US2200063A - Space discharge apparatus and circuits - Google Patents

Description

KSheets-Sheet 1 EL L'cmb/v GROUPS v v v v I AAA AAA VI Ivv R. A. HEISING' SPACE DISCHARGE APPARATUS AND CIRCUITS 0 Filed July 51, 1956 May 7, 1940.

FIG. 4

A TTORNE V R m M w R. A. HE/S/NG I FOCUSS/NG VOLTAGE May 7, 1940.

R. A. HEISING SPACE DISCHARGE APPARATUS AND CIRCUITS Filed July 31, 1936 FIG.7

2 Sheets-She et 2 ATTORNEY 'Patented May 7, 1940 UNITED STATES srAca DISCHARGE APPARATUS Ann cnwurrs Raymond A. Heising, Summit, N. .L, asslgnor to Bell Telephone Laboratories, Incorporated, New York, N. Y., a corporation of New York Application July 31, 1936, Serial No. 93,546

9 Claims.

The present invention relates to space discharge apparatus involving secondary electron emission, preferably a succession of times to secure large amplification, and to circuits for such apparatus. This general type of apparatus is referred to in the art as an electron multiplier, and will be so termed in this application.

. A general object isto improve upon the structure and performance of apparatus or systems 10 using secondary electron emission or systems using electron multipliers.

The invention is particularly, but not exc1usively, directed to the use of electron multiplier apparatus at ultra-high frequencies, especially at frequencies so high that at those frequencies it is diflicult or impossible for the usual grid-controlled vacuum tube to operate effectively to give high amplification or high output power.

High amplification at very high frequencies is difficult of attainment with the usual type of vacuum tube on account of the inherent capacities between elements of the tube or circuit or capacities to ground from the various elements.

A limitation on the output power at very high frequencies is placed by the difliculty that the spacings between grid, filament and plate are so small that the plate itself tends to be small and therefore to have but slight dissipative capacity. These dimculties are overcome to a considerable extent by the invention, according .to which an electron multiplier structure is combined with suitable input and output wave structures to enable high amplification-of waves of high frequency and, if desired, at high output power. The input structure may comprise a cathode and grid similar to those used in ordinary vacuum tubes, with the grid preferably located very close to the cathode as inthe case of the short wave tube disclosed in my prior application Serial No. 64,697 filed disclosed in that application. In the device of the invention, only a relatively small number of electrons need be passed from the cathode through the grid as compared with grid-controlled vacuum tubes of usual eonstruction, for the electrons instead of constituting the plate current of the tube, enter the electron multiplier structure and produce an eventual output current which may be 'many times larger than the current passed through the grid. This fact is taken advantage of in the design of the apparatus to secure th e necessary control of the initial electron stream by the grid at all frequencies while at the same time keeping the active grid losses small. I

- of the device of Fig. 2;

February 19, 1936 and for the same reason as Another condition favorable to the :use ofan electron multiplier in a system of high frequency amplification is that the amplification takes place in the successive stages without the necessity of imposing a high frequency potential between ele- 5 be relatively large without adversely affecting the operation even at high frequencies.

While the invention, as stated, is especially advantageous at very high frequencies,- it is ef-' fective at low frequencies also and certain fea- 15' tures of the invention comprise methods and means of wave transmission or reception, and of generating, modulating and controlling waves which are capable of general application without regard to frequency range or type of system. 20

The various features and objects of the inventlon will be more apparent from the following detailed description of typical embodiments as illustrated in the accompanying drawings.

- In the drawings,

Fig. 1 is a schematic sketch of one embodiment;

Fig. 2 shows more in detail a. tube construction embodying the features shown schematically in Fig. 1;

Fig. 3 is a functional diagram toillustrate the 30 method of operation of the device of Fig. 2;

Fig. 4 shows curves illustrative of the operation of the apparatus of the invention;

Figs. 5 and 6 show typical circuit applications Fig. '7 is a circuit diagram of one method of modulating or controlling transmission through an electron multiplier;

Figs. 8 and 9 are fragmentary diagrams illustrating modified types of apparatus for modulat- 40 ing or controlling the transmission; and

Fig. 8A is a detail plan view of an alternative construction of emitting plate.

