US2294782A - Thermionic valve apparatus - Google Patents

Description

p 1942- s. a. JACOBSEN 2,294,782

THERMIONIC VALVE APPARATUS Filed Jan. 25, 1941 3 Sheets-Sheet l Fig. 1. E'mifler of .iegondary E led/ons- 'L Th Emff Zer of .iecondary E/ecfrons vvwwrop Patented Sept. 1, 1942 UNETED STATES TNT OFFICE THERMIONIC VALVE APPARATUS Application January 23, 1941, Serial No. 375,594 In Great Britain January 30, 1940 (C1. ZED-27) '7 Claims.

This invention relates to thermionic valve systems and one of its objects is to provide thermionic valve systems having a higher mutual conductance than can normally be obtained.

According to the invention, there is provided a thermionic valve system comprising a valve having at least a cathode, an anode and a control grid, in which an additional electrode adapted to give secondary emission eiiect is provided in such position in the primary electron stream and is provided with such direct current potential relative to the other electrodes that on the average the number of secondary electrons emitted from it exceeds the number of primary electrons striking it, and in which an alternating current impedance is connected in the supply lead to said additional electrode, the voltage developed across such impedance being employed (a) for direct positive feedback or feedback to a further additional electrode located in av position in the electron stream where the electron velocity is small, and (b) for neutralising the inherent input impedance of the valve and/or for neutralising unwanted'impedances in a portion of the circuit preceding the valve.

Such systems may provide mutual conductance higher than can normally be obtained. They generally add to known valve structures:

(1) An electrode placed at a point where the electron velocity is high and which is adapted to produce secondary electrons when bombarded with primary electrons. This electrode should be at a potential which is sufiiciently high to ensure that a reasonable number of primary electrons are absorbed but on the other hand is sufiiciently lower than the potential of an adjacent electrode to allow this latter to absorb the secondary electrons emitted by the added electrode.

(2) An electrode placed where the electron velocity is low and adapted to influence the electron velocity. This electrode is coupled with the first mentioned added electrode and both are coupled to the cathode via a resistance.

A resistance may be inserted between the cathode of the valve and the external circuit; and a small condenser may be connected from the first mentioned additional electrode to the normal control grid of the Valve.

Other objects and aspects of the invention will be apparent from the following description and claims.

Fig. 1 shows a form of valve system referred to above with positive feed-back from the secondary electron emitter to an auxiliaiy grid and neg- 5.6

ative feedback to the auxiliary grid and the con trol grid;

Fig. 2 shows a modification with additional electrodes and with the positions of the control grid and auxiliary grid interchanged;

Fig. 3 shows a modification also with addition al electrodes but with the control grid and the auxiliary grid coplanar;

Fig. 4. shows a simplified system with the secondary electron emitter serving as the auxiliary grid to which positive feedback voltage is applied;

Fig. 5 shows a modification in which the auxiliary grid is omitted and the positive feedback is from the secondary electron emitter to the control grid; 1

Fig. 6 shows a modification also with the positive feedback from the secondary electron emitter to the control grid but with the input applied to the secondary electron emitter as well as to the control grid; and

Fig. 7 shows a modification having feedback from the secondary electron emitter to the control grid as well as to the auxiliary grid.

Fig. 1 shows a valve system according to the invention incorporating a valve V.

The cathode I may be of any well known type including photo-electric cathodes; if necessary it will be heated by a heater not shown. The auxiliary electrode 2 is similar to the control grid 3. By means of the grid leaks 8 and I! and cathode lead resistor l9 both are kept at an average potential somewhat below the cathode potential.

The electrode 4 is kept at a high potential and serves to accelerate the electrons. This electrode is sufficiently open to absorb only a moderate proportion of the electrons and is tied to earth potential at working frequencies by means of the condenser M.

The electrode 5 is kept at an average potential definitely below the potential of 4, the potential reduction being efiected by the resistances In and I2. This electrode is connected to earth potential for alternating current via the resistance 9 and condenser IS. The valve further contains an anode 6, the mean potential of which is derived from battery 1 via the primary winding 20 of an output transformer.

The electrode 5 is coupled to 2 by means of the condenser I5. The resistance l9 of value R1 in the cathode circuit provides a current-dependent negative feedback, and it also provides a D. C. potential for the electrodes 2 and 3 via the resistances 8 and I1 respectively. These resistances 8 and I! will not always be connected as shown but will be connected to whatever potential is suitable for the particular system.

