US2230483A - Stabilization for vacuum tube amplifiers - Google Patents

Description

15 T I INPUT E 7 VQLI'AGE Feb. 4, 1941. M CAGE 2,230,483

STABILIZATION FOR VACUUEI l TUBE AMPLIFIERS Filed June 5, 1940 I r jg ,Fjgu.

1- a: 6. A g

VOLTAGE 5 4/ 4 INIIUT z W 5' Patented Feb. 4, 1941 UNITED STATES STABILIZATION FOR VACUUM TUBE All/[PLIFIERS John Martin Cage, Boulder, 0010.

Application June 5, 1949, Serial No. 338,912

Claims.

My invention relates to improvements in vacuum tube amplifiers, and more particularly to the maintenance of substantially constant voltage amplification and linearity in amplifiers, re-

5 gardless of changes in vacuum tube characteristics, circuit parameters, or operating conditions. While not limited thereto, my invention is particularly applicable to the purpose of stabilizing regenerative amplifiers and preventing them from bursting into spontaneous uncontrolled oscillation when some small change occurs in the conditions of operation.

Th invention and its objects and advantages can best be understood by reference to the foll5 lowing specification taken in connection with the drawing, in which Fig. l is a schematic diagram of an ordinary radio-frequency amplifier utilizing regeneration, Fig. 2 is a diagram showing how one simple embodiment of my invention can be applied to the regenerative amplifier of Fig. 1, and Fig. 3 is a diagram showing how my invention can be used to obtain an audible-frequency amplifier that afiords high power sensitivity, low

degree of distortion, and substantially constant voltage amplification within the operating range of frequencies.

Referring more particularly to Fig. 1, the vacuum tube l is a multi-grid high vacuum tube such as the hexode section of a type 6K8 receiving tube. Automatic grid bias is secured by means of resistor 2 and condenser 3, and the battery 4 supplies positive voltage to the screen grid, and anode, circuits. The cathode is of the indirectly heated type, and provision not shown on the drawing is made for heating the cathodes of all the vacuum tubes used for explanation. The capacitor 9 is a radio-frequency by-pass. An alternating input voltage of radio frequency, which is supplied through the tuned input circuit I3, is applied between the grid nearest the cathode and the ground terminal, and the amplified output voltage is taken across the anode resonant circuit consisting of inductance coil 5 and capacitor 6. Henceforth the said anode resonant circuit will be referred to as the output circuit. The amplifier is made regenerative by inductively coupling coil 1 to coil 5, so that an induced voltage is applied between the third grid 50 and ground in the proper phase relation to increase the voltage across the output circuit. The regeneration thereby obtained is varied, or adjusted, by controlling the positive direct voltage applied to the second and fourth grids, which act 55 together as a screen grid. The potentiometer 8 controls this screen grid voltage and therefore also controls the degree of said regeneration.

Now, to those skilled in the art it is a wellknown fact that when the regeneration in an amplifier of the sort shown by Fig. 1 is set very 5 close to, or at, the threshold of oscillation, then there are many small and often random changes in the vacuum tube, in the circuit parameters, and in operating voltages which, when they occur, cause the device to break into sustained 10 oscillation, rendering it useless, for the time being, as an amplifier. Still, it is also well known that the oscillating condition must be closely appreached in order to make the regeneration effective. Therefore, in using said amplifier, one 15 would find it necessary to adjust the regeneration control frequently and critically.

It is an object of my invention to stabilize the amplification of regenerative amplifiers. In other words, it is an object of my invention to 20 cause the voltage gain of said regenerative amplifiers to be less subject to violent fluctuations and to cause the regeneration to be more nearly perfectly controllable. For example; consider more particularly Fig. 2 on the drawing, which 5 is a schematic diagram wherein I have applied a simple form of my invention to the regenerative amplifier of Fig. 1. Although at first thought it might seem that the triode vacuum tube in is merely connected in parallel with the tube I, 30 certain conditions are maintained that make the operation of the circuit unique. It is true that said alternating input voltage is applied to a grid of tube l and a grid of tube In in parallel, but the regeneration that takes place through tube 35 I does not affect the input voltage or amplifying properties of tube H), or at least not to the same extent that the regeneration affects the amplifying properties of tube I. In other words, I have in part placed a regenerative amplifier in parallel 40 with a non-regenerative amplifier that operates under conditions that will be described later. More specifically, I have in Fig. 2 placed a more or less conventional regenerative amplifier in parallel with a vacuum tube that can act as 45 either a positive or a negative resistance across said output circuit. The magnitude and algebraic sign of said resistance will depend upon the phase and magnitude of said output voltage, and said phase and magnitude will in turn depend greatly upon the amplifying properties of tube 1 acting as a regenerative amplifier.