In Fig. 1, an evacuated tube 1|0 has an input circuit at H and an output circuit at II. The tube includes a hot cathode or other suitable source of electrons l3, and a control grid I4, these two elements being constructed and arranged, for

" example, as in; the. case of an ordinary vacuum erably has sufiicient voltage to keep the current figures. This multiplier comprises a series of secondary emitter plates SE and an associated system of deflecting plates P, supplied with increasingly positive potential from grid toward anode, as by taps to appropriate points along a potentiometer l6 supplied from battery IT. A screen I8 is interposed between the last emitter and plate and the anode. Bias battery l9 pref- 2| assumed in contact with terminal 22, the input'waves from H vary the potential of the grid ll relative to cathode l3 and control electron flow toward the first emitter and plate, SE

and P, which, as stated, are positive toward the cathode. 'By means of a magnetic field (normal to the plane of the drawings) the'electrons are constrained to follow a curved path somewhat as indicated by the arrows and to' impinge on the first of the secondary emitters SE, from which they drive 'out a muchlarger number, of secondary electrons. It is assumed that the plates SE are each coated with a suitable substance to make them eflicient emitters of secondary elec- The secondary electrons thus liberated trons. from the first emitter are in similar manner caused to strike the second emitter and drive out a still larger number of secondariesand so on, until finally after a sufilcient number of multiplier stages have been traversed, the resultant large stream of electrons passes through the screen l8 to the anode l5 and sets up current flow in output circuit l2. This, eventual out put current, while under control of the grid, is many times larger than the current passed by the grid. This fact makes it possible to secure large outputs by means of an input wave construction (grid and cathode) which at the frequencies utilized might be capable of transmitting only very small current when used in an ordinary type of vacuum tube amplifier.

By throwing swltch2l to connect to terminal 23, feedback coil 25 coupled to the output tuned circuit 26, is included in the grid circuit in series with leak resistor and condenser combination 21. In this condition the circuit is a regenerative amplifier or generator of sustained oscillations for supplying'a load connected'to circuit 12. In this case no input waves at H are necessary.

In this and the subsequent figures, batteries such as H and 20 would in many cases be replaced by other' voltage sources such as direct current generators or rectifiers with suitable filters as is common in the vacuum tube art. It is preferable to separate the supply sources as at I1 and 20 instead of having one source with potentiometer l6 extending across its terminals, for the reason that the current drawn by the emitters SE increases along the series so'that relatively large current needs to be supplied to the anode I5 and somewhat smaller current to the last emitter, etc. Large losses would result in a potentiometer resistor across power source 20, and, moreover, since the current through the potentiometer mightvary in large amount during signaling, undesirable voltage variations mightjbe produced cn-the various plates. The

subdivision of the power supply may be carried further, as indicated for example in Fig. 2.

One form which the physical arrangement of the parts, indicated diagrammatically in Fig. 1, may take is shown in Fig. 2. The cathode I3 is shown as a single straight wire extending perpendicular to the plane of the paper and sur-' rounded by an inner grid M, which is the control grid, and an outer grid 30 maintained more negative than the control grid. There'is a positive field on plate P1 and secondary emitter SE1 tending to draw electrons through the :two grids. Since no other field exists on other sides of the cathode-grid structure, most of the electrons are directed toward the point where they are to be other figures a magnetic field is used perpendi- .cular to the plane ofthe paper to cause the electron path to be curved and directed against the secondary emitters at the proper angle. InFig. 2, the shield 18 is shown as effectively enclosing the anode l5.

