrode 25 are connected direct- 137 to ground as shown. The cathode is held at a high negative potential, for example 1000 volts, relative"to ground and the electrode [2 is held ata negative potential; for example, of less than 1 00 volts,relative to the cathode by way of the potentiometerresistance 5|. The potential of 7 the electrode may be varied to adjust the beam current. The focussing electrode I3 is biased positive relative to the cathode, for example, of the order of 250 volts, by Way of the poteniometer resistance 52.

The deflector plates 20a and 20b are connected to ground through resistors 53 and 54 and the input signal samples are applied to the deflector plate 28a from a suitable circuit 55. The other deflector plates 25a and 2Ib are balanced to ground by the resistor 56 and have a sweep voltage to provide a linear sweep, impressed therebetween from the source 57.

The coding electrode 26 is held positive relative to ground, for example, of the order of 90 volts. The collector electrode 24 is biased positive relative to ground at a potential equal'to or higher than that of the coding electrode, for example 90 volts, and is connected to the deflector plate 200: by an amplifier 58, the electrode 24 being connected to the input side of the amplifier.

The target or output electrode 2? is connected to ground through a resistor 59.

The resistance 53 is made relatively small, for example of the order of 220,000 ohms and the deflector plates are constructed and arranged so that the capacitance therebetween is small, for example of the order of one micromicrofared, whereby the time constant of the resistancecapacitance circuit defined thereby also is Very small so that high frequency input signal pulses may be distinguished.

The operation of a device of the type herein described and illustrated, in a pulse code modulating system, is briefly, as follows: A signal, for example of audio frequency, is amplitude sampled at high frequency and the amplitude samples are impressed across the deflector plates 26a and 20b. The beam is deflected accordingly, horizontally in Figs. 1 and 6, to be opposite one of the apertures in row VII in the coding electrode or opposite a space between two apertures or beyond the last (rightmost in Fig. 6) aperture in the row. It is then swept across the coding electrode, 1. e. in the vertical direction in Figs. 1 and 6, whereby current pulses, or no current for the beam position beyond the last aperture in row VII, are produced at the target or output electrode 21. As noted heretofore, the pulse group produced by sweeping of the beam in any position to which it is deflected is different from the pulse group for any other position. The deflection position, as has been noted, is determined by the amplitude of the input signal sample;

hence, each pulse group corresponds to or represents a respective signal amplitude so that each input sample is coded. Inasmuch as, as noted heretofore, in the device shown and described one hundred and twenty-eight different pulse groups may be obtained, one hundred and twenty-eight different amplitude samples may be distinguished and coded.

The beam position, for no input signal, may be such that it passes to the center of row VII of apertures in the coding electrode 26, so that for positive input signals the beam is stepped in one direction, e. g. to the left in Fig. 6 and for negative signals it is stepped in the opposite direction, e. g., to the right in Fig. 6. Alternatively, the no signal position may be at either end of row VII.

As noted heretofore, the first (leftmost in Fig. 6) aperture in each of rows II to VII in the coding electrode is somewhat larger than the remaining apertures in the respective row. This allows recognition of input samples of somewhat greater amplitude than if the first apertures were of the same size as the remainder and, in effect, provides a peak limiting effect. A similar limiting effect obtains at the other end (rightmost in Fig. 6)of the coding system.

It will be appreciated that in order to realize proper coding of the input signals, once the beam is deflected to any code position it must remain in that position while it is swept across the coding electrode, i. e. it must follow the path (vertical in Figs. 1 and 6) corresponding to the amplitude of the input sample. Each path, except the leftmost one in Fig. 6 is between a respective pair of grid wires. If, due, for example, to slight misalignment of the electrodes or electrical disturbances, the beam departed from the requisite path duringa sweep, a false pulse group might be produced at the output or target electrode and incorrect coding of the input signal would result. The quantizing electrode 25 prevents such incorrect coding as will be understood from the following considerations with reference to Figs. 10 and 11 to 11C inclusive.