Let us analyze the operation of the circuit of Fig. 2. The alternating current that will flow in the anode leading-in conductor M of tube I will produce an alternating potential across the parallel-resonant circuit, or output circuit, composed of inductance 5 and capacitor 6. Now, if said output alternating potential is equal in magnitude to the product of the alternating input voltage and the amplification factor of the triode tube l0, and if said output voltage is electrical degrees out of phase with said input voltage, then no alternating component of current will flow in the anode lead wire ll of tube In; and therefore, tube In will not materially affect the operation of tube I. However, if the alternating potential across said output circuit retains said phase relation of 180 degrees with respect to said input voltage but exceeds the product of said input voltage times the amplification factor a of the tube In, then an alternating current will flow in the lead II when the tube It) is connected in the circuit; and the phase of this current will be such as to enable the tube III to act as some positive value of resistance, connected .across said output resonant circuit. Under this condition, tube It) will absorb power and reduce the alternating p al developed across said output circuit toward a value, n times said input voltage. On the other hand, if the alternating potential across said output circuit should become less than said product of the input voltage' times the amplification factor of thetriode l0, an alternating current opposite in phase to that of the previously cited condition will flow in lead II, and the triode III-will therefore deliver power to said output circuit. This power will increase said output voltage toward the value, p. times input voltage. In other words, stabilization of the output voltage cocurs about this value, which is determined by a, and analysis shows that the greater the output voltage differs from the stabilized value, the greater, usually, is the stabilizing effect. 1

By amplification factor, or I mean the ratio of the change in anode voltage to a small change in grid voltage in a vacuum tube while the anode current remains constant; or in other words, in a vacuum tubethe amplification factor is the ratio between the effect of grid voltage,and the effect of anode voltage, on anode current under given operating conditions.

When I say the product of the amplification factor, or u, times the alternating component of grid voltage, I mean that every successive instantaneous value of the alternating or fluctuating component of grid voltage is to be multiplied by the exact amplification factor existing under the conditions of operation.

A further object of my invention is to utilize the corrective alternating currents that flow in the anode lead I! of the stabilizating vacuum tube to induce voltages electromagnetically or electrostatically that are substantially proportional to said corrective currents and in turn to utilize said induced voltages to aid in the stabilization of the amplifier. This object may be fully understood by reference again to Fig. 2, where said object may be achieved by severing the lead H at point X and connecting the coil 52 in series therewith. With this connection properly made, and with the coil l2 properly wound and adjusted, if the anode current of the tube H] tends to increase the output voltage, then said anode current flows in the coil 12 in the direction which is correct to increase the regenerative feed-back voltage that-is induced in coil 1. This, in turn, increases the alternating anode current in the tube l and thereby aids in stabilizing the total amplification.- Conversely, if the tube iii should draw power from the output circuit, the phase of the current through the coil 12 will be proper to reduce the feed-back voltage induced in the coil 1. The magnetic coupling between the feed-back coil 1 and the output coil 5 is preferably, but not necessarily, loose.

I have constructed the circuit of Fig. 2 with the propel circuit parameters for tuning over a frequency range of from 500 kilocycles to 1500 kilocycles. Although I practised no especial precautions in constructing the amplifier for stability of circuit constants and voltage supplies, I found that I could adjust the regeneration to obtain a voltage gain of approximately one thousand times and then tune over the entire frequency range of one thousand kilocycles without finding it necessary to readjust the regeneration control 8. With the triode l0 inoperative, however, the regeneration could be practically adjusted to obtain a voltage gain of only about two hundred times, and even then, the circuit would burst into violent oscillations whenever the tuning controls or certain operating conditions were very slightly changed.

It is obviously within the scope of the invention to connect the stabilizing tube anode lead ii to some point on coil 5 other than the point to which the anode lead I4 is connected. It is also apparent that the input grids of tubes I0 and i need not be connected to points having the same potential in the input circuit I3. Many other variations that come Within the spirit of the invention, such as the fact that the stabilizing tube Hi need not be a triode, will occur to those skilled in the art. 7

If, under some conditions, troublesome feedback should occur through tube I0, this feedback can be eliminated by means of neutralization or by using a stabilizing tube with internal electrostatic shielding.

