US1903542A

US1903542A - Audion amplifier

Info

Publication number: US1903542A
Application number: US515347A
Authority: US
Inventors: Alfred W Barber
Original assignee: Individual
Current assignee: Individual
Priority date: 1931-02-12
Filing date: 1931-02-12
Publication date: 1933-04-11
Anticipated expiration: 1950-04-11

Description

April 11, 1933.

E flefscfar aufpu/f 1 /73 A. w. BARBER 1,903,542

AUDION AMPLIFIER Filed Feb. 12, 1951 2 Sheets-Sheet 1 S Qace Carrel)? April 11, 1933. w BARBER AUDION AMPLIFIER Filed Feb. 12, 1931 2 Sheets-Sheet 2 Patented Apr. 11, 1933 mam) w. BARBER, or aoonron, new mnsnv AUDION mmrma REISSUED dpplication filed February 12, 1931. Serial 110. 515347.

This invention relates to audion amplifiers and more particularly to improved methods of and circuit arrangements for automatlcally controlling the gain of the amplifier.

In the reception of modulated carrier wave signals, it is desirable to operate the high frequency amplifier at maximum sensitiv ty so long as the received signal energy 1s msuflicient to produce a predetermined input voltage across the detector, and to reduce the sensitivity of the amplifier for stronger signals at such rate that the detector input remains substantially constant. Some types of automatic gain control, and particularly those employing a diode both as a demodulator of the received signal and as the rectifier for producing a direct current gain control voltage, are open to the disadvantage that the sensitivity is automatically reduced for all signal inputs above the threshold value which will actuate the receiver, thus defeatin the purpose of the automatic gain control wEich is to obtain a constant output level over the widest possible range of received 25 signal energy.

Objects of the resent invention are to provide methods 0 and circuit arrangements for automatic gain control which are characterized by the substantial postponement of 9 a reduction in sensitivity until the received signal energy has risento the value corresponding, at maximum sensitivity, to the desired output level. Further objects are to provide methods of and circuit arrangements for opposing the effects of the automatic gain control voltage produced during reception of signals of relatively low magnitude.

More particularly, objects are to provide methods of and circuit arrangements for automatically controlling the gain of an amplifier by the use of two biasing potentials which vary automatically with received signal energy; the effects of the two bias poten- 5 tials being in opposition and substantially nullifying each other for all signals of a magnitude less than that efiective, at maximum sensitivity, to produce normal output.

These and other objects of the invention will be apparent from the following specification when taken with the accompanying drawings, in which Fi l is a curve sheet showing the relations ip between amplifier input and output for an ideal gain control system, and the relationship actually obtained with one embodiment of the invention,

Fig. 2 is an explanatory diagram based. upon the volta e-curr'ent "characteristic of an audion ampli er,

Fi '3 is a schematic circuit diagram of one orm of gain control system constructed in accordance with the invention,

Fig. 4 is a curve sheet showing the variations, with carrier wave voltage, of the auto- (:3 matic bias potentials, and d Fig. 5 is a circuit diagram of one embodiment of the invention.

As usually understood in connection with radio receivers, an automatic volume or au- (3 tomatic gain control would, if ideal operation obtains, maintain maximum sensitivity so long as the received signal energy was insufiicient to' raise the detector input voltage to a predetermined or normal level, and 7 would reduce the sensitivity as the signal energy increased beyond that critical value, .the decrease insensitivity for increasing signal strength being at such rate that the detector input remained substantially constant.

In Fig. 1, the curve A represents the relationship which would exist between the amplifier input and output voltage if an ideal automatic gain control were employed. As the incoming carrier voltage E increases 8 from its threshold value, the amplifier gain should remain constant at its maximum value until the incoming signal reaches that value e which brings the detector input voltage E to the predetermined value -E at which the receiver is" to be operated. In other words, the amplifier output should be proportional to the input for all signals below the normal signal input voltage which, at maximum sensitivity, corresponds to normal ampli--' fier output or detector input voltage. For all higher signal voltages, the amplifier output should remain constant. I

Curve B shows the relationship existing between amplifier input E and amplifier out- 1 put E in one known form of automatic gain control system in which a diode serves both as the signal demodulator and the source of automatic gain control potential. A comparison of curves A and B demonstrates that such systems do not satisfy the essential requirements of a. gain control which, over a wide range of impressed signal voltages, will maintain constant output, since the output does not reach its intended or normal value E until the signal voltage rises to avery high value.

Systems of this old type limit the output to values much less than proportional to input for all signals, and their operation is not characterized by a quick rise to normal output, followed by constant output, as the signal voltage is increased.

In accordance with the present invention, the sensitivity of the amplifier may be kept approximately constant until the carrier input reaches its normal value e and the normal'ouput voltage E is reached at somewhat greater carrier input. The curve C of Fig. 1 illustrates the control obtained with one embodiment of the invention, and it, will be noted that this curve is a close approximation to the ideal operating characteristic A.

To defer the reduction in sensitivity, use is made of two bias potentials which vary automatically with signal strength and whic for any given change in signal strength tend to change the amplifier gain in the opposite sense.