If the beam were swept across the grid wires it, i. e., horizontally in Figs. 1 and 3, the beam current to the quantizing electrode would vary periodically, being a maximum when the beam is centered on any grid wire 44 and a minimum when the beam is centered on the opening between two adjacent grid wires. The secondary electron current from the grid wires varies in like manner so that the current to the collector electrode 24 varies in similar wise. If the beam diameter were less than the width of the grid openings, the current to the collector electrode would vary between zero and a maximum; however, When the beam diameter is somewhat greater than the width of the grid openings, the collector current does not fall to zero but varies between a minimum of amplitude d illustrated in Fig. 10 and a maximum e. The absolute values o-f 'the maximum and minimum collector currents in any device will be determined, of course, primarily by the beam current and the secondary emission coefficient of the grid wires. The minimum value is dependent also upon the relative magnitudes of the beam diameter and the grid opening width and, additionally, the transverse current distribution in the beam. As is known, the current density adjacent the beam boundary is less than that at and near the center of the beam. Although the specific form of the collector current-beam position characteristic thus is dependent upon a number of factors, nevertheless the general relation is as illustrated in Fig. 10 and the collector current is a maximum when the beam is centered upon a grid wire and is a minimum when the beam is centered upon a grid opening.

The collector current is fed to the amplifier 58 and. converted into a voltage between the deflector plates 26a and 2012. Thus, for any position of the beam, in the horizontal dimension in Figs. 1 and 8, the effective potential between the deflector plates 20a and 20b is the sum of the potential due to the input signal and the feedback potential due to the collector current. The feedback potential is dependent, of course, upon the position of the beam relative to the grid wires 44.

As. illustrated by the line D in Fig. 11, the

beam position in the direction normal to the grid wires 44 varies linearly with the deflecting potential effective between the deflector plates ae ssi 9 28a and 20b. The effective deflecting potential for any beam position is the resultant of the signal potential due to the input circuit 55 and the feedback potential. For the condition of negative feedback, i. e., the condition where the potential of the feedback potential is such 1 that the latter opposes the signal potential, the signal potential-beam position relation is as illustrated by the curve D. For simplicity of illustration, in Figs. 11 to 110, the signal curve is shown for the condition when the minimum feedback potential is zero, which condition obtains for a beam diameter less than the width of the grid openings. It will be apparent from what follows that the same analysis follows for conditions where the minimum feedback potential is other than zero, as for the case where the beam diameter is somewhat greater than the grid opening width.

Consider now the conditions extant when the beam is at the position (Fig.11). In order to be at this position, the effective deflecting potential must be of the magnitude E. The feedback potential is of the amplitude E-E so that the signal potential is of the value E. If new the beam is subjected to a disturbance, for example because the signal amplitude increases slightly, the beam tends to move to the right, i. e. toward the grid wire 4 32. However, any such movement would result in an increase in the feedback potential and, consequently, a decrease in the effective deflecting potential. Similarly, if there is a disturbance such that the signal amplitude decreases slightly, the beam tends to move to the left, i. e. toward the grid wire 4 31, Such motion would result in a decrease in the feedback potential and, consequently, an increase in the efifective deflecting potential.

Hence, for the position 0, the conditions are such that the beam is in equilibrium or stable. Similar analysis will show that for any beam position between substantially O1 and 02, the conditions are such that the beam will be held in that position or, stated in another way, the beam will be held in the opening between grid wires 341 and M2 and against the grid wire 442.

Consider, now, the conditions extant for the beam position 03. For this position, the signal voltage is of amplitude E4, the feedback voltage is E4E3 and the effective deflecting potential is Ea. If for any reason the beam is disturbed, for example if the signal voltage increases slightly, the beam moves to the right. As it so moves the feedback voltage decreases. Consequently, the efiective deflecting voltage increases and the beam is deflected further to the right. The action is cumulative and continuous so that the beam continues moving to the right until it reaches a position between 04 and 05, the beam is not in equilibrium. If, at position 03, the signal voltage decreases, the beam tends to move to the left. However, such motion results in an increase in the feedback voltage and a further decrease in the effective deflecting potential so that the beam moves to the left until it reaches a position between 01 and 02.

Similar analysis will show that for any condition resulting in a beam position between 02 and Or the beam will be deflected to a position either between 01 and 02 or between 04 and 05, the direction of deflection being dependent upon the direction from which the position 03 is approached. Thus, for positions between 02 and Or the beam is not in equilibrium and if the beam reaches such a position it is deflected automatically to a stable position, i. e. between 04 and 05 or 01 and O2.

Viewed in another way, if the feedback circuit is closed while the beam is deflected in response to application of a signal from the circuit 55, of amplitude between E2 and E5, the beam would come to rest at a position between either 01 and 02 or 04 and 05 depending upon whether the direction of deflection due to this signal were to the right or to the left respectively. Thus, it is possible that for signals between the amplitudes noted two different pulse groups or codes could be produced.