The potential on electrode 2 may, in some cases, require to be increased up to or beyond the point at which grid current flows, in order that the eiTect of electrodes 2 and 3 shall not be to lower the main space current materially. Also it may be advantageous, for the same reason, to make electrode 2 of coarse pitch, While retaining a fine pitch for control electrode 3. If the secondary emitter electrode is very efiicient, a relatively high voltage will be available on electrode 5 without the use of a very high value of the resistance 9 (R). When this voltage is high the pitch of grid 2 may be coarse: this gives the advantage the electrode 2 does not obstruct the electron stream, and the mutual conductance of grid 3 (before application of the connection via condenser l5) will be higher.

In adjusting the potential of the electrode '5 the aim is to obtain a high current output from 5 and it will generally be found that this will occur when the potential of electrode 5 is 20 or more volts less than that of the electrode 4.

This potential should be measured at 5 itself, using a very high resistance meter, and may also be calculated from the voltage at X and the current from electrode 5. The potential at 5 will be higher than that at X. Consider now the efiect of a small change of potential on the electrode 3, the control grid. An increase in potential will cause an increase in space current and the rate of incidence of electrons on electrode 5 will be increased. Electrode 5 possesses secondary emission properties. It may be specially prepared with this in view but it is quite possible to rely on the secondary emission properties of surfaces found in valves made by well known methods. The secondary emission ratio must in any case exceed the factor I, i. e. each incident electron must on the average free more than one secondary electron.

This being the case, an increase in the number of electrons reaching 5 will cause an increase in the number of secondaries leaving 5 and the balance will be in favour of the secondary electrons, or the change in current of electrode 5 will increase the current already flowing out of electrode 5. Owing to the presence of the resistance 9 the potential of 5 will therefore experience a positive increment. If instead of the original change, an alternating voltage is applied to 3, electrode 5 will also be found to have an alternating potential and moreover this potential will be in phase with the original potential on 3. By means of the condenser l5 the potential is transferred to electrode 2; the resistance 8 being very high has a negligible effect on this potential transfer.

This A. C. potential on electrode 2 will have the same effect as the original potential on 3 and will therefore still further increase the space current component. This condition might at first sight suggest instability but there is a very simple law for the amount of the above effect which may be obtained without instability. Assume a small generator (e) to be connected to 2 as shown in Fig. 1. Assume further that the A. C. current in electrode 5 caused by this voltage is 6777.2,5 where mat is the mutual conductance between electrodes 2 and 5. The voltage developed on 5 will therefore in the absence of condenser l5 be eRm2,.=,. It is assumed that this voltage is small enough not to disturb materially the potential difference between 4 and 5.

Assume now that a voltage vRmz,5 is required across resistance 9 (of value R). The voltage of the electrode 2 must therefore be 21 volts compared to the cathode. The resistance I9 is disregarded for the present, but of these 1) volts, 12Rmz,5 volts are already provided across R so that the value of e necessary to obtain 1) volts will be v(1Rm2,5). The presence of R therefore reduces the voltage required on 2 to obtain a given effect on 5. The increase in sensitivity is and provided this increase is finite the process will be stable. In other words, the voltage fed back from 5 must be just less than the original voltage. It is therefore clear that as far as the electrode 2 is concerned it should be possible to increase the amplification indefinitely. This is confirmed by experiment: values of mutual conductance as high as 4 amperes per volt have been measured. It will be clear that the mutual conductance for electrode 3 to the anode 6 will be increased in the same manner as the mutual conductance between electrodes 2 and 5. Assume this mutual conductance between the former pair of electrodes to be bm3,s in the absence of coupling condenser l5. When this condenser is included the mutual conductance will be am 1Rm and will provide a voltage m -0.13 1R.m

across resistance l9 (R1). The input voltage E required to provide 1; volts between 3 and I will therefore be The A. C. output current for E volts input will be ay; 1 l Rm R mg e The previous condition for instability Rm2,5==1 no longer holds. Bigger values of Rmas may be used without instability, the criterion now being Rmz,5 1+m3,eR1. Preferably Rmas is made approximately equal to unity and the efiective overall mutual conductance then becomes approximately equal to 1/R1 and is independent to a first approximation of small variations in mas and 171,3,6. This independence is most marked if m3,s is large when in other words the effective mutual is not increased very much. As an example, assume a mutual conductance of mA. per volt to be required. R1 then must be 20 ohms. If ms,e=5 mA./volt and mas deviates by i 5% from the ideal value of l/R, the actual conductance will be assuming mas not to change. If however ma changes to the same extent as mas the correspondingfiguresbecome .0953-or .0328. From this approximate calculation it is clear that the supplies to valves of this sort must be kept fairly constant. The resistance R1, however, helps to adjust the space currents to an approximately constant value.