One bias potential is obtained, in the known manner, by rectification of the carrier and this direct current voltage increases as a substantially linear function of the carrier when, as is the usual case, a linear rectifier is employed.- The other component of the control potential comprises the direct current drop across a. resistor in that part of the cathode circuit which is common to the grid and plate circuits of the controlled amplifier, i. e'., an autoor cathode bias. Although the cathode bias potential doesnot vary as a direct result ofchanging carrier wave energy, its magnitude is dependent, at least in part, upon the magnitude of the rectifier bias potential which varies automatically with signal strength. i v

- The relative effects of these two bias potentials will be apparent from a consideration of the amplifier voltage-current characteristic, Fig. 2, and the circuit diagram, Fig. 3. The curve D of Fig. 2 shows the relationship between grid bias E and space current I for an amplifier tube and, as shown in Fig. 3, the effective grid bias, potential E is determined by the algebraic sum of three components, two of which vary automatically with signal strength.

As shown diagrammatically in Fig. 3, the input circuit 1 of the amplifier tube 2 is connected to the tube grid and, through a.

blocking condenser 3, to ground. The resistor 4 is connected between thetube cathode and ground, and therefore the direct current potential established across this resistor by the space current flow provides a bias potential efi'ective between the amplifier grid and cathode. The plate circuit of tube 2 is connected by a network 5, (which may be simply an interstage cou ling or may include additional amplifiers; to the plate of the diode rectifier 6. The circuit between the diode plate and cathode includes the radio frequency circuit 7 and the resistance 8 which is by-passed for radio frequency currents by a condenser 9. A lead 10 connects the low potential terminal of the amplifier input circuit 1 to that end of resistance 8 which is spaced from the diode cathode, and the latter is connected to ground through a battery 11. No filter network is shown for eliminating audio or radio frequency components from the direct current bias potential feed back to amplifier 1 since the exact form of such network has no bearing upon the theory of op-. eration of the gain control system.

In the absence of any signal voltage E across the amplifier input, the bias E between the amplifier grid is:

where E, is the potential drop due to the flow of space current through the cathode bias resistor 4, E is the potential drop arising from the flow of space current in the diode, and E, is the voltage across that portion of battery 11 which is between the diode cathode and ground.

Referring now to Fig. 2, the point 0 on curve -D is the operating point for maximum sensitivity of the amplifier tube, and operation about this point of the characteristic is effected by the application of a negative bias voltage e As is well known, the slope of a line b drawn through point 0 and the origin 0 gives the magnitude of the resistance 4 which would, for space current flow corresponding to point 0, give a cathode bias potential of e In accordance with the invention, the cathode bias resistance 4 is substantially larger than the magnitude indicated by the slope of line b, the slope of line 0 being so chosen, with respect tothe curvature of the characteristic D, that the cathode bias voltage E represented graphically by the projection of line a on the voltage axis, decreases rapidly as the bias potential is increased negatively to shift line 0 towards the left.

As indicated in Fig. 2, the initial bias e is the algebraic sum of the negative cathode bias E,, the negative rectifier bias E which is established across rectifier circuit resistance 8 by the. space current for zero input voltage E and the positive bias potential E, from the source 11.

As shown graphically in Fig. ,2,-the highcathode bias resistor 4 effects a substantial reduction in the net change in'bias potential a, when, for small signal inputs less than the normal input, the rectifier bias E increases with increasing signal stren When the rectifier bias E, is numerica ly equal to the fixed bias E the operating point 0' is located by drawing through origin 0, a line 0' parallel to the line 0. It ,will be noted that the cathode bias potential has dropped materially and therefore the net change in the gain control bias e is substantially less than the change in the rectifier bias E The relationship between the cathode bias potential E the rectifier bias potential E and the gain control bias E for increasing signal strength E, is illustrated in Fig. 4. For the particular gain control system from which the data was obtained, the decrease in cathode bias was somewhat less than the increase in rectifier bias for values of E less than normal input e thus accounting for the fact that the control curve C, Fig. 1, is slightly lower than the inclined branch of the ideal control curve A. For somewhat greater values of signal input E, but little change in the cathode bias otential E is produced by increasing recti er potentials E and therefore the curve e is substantially parallel to the rectifier potential curve E One practical embodiment of the invention, as illustrated in Fig. 5, comprises a twostage tetrode amplifier working into a diode rectifier. The input circuit of each tetrode amplifier 12 includes in series, the inductance 13, tuning condenser 14 and a blocking condenser 15 of 0.1 microfarad capacity.

A cathode resistance 16, by-passed for radio frequency currents, is included in each amplifier circuit. The circuit of diode 17 comprises the tuned radio frequency impedance 18 in series with a resistance 19 of approximately 1 megohm which is by-passed for radio frequency by condenser 20. The audio frequency potential developed across resistance 19 may be passed to an audio frequency amplifier, not shown, and the rectified direct current potential is passed, by lead 21 to the grid circuits of the amplifiers, resistances 22 of a half megohm being included to suppress audio frequency components from the bias potentials passed to the amplifier grids. The positive bias potential necessary to reduce the high negative bias components, due to the noload space current fiow in the respective tubes, to the appropriate value for maximum sensitivity is supplied by a battery 23. The values stated are appropriate when the'tetrodes are of the commercial 224 type and the diode is of the commercial 227 type, with grid and plate connected together.

