US2081127A - Electrical system - Google Patents

Description

\ `Maly 18, 1937. P. J. KoNKLE 2,081,127

ELECTRICAL SYSTEM Filed May 14. 1956 2 Sheena-shamA 1 May 1s, 1937'.

P. J. KONKLE ELECTRICAL SYSTEM Filed May 14, 1936 2 sheets-sheet. 2

arie.

Patented `May I8-, 1937 OFFICE ELECTRICAL SYSTEM Philip J, Konkle, Philadelphia, Pa., assignor to Philco Radio & Television Corporation, Philadelphia, Pa., a corporation of-Delawarc Application May 14, 1936,'Serial No. 79,808

7 Claims.

This invention relates to the control ofthe gain or amplication of variable gain ampliers and, more particularly. to a novel gain control device by means of which an input signal may be ampliied to an extent which is governed by a second electrical signal. The invention is capable of general application and may be used in any instance where it is desired to control the ampliflcation oi a signal in the manner above men- 1'0 tioned. For example, the invention is well adapted for use in television systems, and ilnds several different applications therein. In order to illustrate the invention, a specific example ofits use for one particular purpose in a television system will be given. It will be understood, oi' course, that specic embodiments oi the invention other than those disclosed herein may be used and are, therefore, to be considered within the scope of the invention.

20 One object of the invention is to provide a variable gain amplier, whose gain may befcontrolled by an electrical signal whilepreventing.

such control signal from appearing in the output circuit of the device. 25 Another object of the invention is to provide a variable gain device, whose gain may be controlled without the introduction of a transient voltage or other extraneous voltage in the output circuit of the device, as the-gain thereof is varied.

Another object of the invention is to provide an amplifier having a high impedance input circuit, which when energized by two input signals, will form in its output circuit, an output signal proportional to the difference between th input signals.

l Still another object of the invention is to provide an amplier having one input circuit which may be adapted for push-pull input signals, and another input circuit` for a control signal, the amplifier being so arranged that, first, the control signal is denitely established in such amanner as to control the mutual conduc- 1 tance and thus the gain of the amplifier tubes 45 without introducing degeneration of the input signals and second, the output signal of the amplifier comprises only'the input signal amplified by an amount dependent upon the control signal.

A still further object of the invention is to pro- 55 Other objects `and features of the invention will appear hereinafter as the description .procaeds.

In the accompanying drawings:

Fig. 1 is a circuitdiagram of a device employing-the basic principles of the invention; 5

Fig. l2 shows the characteristic curves of certain tubes which may be used in the device of the invention;

Fig. 3 is a. schematic illustration o! a scanning signal generator employing an embodiment of the invention which may be used to actuate the scanning means in a conventional television system ,wand

Fig. 4 is a circuit diagram illustrating a modication of the invention.

The general principles of the invention maybe understood from a consideration of Fig. 1 wherein the output signal Eo appearing in the output circuit corresponds to the amplified input signal E1, the degree of ampliilcation depending upon a second or auxiliary input signal En. The input stage oi' the device comprises two tubes V1 and V2, which tubes have their cathodes connected in a common cathode circuit including the resistor R1. Tubes V1 and Vz are preferably iden- 25 tical, one withthe other, with respect to their electrical characteristics. The` average bias voltages applied to the grids of these tubes are maintained at corresponding .values with respect to some predetermined voltage level, such 'as 30 l ground, by means of the bias battery C and the grid-leak resistor R. Due to this resistor, the voltage of the grid of tube V1 iluctuates about the predetermined bias voltage -level in response to the signal E1, whereas the grid o1' tube Va 35 is maintained at the bias voltage level, since the signal E1 is not applied to the grid of tube Vz. The cathodes of tubes V1 and V: are maintained at some voltage greater than ground potential, depending upon the voltage drop across the re- 40 sistance R1. Thus, the input signal for tube V2 comprises the bias voltage due to battery C and the voltage drop in the resistor R1, while the input signal for the tube V1 comprises the same 45 voltages plus the alternating component of the signal E1, which causes the grid voltage of tube V1 to iluctuate about themean value of the grid 'voltage of tube Vz.