For the particular amplitude of feedback voltage illustrated in Fig. 11, substantially equal to the voltage increment corresponding to Es-E1, i. e. substantially equal to that corresponding to a beam deflection from the center of one grid opening to the next adjacent one, two different pulse groups or codes could be produced for the range of signal voltages between E5 and E2. The two codes would be successive ones. That is, if the pulse group associated with the opening between

grid wires

441 and 442 is'identifled as code n, that corresponding to the grid opening between

wires

442 and 443 could be designated as n+1.

If the amplitude of the feedback voltage is increased and the feedback circuit is closed as the beam is deflected in response to an input signal, the separation between the two possible pulse groups increases. This may be seen from Fig. 11A wherein stable beam positions for several input signals are indicated by points on the curve D1 and several codes are designated for the respective grid openings, the feedback potential being of maximum amplitude several times the deflection voltage movement requisite to swing the beam from one grid opening to the next adjacent one. For the signal E1, under the conditions postulated, it will be seen that codes n+3 or n2 may be produced; for an input signal Es, codes n+4 or n-l may be produced;. for an inputsignal E9, codes n+2 or n3 are possible; and for an input signal E10, codes n+1 or n-4 are possible.

For maximum amplitudes of feedback po ential less than that equal to the deflection voltage increment correspondingto the center-to-center spacing of the grid openings, the conditions possible may be seen from Fig. 11B. In this figure, the feedback voltage amplitude is approximately one-half that of the deflection voltage increment and such that the curve D1 has a slightly negative slope for beam positions between 02 and O4 and corresponding positions relative to other grid openings. From what has been said heretofore it will be appreciated that for the conditions illustrated in Fig. 113, the equilibrium or stable positions of .the beam fall between 01 and 02, between 04 and 05, etc., and that positions, such as between 02 and 04, for which the curve D1 has negative slope are unstable beam positions. For such positions, or more accurately for the signal voltages corresponding thereto, two codes are possible. However, it will be noted that the rangeof signal voltages, e. g. between E5 andEz, for which such double coding is possible is considerably smaller than that for the condition or feedback potential amplitude resulting in the curve D1 in Fig. 11. 1

If the amplitude of the feed back potential is decreased still further, i. e. if the maximum amplitude thereof is less than one-half the potentialincrement corresponding to a beam deflection equal to the center-to-center spacing of adjacent grid openings, the conditions possible will be understood with reference to Fig. 110. For such amplitude of feedback potential, all portions of the curve D1 have a positive slope; for points between 01 and 02, between 04 and O5 and similar regions, the slope is greater than that of line D; for other points, e. g. between 02 and 04, the slope is less than that of line D. Beam positions corresponding to the first points noted, e. g. position 0, are equilibrium or stable positions, as is apparent from what has been said heretofore. Additionally, beam positions at the other points noted, say the position 03 or other position between 02 and 04, also are equilibrium or stable positions because of the small positive slope of the curve D1 at these points. That is to say, the change in effective deflecting potential associated with departure from the beam in either direction from position 03 is insufficient to drive the beam further in that direction. Consequently, for the feedback potential amplitude-deflection potential amplitude relation illustrated in Fig. 110, every beam position is an equilibrium or stable one.

From the conditions and relationships illustrated in Figs. 11 and 11C and pointed out hereinabove, certain characteristics of devices including a quantizing electrode or grid will be apparent. First, it will be noted that in such a device, once the beam has been deflected to any code position, by the application of a potential between the deflector plates 20, it will be held in a stable position by the action of the grid as the beam is swept across the coding electrode by the potential applied between the deflector plates 2|. If, for example, any two adjacent grid wires 44 are misaligned relative to the direction of the beam sweep, once the beam is deflected to a stable position to pass between these two wires it will be held between them during the sweep cycle by virtue of the feedback potential due to secondary emission from these wires. Thus, once the beam reaches a stable or equilibrium position between two wires, when it is swept, the code pulse group obtained is that corresponding to the beam position between these two wires.

Furthermore, it will be noted that the beam will remain in any stable position it reaches despite variations in the applied signal, i. e., that applied between the deflector plates 20 from the input circuit 55, during the sweeping cycle. The magnitude of variation for which this will obtain is determined, of course, by the amplitude of the feedback potential and will be greater the larger the feedback potential. Thus, for the condition illustrated in, Fig. 11, if the beam reaches a stable position, e. g., between 01 and 02, it will remain between the wires 44 1 and 442 during the sweeping cycle and produce the corresponding code pulse group for variations in input signal, during the sweeping period, resulting in efiective deflecting potentials between about E1 and E2. For the conditions illustrated in Fig. 11A, the range of voltage variations for which the beam will remain in a stable position between two grid wires obviously is greater than that for the conditions illustrated in Fig. 11, and for the conditions illustrated in Fig. 11B, this range is smaller than that for the relations illustrated in Figs. 11 and 11A.