In order to increase this efiect, a further resistance may be connected in series with R1 but shunted by a condenser to render this additional resistance ineffective for A. C. components. If a very high additional resistance is used it may be necessary to connect the grid resistances 8 and I 1 to a point of potential higher than the negative terminal of I. This may be done conveniently by a potential divider connected across 1.

A constant space current helps to stabilise the performance of the valve circuit. The presence of the resistance R1 also helps to reduce the effect of unavoidable inductance in the lead from the cathode to the wire 22.

Another important factor in stabilising the performance of the circuit is the potential difference between the electrodes 4 and 5. The maximum secondary emission effect is obtained for a certain potential difference which depends on the shape of the electrodes. For stable operation it is generally preferable to use a potential on 5 rather higher than the value for optimum secondary emission. The resistances l and I2 which serve as a potential divider can often with advantage be made fairly high so that the resistance between and 22 at zero frequency has a value of several thousands ohms even if the R value required is only quite small. If this is the case the potential of electrode 5 will tend to rise with increased activity in the valve and in rising the secondary emission current will be reduced since the potential was already above the optimum value. The reduced secondary emission will in turn help to make the overall performance more nearly what it was before the increased activity occurred.

Fig. 1 shows also the D. C. potential of electrode 5 as being derived from the potential of 4 rather than directly from battery 1. This helps to keep constant the potential difference between these electrodes. A high value of the resistance II is not always an advantage but will generally be found to make the mutual conductance less dependent on the voltage of battery 1 and of changes in the activity of the cathode I.

The rules stated above will not necessarily hold good for any valve incorporating the principles of the invention but illustrate the type of circuit changes that may be used to make overall gain less dependent on supply voltages, etc.

Fig. 2 shows another valve system in accordance with the invention. This valve V differs from that shown in Fig. 1 in the following points. (1) The positions of the control and auxiliary grids 2 and 3 are interchanged. (2) A further electrode 4a has been added which helps still further to withdraw electrons from the secondary emitter 5 and to screen 5 from the anode A. C. potential and which also has some acceleration effect. (3) A further electrode 23 which is kept at a low potential and repels secondary emission from the anode 6. The condenser I5 and resistance 9 are shown again since they are important details; other battery supplies are not shown but may be arranged in any known manner. The potential of electrode 5 must however be lower than that of electrode 4 and/or 40..

Fig. 3 shows a further embodiment of the invention. This valve V differs from those of Figs. 1 and 2 in that the control grid 3 and auxiliary grid 2 are arranged in a single plane. The electrode 2 is conductively connected to 5, the secondary emitting electrode, and to counteract the high potential on 2 the control electrode 3 is supplied with a negative bias such that the mean of the potentials on 2 and 3 is negative compared to the cathode I. The special advantage of Fig. 3 is that the positive feedback effect from electrode 5 does not depend on a condenser coupling and is therefore effective at all frequencies including zero frequency. In other respects this valve is similar to Figs. 1 and 2. The coupling condenser shown in Figs. 1 and 2 would generally be used rather than the conductive connection shown in Fig. 3 unless the high mutual conductance is required at zero frequency.

Fig. 3 may also be used with a coupling condenser in place of the direct connection between 2 and 5. The potential on 2 and 3 may then be of the same order as in Figs. 1 and 2.

In all valve systems according to the invention, the cathode resistance I9 may consist of a material such as iron or tungsten in the form of a fine filament mounted in a separate evacuated envelope or in the main envelope 8. When the supply voltage rises and more current flows through the resistance [9 the temperature of the wire will rise and the resistance increase, whereby the overall mutual conductance will be decreased and tend to compensate for the increase in mutual conductance caused by increased supply voltages.

Fig. 4 shows a simple method of obtaining effects similar to those obtained with the previous arrangement. The reference characters are the same as in Figs. 1, 2 and 3, when applied to similar parts. The secondary emitter 5 is not connected back to any special auxiliary grid but itself serves as an auxiliary grid. To increase this effect 5 should be rather closer to the grid 3 than it would be in the valves of Figs. 1, 2 and 3.