A portion of the output signal from tube V1 is 50 taken across a variable resistance Rz in the anode circuit of the tube and is transferred to an additional tube V3 by a conventional resistance-ca-v -pacitance coupling circuit. `Output signals from tubes V: and V3 are-added together and the comu y these tubes.

bined output signal is taken across a common resistance R3 connected in the anode circuits of The resistance R2 may be so adjusted that the amplification of any input signal through the branch comprising tubes V1 and Vs is the same as the ampliiication of any input signal through the other branch comprising tube V2 alone. It will be noted, however, that the phase of the signal transferred through the first branch will be opposite to the phase of the signal transferred through the second branch. In y other Words, due to the additional tube V3 in the rst branch, there will be a shiftv or difference in phase of betweenthe signals transferred through the two branches; Consequently, any input signal or voltage variation introduced in both tubes V1 and V2, which signal might, for example, be one introduced across the resistor R1, will cancel in the output circuit. 'Ihat is to say, the amplified signal from tubes V1 and V2 appearing across resistance R3 will be of equal amplitude andopposite phase with respect to the corresponding signal transferred `through the branch comprising tube V2, and the two signals will thus cancel each other. It will be seen, however, that any signal introduced in one of the branches but not in the other branch will be transferred and will appear in the output circuit. Likewise two signals of opposite phase applied to. the input circuits of tubes V1 and V2 would add together in the output circuit. Thus an input signal of opposite phase with respect to signal E1 might be applied to tube V2 and would produce a signal E0 inthe output circuit similar to the output signal obtained by applying the signal E1 to tube V1.

As noted above, the A. C. components of the input signal E1 are applied only to tube V1, and the biasing voltages for the grids of tubes V1 and V2 lare identical. Consequentlyfthe signal E1 will be transferred through the system, but variations in the bias voltage of the tubes, which will appearas input signals of the same phase to both tubes, will cancel in the output circuit. Tubes V1 and V2 may have the characteristic that their mutual conductance, which is a measure of gain or amplification, is a function of the input voltage. lThe amplification of the systemfmay then be varied by varying the voltage drop across the resistance R1 which will vary the-mutual conductance of 'tubes V1 and V2 and, consequently, will vary the degree of amplification of the signal E1 which, as above pointed out, may be transferred through the branch comprising tubes V1 and V3. The voltage drop across the resistance R1 may be conveniently controlled by means of an additional tube V4 whose cathode is connected in common with the cathodes of tubes V1 and V2 and whose anode is connected to the plate supply source B. Thus, the current drain of tube V4. which may be used to control the voltage drop across the resistance R1, may be varied readily by varying the input signal E12 applied to the tube. It will be seen, therefore, that the input signal E1 may-be amplified in the device by an amount dependent upon a second or auxiliary input signal E2 to obtain an output signal Eo similar in wave shape to the input signalE1 and-convoltage inthe input circuits of these tubes as well as in the output circuits of tubes V1 and V2, and in addition this voltage may constitute the total output voltage of tube V4. The total input signal voltage of tube V1 may comprise the voltage across R1 plus an input signal E1, and that of V2 may comprise only the voltage across R1 or if desired another input signal similar to E1 but opposite in phase may be added to the voltage of R1 (see Fig. 4). Preferably, the signals in the output circuits of V1 and V2, which are combined tov produce the desired difference or output signal, may comprise the total output signals less the voltage across R1, although the total ouput signals migh be used.

In the preferred mode of operation of the device, the signal E1 will be small as compared with the signal E2. Further, in general, the combined space current of tubes V1 and V2 will be of the, same order as that of tube V4. The cathode may then offer certain unusual advantages. will be seen that when a signal is applied tothe input circuit of tube V4, the tube will tend to',v

draw enough current so that the voltage across R1 is. the same as the impressed voltage. Under these conditions, the actual grid-to-cathode or tube input voltage would be simply the bias voltage. The tube will not, in general, however, draw suiiicient current to bring about this condition, but will fail by the differential grid voltage required to cause the change in space current. In other words, the tube is not actuated by the input voltage but bythe dinerence between that voltage and the voltage obtained across the resistor R1. By designing the circuit so that a small change in actual grid voltage of tube V4 will cause a large change in the voltage drop across the resistor R1 due to the change in space current of tube V4, the actual voltage across R1 may be made to follow the input voltage E2 very closely.