From this standpoint, therefore, it is advantageous that the feedback potential be large, specifically several times as great as the deflecting potential increment necessary to displace the beam from one grid wire to the next in the absence of feedback. Although, as has been noted heretofore, for such large amplitude of feedback potential the possibility of ambiguity in coding is present in the case where two directional displacement of the beam in response to signal pulses applied to the deflector plates 20 is utilized, such ambiguity may be avoided by effectively blanking the beam or by opening the feedback circuit until the potential across the deflector plates 20 is raised or lowered by an input signal pulse to the value requisite to deflect the beam to the code position corresponding to the amplitude of the signal pulse. For example, referring to Fig. 11A, if the signal voltage is E9 correspondin to the beam position 09, and the beam is blanked or the feedback circuit effectively opened until the voltage reaches thisvalue and the beam is then turned on or the feedback circuit closed, the beam will be held in a stable position to produce the code pulse group n+2.

A further important function of the electrode or grid 25, which leads to its designation as a quantizing electrode or grid, is to be noted. As is apparent, there is a range of efiective deflecting potentials between the deflector electrodes 20 corresponding to a beam position such that it passes between a pair of grid wires. For example, referring to Fig. 11, it will be seen that for all input signals of values such that the beam passes between

grid wires

441 and 442, the beam will reach a stable position between 01 and 02, remain there during the sweeping cycle and result in the corresponding code pulse group. Stated in another way, all input pulses effective to deflect the beam to a position between

wires

441 and 442 result in the code pulse group corresponding to this position. Similarly, all input pulses effective to deflect the beam to a position between 02 and 05 result in the code pulse group associated with the opening between the

wires

442 and 443.

Thus, the quantizing electrode or grid in effect, divides the range of input signals into a multiplicity of discrete bands each of which corresponds to a respective code pulse group.

The electrode or grid 25, therefore, performs the dual functions of quantizing the input signal and of holding the beam in the proper code position during each sweeping cycle.

As has been noted heretofore, advantageously the feedback potential is relatively large. The provision of the electrode or grid 25 having a secondary electron emission coefficient greater than unity facilitates attainment of this desideratum in that it not only provides an initially greater feedback current but also reduces the amplification requisite to produce a given feedback potential, and thus simplifies the amplifier design.

Additionally, the provision of such a grid has other advantages. Secondary emission from the coding electrode 26 cannot be entirely eliminated, at least not readily. However, in the case where the electrode 25 has a secondary emission coefiicient greater than unity, the secondary current from the coding electrode to the collector electrode 24 is but a very small fraction, in any event, of the total current to the collector electrode so that the feedback potential is accurately determined by and representative of the beam position.

Further, the provision of such a grid enables a substantial reduction in the size of the device and particularly of the quantizing electrode or grid for any desired number of code groups,

whereby very closely spaced grid -wires'maybe employed and high resolution of input signals Additionally, the collector electrode 24 has a small capacitance to the other electrodes whereby fast quantizing action is'obtained.

Although, a specific embodiment of this invention has been shownand described, it willbe understood that it-is but illustrative and that various modifications may be made therein without departing from'the scope and spirit of this invention asdefined in theappended claims.

What is claimed is:

1, An electron discharge device comprising a member having a plurality of rows of apertures therein, means opposite thereto for projecting an electron beam toward said member, means for deflecting said beam in the direction along one of said rows, means for deflecting said beam in a direction different from said first direction, and means for guiding said beam along a path of preassigned configuration while it is deflected by said second deflecting means, said guiding means comprisin an auxiliary electrode between said beam projecting means and said member and having therein apertures conforming to said preassigned configuration, said auxiliary electrode having a coeflicient of secondary electron emission greater than unity, a collector electrode opposite said auxiliary electrode and a feedback coupling between said collector electrode and said first deflecting means.

2. An electron discharge device comprising a member having a plurality of rectilinear rows of apertures therein, means opposite thereto for projecting an electron beam to said member, means for deflecting said beam in the direction parallel to one of said rows, means for sweeping said beam across said member at an angle to said direction, and means for holding said beam to a substantially rectilinear path while it is deflected by said sweeping means, said holding means com prising an auxiliary electrode between said beam projecting means and said member, extending across the beam paths therebetween and having a plurality of restilinear openings therein, the face of said auxiliary electrode toward said beam projecting means having a coefficient of secondary emission greater than unity, a collector electrode opposite said auxiliary electrode and a feedback coupling between said collector electrode and said deflecting means.