As indicated in connection with Fig. 1, such a high negative grid bias is preferably used as to break up the electron stream into a succession of electron groups or clouds somewhat as shown diagrammatically'in Fig. 3. In this figure the groups -of electrons are indicated by small dots crowded together and occurring one each cycle, by way of example. The numbers in a, group are shown increasing toward the right without attempting to depict the actual rate of increase. .The time required foran electron leaving the cathode to reachthe' anode IE (or more exactly to cause its corresponding successor electrons to arrive at the anode l5) may be a large number of cycles of the impressed waves. rive at the screen l8 during that part of the alternating cycle at which,' after passing through the screen, they will deliver power to the load circuit when they move across to-the plate l5.

(This feature of my invention is claimed in my copending application Serial No. 280,702, filed June 23, 1939.) Several cycles of time may elapse between the instant when an electron leaves the cathode-grid region and the instant of arrival of the corresponding electron group at the screenanode region without interfering with the power delivering properties if the time and frequency relations are properly observed. The proper time relations can be determined by trial by varying the voltages applied to the tube elements and observing the maximum output for given input.

The time of travel is of first order importance between cathode and grid l4 and also between screen l8 and anode l5.' The grid It should be near the cathode to effectively control the electrons without too great an active grid loss. The spacing between screen and anode must not be The electrons should artoo large since this part of the electron stream delivers power to the load circuit. However, the spacing of screen and anode need not be as small as in the ordinary three-electrode tube for the reason that in the tube of the invention the elecgion in the ordinary vacuum tube where they 7 utilize a large part of a cycle in attaining their speed in passing the'grid.

In Fig. 3, a number of capacities are shown between the various emitter plates and ground.

Large capacities at these points are not only tolerated but are desired in order to prevent variations in high frequency potential from occurring between shields, secondary emitters, plates, etc. The only points at which it'is desirable to keep the capacity between elements as small as possible is between grid and cathode, and between screen 18 and anode l5. Low capacity between cathode and grid is desirable since potential variations must occur between these elements in order to control the electrons passing to the multiplier. Low capacity between shield and anode is desirable since this part of the circuit acts like the plate impedance of a tube in delivering power to a load, and it is important not to have. a material amount of shunting capacity. In Fig. 3 a ground plate 3|. is shown extending from the screen I8 in such manner as to shield the secondary emitters SE1. etc. and plates P1, etc. from the output circuit. This plate 3| may have inherent capacity as indicated at 32, 33-39, with respect to the secondary emitters and plates to prevent potential fluctuations on these elements. These capacities may be partly or wholly inherent or may consist of condensers connected at the points shown. The shield 3| may be tubular and surround the electron multiplier apparatus completely. One such capacity 39 is shown between the cathode and the ground plate. If desired the capacity at this point can .be placed between the control grid and the ground plate in which case the signal waves vary the cathode potential with respect to ground, necessitating use in the cathode heating circuit of some suitabledirect current filter, (which may be'of usual construction) to prevent shunting off the signal through the current supply circuit.

In Fig. 4 the curve A givesthe general shape of the characteristic of the device of the previous figures when used as an amplifier. As a generator of ultra-high frequency waves the characteristic has more nearly the shape of curve- B or C. With relatively few emitters or multiplier stages, the curve takes the form of characteristic B. As the number of stages is increased the curve has some such form as that of C. The reason for this is that the timing becomes important, as well as the voltage. In order to produce maximum output current not .only must the yoltage be such as to cause the electrons to travel from emitter to emitter in proper paths but the groups of electrons must eventually arrive at the screen l8 and anode IS in proper instants of time or, in other words, must have the right phase relation to the oscillations being generated. Where the number of stages is great there may be several values of applied voltage at. which the timing is such as to produce a current maximum, as shown by curve C.

The remainder of the figures disclose methods of utilizing the device of the invention in signaling circuits, particuiarly'in modulating and detecting circuits. g

In Fig. 5 the tube III may have the same construction as in Fig. 2 and this construction is indicated in outline. A radio frequency wave is produced at 40 by suitable means and is applied through the tuned coupling 4|, 42 to the grid circuit in* series with the speech input coupling shown at 44 leading from microphone 43. Condenser'45 affords a by-pass for the high frequenthe anode of a vacuum tube in drawing electrons through the grid. The modulated current is amplified to the desired extent by the electron multiplier structure as described in connection with Figs. 1 and 2. The output current is shown as being led to tuned output circuit 26 which is coupled to any suitable transmitting circuit such as an antenna or line.