It will be observed that the operating range of input voltages may be-considerably larger than the total rated grid swing of the tube V4, since the actual grid-to-cathode voltage is the difference between the voltage E2 and the voltage drop across the resistor R1, neglecting, of course, the unidirectional component which may be controlled by the average bias on the grid of tube V4. Thus, for example, the actual grid-to-cathode voltage on. tube V4 may be a small fraction, say

. 116, of the actual input voltage. Since the circuit may be designed so that the tube operates over only part of its grid voltage-plate current characteristic, signal distortion due to non-uniformity of tube characteristic may be minimized to a considerable extent.

It is worthy of note that .degeneration of the signal E1, which is'usually present when an appreciable impedance is introduced in the cathode circuit of a vacuum tube amplifier having an anode load circuit, is avoided due to the property of this circuit of balancing out signals introduced in both branches. It should be noted further that the effect of variations in space current of tubes V1 and V2, which variations are small, is further materially reduced by the use of the cathode load for tube V4. Since this tube is actuated by the difference in voltage between the input signal E2 and the voltage drop across the resistor R1, which difference voltage will be small, it makes little difference whether the latter voltage is due to space current in tube V4 or to some other cause, and in any event, the tube V4 will operate to minimize the discrepancy in voltage between signal Ez and the resistor voltage. Thus, the circuitof tube V4 not only establishes the bias on tubes Visand V2 in accordance with the signal E2, but it also `functions to minimize any tendency of these tubes to have any other bias, and thus minimizes any degeneration of the signal Ei due to the cathode resistor R1.

In order to obtain full advantage of the cathode load feature of the inventin, it is desirable that tubes V1 and V2 have an appreciable load in their respective anode circuits` In the circuit of Fig. 1, the anode circuits of tubes Vi and V2 are loaded to different extents, the/.loading of tube V: being less than that of tube Vi, due to the operation 4of tube V3 which tends to reducethe effective impedance of Ra. In Fig. 4, there is shown a modification of the invention which provides for loading to the full extent. As before, it is desirable to obtain in an output circuit the difference between the output signals of tubes V1 and V2.

' circuit of Fig. 4, tubes Vi and V2, constituting one group, are supplied with load impedances Ra and R4, respectively, which are of suitable magnitudes to act as proper loads for their respective tubes. The amplified signals obtained across Ra and R4 are supplied through blocking condensers to the grids of tubes V5 and Ve, constituting another group. 'Ihe cathode of tube Vs may be connected to the plate of tube Ve, the anode of tube V5 may be connected to the B supply, and the cathode of tube V5 may be connected to ground. If desired,

the plate voltages of tubes V5 and Ve may be increased by connecting the cathode of tube V5 to the negative side of a second B voltage supply unit whose positive side is connectedv to ground. The common junction of the tubes V5 and V5 is connected to an output load resistance R5 which may be connected to an intermediate point on the whole B voltage supply. Obviously, this intermediate point might alternatively be provided by a drop wire or other similar device. Tubes V5 and V6 should preferably have identical electrical characteristics, and their respective input Ygrids may be suitably biased to similar operating points by the grid-leak resistors Re and R1, as shown. 'Y

Tubes V5 and Va are thus connected so that their output circuits are in series with each other and the battery B. The output circuit for the group, however, is the resistance R5 which for A. C. components is effectively connected across each tube output circuit. Signals across R. are supplied to the input circuit ofytube Vs while signals across R3 are combined with the output voltage across R5 and then supplied to the input circuit of tube V5.

By this interconnection of input and output circuits, a circuit is obtained such that the output signal is proportional to the difference bev tween the input signals, and yet the system is perfectly stable.