The positions of the accelerating electrode 4 and the secondary emitter 5 have been interchanged in order that 5 shall have more control of the electron velocity. The main advantage of the valve of Fig. 4 is that it is simple to manufacture. Electrode 5 in the arrangement of Fig. 4 must have a considerably higher A. C. potential and there is a tendency to non-linearity of response since the potential difference between 4 and 5 will be less constant than in the other arrangements. Unless 5 has a high A. C. voltage the positive feed back eifect will be small owing to the velocity of the electrons at 5 being much higher than the electrons near 3 in Figs. 1, 2 and 3.

Fig. 5 shows a simplified method of carrying outthe invention. This method is suitable mainly for low frequency work. The current from the secondary emitter 5 produces an in--phase voltage across resistance 9 (of value R) and this voltage is applied to the control grid 3 via secondary of transformer 24, the primary of which forms the input circuit to the valve V and is terminated in resistance 25. The potential transfer from 5 gives a positive feedback efiect as in Figs. 1, 2 and 3. The effective input impedance of the valve V is reduced by the positive feedback and this is a disadvantage at high frequencies.

Fig. 6 shows a still further modification. The positive feedback is applied via condenser l5 to the control grid 3 but the input is applied between 22 and 5 and also via l5 to 3. The input impedance, 1. e. the impedance into 5 and 22, is higher than the resistance 9 since the A. C. current for the secondary emitting electrode 5 produced by a voltage on 3 is in such a direction that it helps the input potential. Capacity between 3 and 22 (or I) which would normally constitute input capacity is similarly reduced by the effect of the A. C. current for electrode 5. The valve V' on the left is shown only by way of example and this valve may also be one incorporating the invention.

Although Fig. 6 shows a principal control grid the method would apply equally to the valve shown in Fig. 3. The input would be applied to 5 and 2 which are joined together, while 3 would be connected to a suitably high negative potential or else 3 may be used for an independent input or for automatic gain control by change of the potential applied.

The valves shown in Figs. 1, 2, 3 and 6 may conveniently be used for generation of oscillations. To this end a parallel resonant circuit is connected across the resistance 9 which must now be of a higher value than before. Similarly in Fig. 5, by removing the input transformer 24 and connecting 5 to the left hand terminal of condenser I 6, and further by connecting a parallel resonant circuit across 9 a circuit is obtained suitable for generation of oscillations.

In all the circuits shown resistance 9 may be replaced by a parallel resonant circuit shunted by a resistance for the purpose of using the valve as a selective amplifier. The positive feedback effect will be frequency dependent and the circuit will show a much higher selectivity than would be obtained from the tuned circuit used with an ordinary amplifying valve.

An important feature of the invention will now be described in connection with Fig. '7, but which applies equally to Figs. 1, 2, 3 and 4. In order to increase the effective input impedance of the valve (impedance into 3 and 22), a network 24 will be connected between electrodes 5 and 3. If the input impedance is mainly capacitive (i. e. low frequency amplifiers) the said network will consist of a capacity only, or if the input impedance has a considerable resistance component the network 24 will consist of a resistance and capacity (e. g. high frequency amplifiers). Other networks will be used for instance when negative impedance effects are required.

The effects obtained rely on the fact that the A. C. potentials on electrodes 3 and. 5 are inphase and that the A. C. potential on 5 is higher than that on 3. To ensure that the A. C. potential on 5 shall be the higher the resistance 28 is used. Resistance 9 is used as before to develop the positive feedback voltage. The voltage for the network connected between 5 and 3 is developed across resistances 9 and 23 and may be fixed independently of the voltage required on electrode 3 by choice of the resistance 23. It is preferred that the A. C. voltage on 5 shall be double or more than double the A. C. voltage on 3. If in a specific case it is impossible to attain such an A. C. voltage on 5 it would still be possible to use the network 24 but it would have to be of rather low impedance.

In general when used in an amplifier the network 24 and the sum of the resistances 9 and 23 would be such that the input capacity of the valve 8 (between the left-hand terminal of condenser IB and 22) would be negative to an extent suflicient to neutralise the anode capacity of a preceding valve or of an input transformer. It is understood that the network 24 may be used also in the absence of positive feedback effect (e. g. with condenser l5 and electrode 2 omitted).

The invention may be usefully employed in broadcast receivers e. g. in the system of Fig. 6 a tuned aerial input circuit could be applied between 5 and 22 and the grid potential via resistance I! could be used for automatic volume control. This kind of system is particularly useful for automatic volume control work as the mutual conductance decreases very rapidly with increase in grid bias and a large control range can be obtained.