In Fig. 6. a receiving circuit orantenna 46 is coupled through suitable tuned input circuit 41 to the grid'of tube 10, the grid circuit of which includes a grid leak and condenser combination 48. Detection occurs as in the ordinary grid leak detector and the resulting detected or low frequency current passes into the electron multiplier apparatus where it is amplified to the desired extent. The output is transmitted through transformer 49 to loud-speaker 50 or other utilization circuit.

In Fig. 7 a high frequencywave from source 40 is applied to the grid of tube In through tuned input coupling 4|, 42 and the resulting high frequency wave is amplified in the electron multiplier portion of the tube. Modulation is effected by applying the speech or other low frequency waves originating in microphone 43 through transformer 44, between two of the emitter plates such as SE4 and SE5 and their associated deflectingplates P4 and P5. Emitter plate SE5 is made in two parts 5| and 52, such that only the part 5| is eflicient in emitting secondary electrons, the part 52 being coated with carbon or otherwise rendered non-emissive. It may be constructed as a depressed portion into which about half of the electrons from the previous emitter fall when there are no signal voltages applied. The signal voltages vary the focusing voltage which determines the point at which the electrons from plate SE4 fall on plate SE5 and cause a variable portion of these to fall on the active part Slof plate SE5.

' Thus, the quantity of secondaries emitted from from thedrawings and from the foregoing description that the voltages between the emitter.

plates to the left and to the right of the modulat ing plates remain relatively fixed.

If desired, the low frequency waves can be applied to the grid and the high frequency waves to be modulated can be introduced between a pair of emitter plates this representing the converse relationship of that shown in Fig. '7.

A modified type of controlling means is shown in Fig. 8 in which the high frequency or low frequency control or modulating wave is impressed on one emitter SE3 and plate Pa. (The successive electron streams are differently hatched in the drawings to distinguish them apart.) Emitter SE3 may be constructed as in Fig. 7. Emitter SE: is non-emissive except fora small central strip which maybe pictured as extending across the tube in the direction perpendicular to the plane of the paper. emitted from this active strip are under normal conditions directed so that half of them fall on The secondaries the active part of emitter sm, the other halt 7 these waves.

obtained between plates SE2 and SE3. The modulating waves applied to plate SE3 cause a greater or less portion of this electron beam to impinge on the active portion of plate SE3. A modulated electron beam appears therefore from plate SE: onward throughout the rest of the multiplier lying to the right in this figure. This method of modulation is adapted for use where the modulating wave is either of low or high frequency. Examples of use with a high frequency modulating wave would be the 'case where the source at 54 is a beating oscillator for heterodyning or for modulating, detecting or amplifying purposes.-

Instead of rendering parts of a secondary plate non-emissive by coating it with carbon or other suitable substance, the part to be rendered nonemissive can be covered with a screen consisting preferably of wires running parallel to the direction of general electron motion (parallelto the paper in Fig. 8). Such construction is illustrated in Fig. 8A which is a plan view of plate SE2 of Fig. 8 with the alternative construction just described. The screens are shown at 55 and 56. '-The portions of the plate covered by the screens need not be depressed as much as in Fig. 8. The potential on the screens 55 and 56 may be the same as that of the emitter plate SE2 which they shield.

An alternative modulating method is illustrated in Fig. 9 in which a pair of metal electrodes 51 and 58 extend across the path of the electron beam and have impressed between them the modulating potential, from input-54.

These variations in potential applied to electrodes 51 and 58 shift the electron beam back and forth so that a variable part of it strikes the active portion of emitter plate SE3 producing a moduflated beam of secondaries to be emitted from SE3. The modulating wave may have low or high frequency.