This result is due to the fact that as far as results are concerned the output circuit has the properties of a cathode load output circuit, but the circuit interconnections are such that sig- `nals introduced in one input circuit are transferred without any phase reversal, while signals in the other circuit are transferred'reversed in phase. Thus signals supplied to the input circuit of tube V5 are reversed in phase and transferred to the output circuit from which they are fed back to the input of tube V5 which tube oporates to control the amplitude of the output signal so that it will be the same as that due to an input 3 signal supplied to tube V5 which signal, however, would be transferred to the output circuit without reversal of its phase. Thus the output signal will be vproportional to Athe diierence between thev input signals. For example, in the case in which the same signal is applied in each input circuit, the current in each tube would change by the 'same amount and there would be no current change in R5 since the current through this ime pendance is the difference between the current in each tube. Whereas, if equal and opposite signals were applied to' each/input circuit/ the current in one tube would increase while that of the other would decrease, the difference current through R5 would produce an output signal which would-feed back affecting the grid voltage of the one tube and the plate voltage of bothV in such a way as to cause equal amplification of both input signals. The resultant output :signal would then' be proportional/,to twice the input signal.-

Consider no'w the operation of the circuit of Fig. 4 as a whole. The groupfof tubes Vi and V2 may be biased to the same point by means of the common resistance Rand the common voltage across Rn A signal E1 may be combined with the voltage across Ri and supplied to tube Vi. If desired, a voltage of opposite phase with respect to Ei may be similarly introduced in the input circuit 'of the tube V2 by the push-pull input circuit shown. Output signals from tubes Vi to-cathode voltage of tube Ve will correspond to the voltage across R4 while that of tube V5 will l.

be the voltage across R3 combined with that across R5. f

Preferably R3 and R4A should have the same vaue, and V5 and Vs should be similar tubes. Under these conditions, the output signal across R5 will be proportional to the difference signals supplied to the group of tubes Vi and V2, and the proportionality factor will be dependenty upon the control signal Ei which will not otherwise appear in the output circuit. The input circuits of tubes V5 "and Vs constitute high impedance sources and consequently do not tend to restrict the use of proper load impedances R3 and R4 in the output circuits of tubes V1 and V2. Hence. the optimum operating conditions for the circuit as a whole may be obtained.

Thus, as before, an input signal Ei is supplied to one of two amplifying branches each of which includes-an amplifier'tube whose mutual con-` ductance is a function of its grid voltage. II desired. a similar signal of opposite phase may be supplied to the input circuit of the other branch. n A second input signal is supplied to the input circuits of both branches in such a way that the mutual conductance of an amplifier tube in veach branch is modified, producing in each il J branch an identical undesired signal and d'eter-A/z mining the degree of amplification of the signal E1.l The output of-each of these 'branches is then combinedin a circuit which forms in its output circuit a signal proportional to the difference between the two input signals, that is, a .signal corresponding to E1 but amplified by an amount determined by E2, and which permits'thc use of load impedances in the anode circuits of the input tubes, thus enabling better control of the gain of these tubes by the control tube.

The amplifier shown in Fig. 1, and described above, is adaptable for use over a wide frequency range due to the fact that, in general, resistancecapacitance coupling is used and, as well known, such coupling devices are substantially uniform over a frequency range extending from some low value to reasonably high values. For use in television system. however, where extremely wide frequency ranges are encountered, certain features are desirable, as will now be described.

To obtain uniform amplification over the videofrequency range, which may extend upward to two or thru magacycles in a high definition television system, it is preferable to add a small amount of inductance to the load resistance of each video-frequency stage in order to minimize the effect of 'the capacity to ground of the coupling condenser and the inter-electrode capacity of the amplifier tube. Such inductances are shown at L and L1 in Fig. 1.

Referring to Fig. 1 for the moment and considering the operation of the device, it will be seen that with the inductances present, the frequency characteristics of the two branches are not similar and, consequently, in order to balance out very high frequency components introduced in both branches, it is necessary to modify the high frequency response of one branch, for example, the branch including tube V2, to make the frequency response of that branch conform to the frequency response of the other branch; or,

alternatively, it is necessary to take precautions' to prevent the introduction of such high frequency components in the circuit itself. Where such high frequency components are outside the range of frequencies which it is desired to transmit, but may be included in the several input signals, it is preferable to remove the saidcomponents at the point of introduction to the circuit. Where it is desirable to transmit the said co nents, the frequency characteristic of the circ t may be corrected, for example, by inserting a very small inductance L2 in the plate circuit of tube V2. By proper adjustment of the inductances L, L1 and L2, uniform frequency characteristics may be obtained for the two branches. It will be noted that the correction must include not only the effect of inductances added to overcome specific frequency response defects, but also the inductances due the physical length of the connections themselves which are of importance at the high television modulation frequencies.