3. An electron discharge device comprising a member having a plurality of parallel rows of targets therein, means opposite thereto for projecting an electron beam to said member, means for deflecting said beam in the direction parallel to said rows, means for sweeping said beam across said member in the direction at right angles to said rows, an auxiliary electrode between said beam projecting means and said member and comprising a plurality of parallel elements extending normal to said rows and in the direction of the beam sweep, said elements having a coefficient of secondary emission greater than unity, and a collector electrode opposite said auxiliary electrode.

4. An electron discharge device comprising a member having a plurality of adjacent rows of targets therein, means opposite thereto for projecting an electron beam to said member, a first deflecting means for deflecting said beam in the m directionparallel to one of said rows, a'second deflecting means for deflecting said'beam in a direction'across said'rows, an auxiliary electrode between said beam projecting means and said 5 memberand having therein a plurality of openings extending transversely of the beam paths between saidbeam projecting means and said member, there being one opening for each target in said one row and in alignment therewith, said 5 auxiliary electrode having a coefficient of secondary emission greater than unity, and a collector electrode-adjacent said auxiliary electrode.

5. An electron discharge device comprising a member having a plurality of substantially parallel rows of targets therein, an auxiliary electrodeiopposite one face of said member and having therein a plurality of parallel rectilinear openings extending across said rows, there being one opening in alignment with each target in one of said rows,said auxiliary electrode having a coeflicient of secondary electron emission greater than unity, a collector electrode opposite said auxiliary electrode, a first deflection means for deflecting said beam in the direction parallel to said rows, and a second deflection means for deflecting said beam in the direction parallel to said openings.

6. An electrode for electron discharge devices, comprising a foundation member having an aperture therein and having also a groove therein to one side of said aperture, a plurality of wires extending across said aperture and groove and affixed to said foundation member beyond said groove, and means for applying tension to said wires comprising a member overlying said groove and in engagement with said wires and means urging said second member toward said groove.

'7. An electrode for electron discharge devices, comprising a foundation member having an aperture therein and having also a groove therein to one side of said aperture, a plurality of wires extending across said aperture and groove and afiixed to said foundation member beyond said groove, a rigid rod overlying said wires and in alignment with said groove, and resilient means mounted by said foundation member and bearing against said rod to force it toward said groove.

8. An electrode for electron discharge devices, comprising a foundation member having an aperture therein and having also a groove therein to one side of said aperture, a plurality of wires extending across said aperture and groove and affixed to said foundation member beyond said groove, a rigid wire overlying said groove and said wires and engaging said Wires, and a resilient strip substantially coextensive with said rigid wire, secured to said foundation member and bearing against said rigid wire to face it toward said groove.

9. An electrode for electron discharge devices, comprising a foundation plate having an aperture therein and having raised portions on opposite sides of said aperture, both of said raised portions being provided with a plurality of trans verse parallel slots and one of said portions being provided with a longitudinal groove, a plurality of wires extending across said aperture, seated in said slots and secured adjacent their ends to said plate, a rigid wire overlying said wires and groove and in alignment with said groove, and a resilient strip mounted by said plate, substantially coextensive with said rigid wire and bearing against said rigid wire to urge it into said groove.

10. An electron discharge device comprising an enclosing vessel, an electron gun mounted at one end of the enclosing vessel, a unitary assembly supported adjacent the other end of said vessel, said assembly comprising a coding plate having a plurality of rows of apertures therein and having also aligning openings therein, an auxiliary electrode opposite said plate and having openings therein in alignment with said rows of apertures, said auxiliary electrode comprising a foundation member having aligning openings therein one opposite each of said openings in said plate and rigid insulating rods each fitted in respective aligning openings in said plate and said foundation member, and deflecting means between said gun and said assembly.

11. An electron discharge device comprising an enclosing vessel having two opposed cylindrical portions sealed together at their opposing edges, an electron gun mounted in one of said portions, deflector means including a pair of parallel deflector plates supported in said one portion, and a unitary assembly mounted within the other of said portions, said assembly comprising a coding electrode having a plurality of parallel rows of apertures therein, a grid including a plurality of parallel wires extending over one face of said coding electrode and at right angles to said rows, said wires being parallel to said pair of deflector plates, and means joined thereto fixing said grid in relation to said coding electrode.

RAYMOND W. SEARS.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 2,053,268 Davis Sept. 8, 1936 2,144,337 Koch a- Jan. 17, 1939