In each of the Figsfl'? to 9, for ease in illustration the electron current has been represented as substantially continuous, but it will be understood that it could consist of more or less separated groups of electrons as described in connection with Figs. 2 and 3 by giving the grid a sufficiently high negative bias.

Uniform spacing of the emitter plates is indicated in thevarious figures. This is not a necessary condition, however. The path taken by the electrons from one plate to the next is determined by the voltage difference between adjacent plates and by the magnetic field. It is generally more convenient to make the magnetic field substantially constant and uniform throughout the tube. When this is the case, the spacing between any two emitter plates and the voltage difference between them must correspond with each other so as to make the electrons strike the emitterplates in the most effective manner. In general where large output power is desired, a high voltage must be used on the output anode, in which case the path taken by the electrons from the last emitter plate to the anode may be of difi'erent shape from that taken between two emitter plates. The screen grid prevents the relatively high voltage on the anode from affecting the electron travel between the secondary emitter plates.

The claims are not to'be construed as limited to the details of construction that have been disclosed herein as illustrative examples, since many modification and alternative constructions will follow readily from the types that have been disclosed. I

What is claimed is: j 1. In combination, an electron multiplier comprising an evacuated enclosure containing a succession of secondary electron emitters, means to supply successively higher potentials to said emitters, means to release electrons from the first emitter of the succession for impinging on the next emitter, and so on, an-output electrode leading to a utilization circuit and means to control the electron current transmitted through the multiplier comprising a circuit 'forimpressing a variable control voltage between two adjacent emitters in the succession.

2. In combination, an-electronmultiplier comprising an evacuated enclosure containing a succession of secondary electron emitters, means to supply successively higher potentials to said emitters, means to release electrons from the first emitter of the succession for impinging on the next emitter, and so on an output electrode leading to a utilization circuit and means to conmultiplier comprising a circuit for impressing a variable control voltage on one of said emitters, the surface of the emitter on which the control voltage is impressed comprising an emissive portion and an non-emissiveportion on the side upon which the incident electrons impinge.

3. In an electron discharge device, a member having a surface a portion of which is emissive 'trol the electron current transmitted through the a of secondary electrons and a; portion of which is non-emissive, means including a preliminary emitter of secondary electrons for producing a stream of secondary electrons, means to direct said stream of secondary electrons against said surface to cause emission of secondary electrons from the-emissive portion, and means adjacent the path of said stream and controlled by a variable voltage for deflecting the stream so thata variable portion of it is caused to strike the emissive portion of said surface.

4. A combination according to" claim 3 including means to vary tlie intensity of said stream of secondary electrons in accordance with an applied wave.

. 5. In electron discharge apparatus, a plurality of secondary emitter plates, means to cause electrons to impinge on said plates in succession'and to emit secondary electrons from each plate, the

surface of each of a plurality of said plates having only a restricted area that is emissive of secondary electrons, means directing the electron beam from the first of said last-mentionel plates tofall partly on and partly off the emissive area of another of 'said last-mentioned plates, and means to vary the-portion of said beam falling on the emissive area of said other plate.

6. In an electron multiplying device, a secondary electron emitter, means to direct primary electrons against said emitter to cause secondary electron emission current therefrom, a member having a surface a portion of which is a good emitter of secondary electrons and a portion of said surface to produce secondary emission from the emissive portion thereof, and means to deflect variably the secondary emission current incident upon said surface so that a variable portion of.it strikes the-emitting portion of said surface.

7. An electron multiplying device according to claim 6 in which means are provided to control the intensity of the primary electron stream in accordance with an impressed wave to, in turn, vary the strength of the secondary emission current given off from said secondary electron emitter and incident upon said surface, and in which said means to deflect variably the secondary emission current is under control of another wave. 1 1

8. Electron discharge device according to claim 3 in which said means adjacent said path comprises electrodes between which the electron beam passes, and a circuit for impressing a variable a high frequency source is used 'in combination with means for varying the intensity of the electron current thereby.

, RAYMOND A. HEISING.

Images