It is likewise important that the frequency response of each branch be the same at the low frequency as well as the high frequency end, and hence in Fig. 4 the values of the kcondensers and resistors connected to R3 and R4 should be so designed that the transfer circuits will have similar frequency characteristics. As R3 and R4 should preferably be equal, this may easily be accomplished by using condensers and grid leak resistors of equal values.

It is worthy of note that in the circuits of Figs. 1 and 4, no unusual restrictions are placed upon the source of B voltage for the circuits, and the same B supply as that used for associated apparatus may be employed. In addition, one side of each of the several input and output circuits may be grounded for A. C. components which facilitates the use of the circuit. Furthermore, the use of transformerswhich would tend to restrict the operable frequency range cf the circuit is avoided `plus the signal E2.

except for the input of Fig. 4 which might be obtained from any vacuum tube phase inversion circuit, thus completely avoiding the use of transformers.

The variable gain amplifier provided by the invention finds several uses in television systems. For example, due to difference between the center line of the optical system of a television camera tube and the axis of the electron beam generator which supplies the scanning beam, an effect known as the keystone effect occurs and this effect may be overcome by causing the amplitude of the horizontal deflection of the beam to vary in accordance with the amount of vertical deflection. This may be accomplished by amplifying the signal for producing horizontal deflection in a variable gain amplifier, the gain of which is determined by a second signal derived from the vertical deiiecting system. In this instance it is desirable to have the amplitude of the signal E1 substantially directly proportional to a constant 'I'his may be accomplished byv using for V1 and V2, tubes having'the mutual conductance characteristic M shown in Fig. 2. If, for example, the signal E2 varies the operating points of the tubes V1 and V2 from the grid voltage value GVi in Fig. 2 to the value GVz, the amplification or gain of the amplified signal E1 will vary from gm to gms since the stage gain is substantially proportional to the mutual conductance of the tube.

In another instance it may be desirable to have the gain constant ai. one Value Ypart of the time, and zero or some lcw value other times. This may be done by using for V1 and V2, tubes having a mutual conductance characteristic such as Nin Fig. 2. In this casethe control signal Ez may vary the operating point from GV4 to GV: for which points the gain will be zero and proportional to gms, respectively. It will be noted that if the gain curve N is flat topped from l1 to I8,

the voltage of the control signal plus the input signal E1 may be anywhere in the interval il to I8, thus permitting the use of larger input signals Ei. Where a push-pull input circuit such as that of Fig. 4 is used, distortion of the signal El due to the slope of the gm curve is largely balanced out, and larger input signals may be employed. Tubes having characteristics as shown in Fig. 2 are' now available commercially. For example, tubes of the type known as 42 have a characteristic similar to M while tubes of the type known as 47 have a characteristic similar to N. Other tubes having similar or different characteristics may, of course, be used, de-

pending upon the` particular application.

In Fig. 3, there is shown in block diagram a schematic outline of a circuit which may be used to avoid the keystone effect above mentioned. In the figure, horizontal and vertical deecting signals En and Ev respectively are supplied to blocking tube oscillators l and 2 which in turn energize generators 3 and 4 which form the proper scanning signals. In the vertical signal circuit, the blocking tube oscillator 4 also energizes a sawtooth signal generator 5. The output of the generator 5 is supplied to a variable gain amplifier 6 of the type show n in Figs. 1 or 4 which amplifier is included in the horizontal signal circuit and serves to control the amplitude of the horizontal scanning signal in accordance with the vertical scanning signal, by which the keystone effect is eliminated. Both scanning signals are then amplified by power amplifiers l and -8 and used to energize appropriate deflecting an output circuit, and an effective mutual conductance, the mutual conductance of each of said space discharge devices being dependent upon the voltage in the input circuit of the respective discharge device, another space discharge device having an input circuit and an output circuit, an impedance having a voltage thereacross in said input circuits and said output circuits, for establishing said voltage in each of said input circuits and each of said output circuits, means for introducing an input signal in the input circuit of y at least one of said plurality of space discharge devices, means for introducing an ,input signal in the input circuit of said other space discharge device, and means for deriving signals from the output circuits of said plurality of space discharge devices and for combining said derived signals in opposed relation to form an output signal.