It will be apparent that the use of two bias potentials which vary automatically with signal strength affords wide latitude in the design and construction of automatic ain control systems. While I have describe one par ticular type ofcontrol, it will be apparent that the effects of changing cathode blas p0- tentiak and changing recti r potential may, if desired, be combined to produce control characteristics substantially different from that indicated by curve C of Fig. 1.

It is also apparent that the invention is not limited to systems in which the rectifier has the further function of demodulating the signals transmitted by the amplifier, or in which the rectifier input voltage is derived from the amplifier output.

I claim:

1. In the operation of an electrical wave amplifier working into a rectifier and subject to signals of varying magnitude, the method of automatically controlling the gain of said amplifier, which comprises impressingupon the amplifier, input circuittwo discretedirect current potentials which vary automatically and according to different functions of the magnitude of the incoming electrical wave, one of said potentials being predominately effective for weak signals and the other predominately effective for strong signals.

2. In automatic Volume control apparatus wherein two discrete potentials are employed to control amplifier gain, means whereby one of said discrete potentials is obtained by rectification of the amplifier output, and means whereby the second potential is derived from the space current flow in the cathode circuit of the amplifier.

3. The method of operation of an electrical wave amplifier so as to secure automatic gain control thereof, which comprises applying to said amplifier a gain control potential which varies automatically with the magnitude of the input of said amplifier and applying to said amplifier a second gain control potential which varies automatically as a function of said first gain control potential.

4. The method of operation of an electrical wave amplifier so as to secure automatic gain control thereof, which comprises applying to said amplifier a gain control potential which varies automatically with the magnitude of the input of said amplifier and applying to said amplifier a second gain control potential which Varies automatically as a function of said first gain control potential, said second gain control potential being obtained by the voltage drop due to the flow of current in the cathode circuit of said amplifier.

5. The method of operation of an electrical wave amplifier so as to secure automatic gain control thereof, which comprises applying to said amplifier a gain control potential which varies automatically with the magnitude of a the input of said amplifier and applying to varying magnitudes so as to secure an automatic control of the gain of said amplifier, which comprises impressing a direct current potential varying with signal strength upon the input circuit thereof and impressing upon said input circuit a second direct current potential automatically decreasing in magnitude as the strength of the received electric signals increases, both of said direct current potentials functioning as gain control voltages.

7 In an electrical wave transmission system, the combination with an electron tube amplifier having input and output circuits, and means for automatically impressing u on said amplifier a gain control potential which varies in magnitude with changes in the magnitude of the electrical waves to be amplified, of means substantially preventing said automatic gain control potential from affecting the amplifier gain until the electrical waves increase in magnitude to a predetermined value, said last mentioned means including a cathode bias resistance metallically connected in the input circuit of said amplifier.

8. In an electrical wave transmission system, the combination with an electron tube amplifier, a rectifier, and means impressing upon said amplifier a gain control voltage derived from said rectifier by rectification of alternating currents traversing said s stem, of means impressing upon said ampli er a second gain-control potential of a ma nitude determined as a function of said 'rst control voltage and operative conjointly therewith so as to prevent said first control voltage from being substantially effective un-- til it reaches a predetermined value.

9. In an electrical wave amplifier, the com- 7 bination with a vacuum tube and input and output circuits therefor, a rectifier and means impressing 'upon the input circuit thereof electrical waves varying in magnitude with v the electrical wave input to said vacuum tube, and circuit elements associated with the rectifier for impressing upon said tube a gain control potential varying with rectifier out put, of means in circuit with said tube for developing a second gain control potential which varies in magnitude with the gain control potential impressed thereon by said rectifier and operates conjointly therewith so as to prevent said first controlled voltage from being substantially effective until i it reaches a predetermined value.

10. An electrical wave transmission system of the type including an am lifier and a rectifier for'impressing upon said amplifier a control grid bias potential varying automatically with changes in the amplitude of the electrical waves impressed upon said amplifier, characterized by the fact that the oathode circuit of said amplifier includes a bias resistor across which the space current flow establishes a bias potential, said resistor having a magnitude such that for a predetermined range of amplifier input voltages the changes in, cathode bias potential substantiall ofl'set changes in the rectifier bias potentlal.

11. In a carrier wave receiver, the combination with a carrier wave amplifier, a diode rectifier having input and output im edances associated therewith, means coup ing the amplifier output circuit to the input impedance of said. diode rectifier, and circuit elements for transferring back to said amplifie'r a gain control voltage developed across the output impedance of said diode rectifier, of a cathode bias resistor in said amplifier of such magnitude as substantially to postpone reduction in amplifier gain until said gain control voltage reaches a predetermined value.

- 12. In a carrier wave amplifier, the combination with an amplifier tube having input and output circuits, a diode rectifier having an input circuit across which is developed av carrier voltage varying with received carr1er wave voltages, of means operative in the absence of recelved carrier waves to has said amplifier tube to maximum sensitivity, said