What is claimed is:

1. A thermionic valve system comprising a valve having at least a cathode, an anode and a control grid, an additional electrode adapted to give a secondary emission effect so positioned in the primary electron stream from said cathode and provided with such direct current potential relative to the other electrodes that on the average the number of secondary electrons emitted from it exceeds the number of primary electrons striking it, an alternating current impedance connected between said secondary electron emitting electrode and in series between said secondary electron emitting electrode and said anode and traversed by secondary electron current of said secondary electron emitting electrode, a further additional electrode located in a position in the electron stream from said cathode Where the electron velocity is small, and means for transmitting voltage variations of said secondary electron emitting electrode produced by said secondary electron current to said further additional electrode in such phase as to augment said variations.

2. A thermionic valve system comprising a valve having at least a. cathode, a control grid, a positive accelerator grid, an anode and an additional grid adapted to give secondary emission effect located in the primary electron stream from said cathode in the vicinity of said accelerator grid, means for giving said additional grid such a direct current potential relative to the other electrodes that on the average more secondary electrons are released by it than primary electrons impinge against it, an alternating current impedance connected between said cathode and said additional grid and in series between said additional grid and said accelerator grid and traversed by secondary electron current of said additional grid whereby upon the application of an alternating potential to said control grid an in-phase alternating potential appears across said impedance, an auxiliary control grid located in a position in the electron stream from said cathode at which the electron velocity is small, and means for transmitting said in-phase alternating potential to said auxiliary control grid.

3. A thermionic valve system comprising a valve having a cathode, a control grid, a positive accelerator grid and an anode, an additional grid adapted to give secondary emission effect located in the primary electron stream from said cathode in the vicinity of said accelerator grid, means comprising a supply lead for supplying to said emissive grid a direct current potential lower than that of said accelerator grid, an alternating current impedance connected in said supply lead between said emissive grid and said accelerator grid and traversed by secondary electron current of said emissive grid, a further additional grid located in a position in the primary electron stream from said cathode at which the electron velocity is low, means connecting said emissive grid to said further additional grid, a second impedance, and means connecting said emissive grid through said second impedance to said control grid.

4. The combination with an electron tube having an anode, a cathode and a control grid, of means for increasing the transconductance between said grid and said anode comprising an auxiliary control grid for controlling the electron stream between said cathode and said anode, a secondary electron emitter in said tube, means for producing potential variations of said emitter and said anode in opposite phase, and means for transmitting said potential variations of said emitter to said auxiliary control grid.

5. A wave translating system comprising an electron discharge tube having a cathode, a signal control grid, an anode, a secondary electron emitter and an auxiliary control grid positioned in the electron stream between said cathode and said anode for controlling said stream, a source of positive potentials, means connecting said cathode to the negative pole of said source, a circuit including a voltage dropping resistance connected between said emitter and a point of positive potential on said source, means connecting said anode to a point of higher potential on said source, means for supplying to said signal control grid and said auxiliary control grid biasing voltages negative with respect to said cathode, and means connecting said auxiliary control grid to said circuit at a point between said resistance and said emitter.

6. A circuit comprising an electron discharge tube provided with a source of primaryelectrons, a signal control grid, a source of secondary electrons, an anode, a secondary electron collecting electrode and an auxiliary control grid, a source of positive potentials, means connecting said cathode to the negative pole of said potential source, a connection including a voltage dropping resistance between said secondary electron source and a point of positive potential on said potential source, means connecting said anode I and said collecting electrode to points of higher potential on said potential source and maintaining said collecting electrode substantially at the potential of said first point for alternating currents, a load circuit connected between said anode and said cathode, means for supplying to said signal control grid and said auxiliary control grid biasing voltages negative with respect to said source of primary electrons, and means connecting said auxiliary control grid to a point of said connection between said voltage dropping resistance and said secondary electron source.

7. A wave translating system comprising an electron tube having an anode, a cathode, a signal control grid, an auxiliary control grid positioned in the electron stream between said cathode and said anode for controlling said stream, a secondary electron emitter in said tube, means for producing potential variations of said emitter and said anode in opposite phase with respect to said cathode, and means for transmitting potential variations of said emitter to said auxiliary control grid to produce positive feedback and transmitting potential variations from said anode to said signal control grid to produce negative feedback.

BENT BULOW JACOBSE'N.

Images