2. In an electrical system, a plurality of space discharge devices each having an input circuit, an output circuit, and an effective mutual conductance, the mutual conductance of each of said space discharge devices being dependent upon the voltage in the input circuit of the respective discharge device, said discharge devices having similar electrical properties, another space discharge device having an input circuit and an output circuit, an impedance having a voltage thereacross in said input circuits and said output circuits, for establishing said voltage in each of said input circuits and each of said output circuits, means for introducing an-input signal in the input circuit of at least one of said plurality of space discharge devices, means for introducing an input signal in the input circuit of said other space discharge device, and means for deriving signals from the output circuits of said plurality of space discharge devices and for combining said derived signals in opposed relation to form an output signal.

3. In an electrical system, a plurality of space discharge devices each having an input circuit, an output circuit and an effective mutual conductance, the mutual conductance of each of said space discharge devices being dependent upon the voltage in the input circuit of the respective discharge device, another space discharge device having an input circuit and an output circuit, an impedance having a voltage thereacross in said'input circuits and said output circuits, for establishing said voltage in each of said input circuits and each of said output circuits, means for introducing an input signal in the input circuit of at least one of said plurality of space discharge devices, means for introducing an input signal in the input circuit of said other space discharge device, and means including an impedance in at least one of said output circuits for deriving signals from the output circuits of said plurality of space discharge devices and for combining said derived slgnalsin opposed relation to form an output signal.

4. In an electrical systeni, a plurality oi space discharge devices each having an input circuit, an output circuit and an effective mutual conductance, the mutual conductance of each of said space discharge devices being dependent upon the voltage in the input circuit of the respective discharge device, another space discharge device having an input circuit and an output circuit, an impedance having a voltage thereacross in said input circuits and said output circuits, for establishing said voltage in each of said input circuits and each of said output-circuits, means for introducing an input signal in the input circuit of at least one of said plurality of space discharge devices, means for introducingan input signal in the input circuit of said other space d scharge device, and means including an impedance in each of said. output circuits for deriving signals from the output circuits of said plurality of space discharge devices and for combining said derived signals in opposed relation to form an output signal.

5. In an electrical system, a plurality of space discharge devices each having -an input circuit, an output circuit and an effective mutual conductance, the mutual conductance of each of said space discharge devices being dependent upon the voltage in the input circuit of the respective discharge device, said discharge devices having similar electrical properties, another space discharge device having an input circuit and an output circuit, an impedance having a voltage thereacross in said input circuitsl and said output circuits, for establishing said voltage in each of said input circuits and each of said output circuits, means for introducing an input signal in the input circuit of at least one of said plurality of space discharge devices, means for introducing an input signal in the input circuit of said.

other space discharge device, and means including an impedance in each of said output circuits for deriving signals from the output circuits of said plurality o space discharge devices and for combining said derived signals in opposed relation to form an output signal.

6. In an electrical system, a plurality of space discharge devices for providing signal channels, each having an input circuit, an output circuit and an effective mutual conductance, the mutual conductance of each of said space discharge de- .vices being dependent upon the voltage in the input circuit of the respective discharge device, said discharge devices having similar electrical properties, another space discharge device having an input circuit and an output circuit, animpedance having a voltage thereacross connected to said input circuits and said output circuits, for establishing said voltage in each of said input circuits and each of said output circuits, means for introducing an input signal in the input circuit of at least one of said plurality ci space discharge devices, means for introducing an input signal in the input circuit of said other space discharge device, means including an impedance in at least one of said output circuits for deriving signals from the output circuits of said plurality of space discharge devices and for combining said derived signals in opposed relation to form an output signal, and means associated with at least one of said plurality of spa discharge devices for balancing the frequency characteristic of the saidoutput signal channels.

7. In an electrical system, a. plurality of space discharge devices each having an input circuit, an output circuit, and an effective mutual conductance, the mutual ,'conductanceK of each of said and for introducing an yinput: signal of opposite phase relation in the input; circuit of another of PHILIP J. KON'KLE.

Images