It is the previous difiiculty in securing a high degree of amplification at high frequencies which has made it desirable for engineers to resort to frequency conversion, or in other Words, the superheterodyne principle, in designing radio receivers. Whereas frequency conversion requires a comparatively large assortment of circuit components, my invention enables one to secure high voltage gains with greater simplicity and economy. I i I Although, as I have demonstrated, my invention is especially useful for the amplification of voltages at high frequencies, it is not limited thereto. Refer now more-particularly to Fig. 3 on the drawing, which is a schematic diagram of an audio amplifier incorporating myv inven-. tion. The tube I 5 is the hexode section of a multi-grid receiving tube of type 6K8 or a tube similar thereto. The tube 16 is a power output tube. The tube I! is atriode which produces stabilization. The anode of the tube I5 is coupled to the control grid of thetube IS in a conventional manner that will be fully understood by any skilled person upon reference to the drawing. Theanode circuit of tube It is in turn coupled through a transformer l9 to a'loudspeaker iii in a conventional manner. Other connections are made as'shown on the drawingQand the amplifier can be readily duplicated by anyone skilled in the art. Resistor 2| and capacitor 22 effect automatic direct current grid bias for tube 16. g I

Now let-us an'alyze the operation of the; circuit. A signal voltage of audio frequency is introduced at the input terminals and coupled through capacitor 20 to a grid circuit of tube l5, where said signal voltage causes a variation of the anode current and thus an alternating voltage in the control-grid circuit of tube l6. Further, an alternating voltage will appear across the windings of the transformer l9; and the loudspeaker IE will respond. In order to cause regeneration in the amplifier, a proportional part of the alternating voltage across the primary of the transformer 19 is placed in series with the screen grid supply voltage battery 23 by means of capacitor 24 and potentiometer 25. The slider on the potentiometer 25 allows adjustment of the regeneration. Now, a proportional part of said signal voltage is applied also to the control grid circuit of the stabilizing tube H, by means of resistors 26 and 21'. The anode of said stabilizing tube is connected to the primary winding of the transformer 28, and the other side of said primary is connected to a voltage composed of a proportional part of the loudspeaker voltage superimposed upon the voltage of the battery 29. The alternating component of said anode voltage supply is controlled by the setting on the potentiometer 30. Now, if the amplifier is properly adjusted, said alternating component of voltage will be 180 electrical degrees out of phase with said input signal voltage and will also be equal in magnitude to the product of the amplification factor, times the grid signal voltage component of said stabilizing tube l1; and therefore n0 alternating current will flow in the primary of the transformer 28 and no voltage will be induced in the transformer secondary. However, if because of some tube or circuit instability the magnitude of the alternating voltage across the potentiometer 30 should change, an alternating current will flow in the primary of the transformer 28, a voltage will be induced in the secondary thereof, and said secondary voltage will be supplied to an auxiliary grid circuit of the tube IS in the proper phase to oppose and limit the instability. Likewise, any harmonic distortion components of voltage that may appear across the loudspeaker will be similarly opposed and limited to magnitudes that are small fractions of the magnitudes that they would reach in a similar amplifier without the benefits of my invention.

Of course, circuits that will apply some sort of corrective influence to an amplifier are known to the prior art; for instance, one such circuit is described by P. O. Pedersen at page 59 of vol. 28 of the Proceedings of the I. R. E. for February, 1940. One of my specific improvements in devices that reduce distortion and improve stability in vacuum tube amplifiers is that I cause the output voltage of an amplifier to be stabilized about a magnitude that is determined by the input voltage and the amplification factor of a vacuum tube; and since said amplification factor can be made to remain substantially constant over a wide range of operating conditions, said magnitude of stabilizing value, or in other words, the comparison value, will at any given input voltage remain constant also. Furthermore, since the inclusion of my invention in an amplifier will produce negligible increase or no increase in the power taken from the source of input voltage, said inclusion will have no deleterious efiect on said input voltage. This is a distinct improvement over most of the methods employed heretofore.

While I have described and illustrated my invention in connection with two particular kinds of amplifiers, it is to be understood that the invention is not limited to amplifiers such as these which I have shown herein. Obviously, the corrective current that will flow in the anode circuit of a vacuum tube used in accordance with my invention can be utilized by skilled persons to stabilize the operation of many types of amplifiers and over a wide range of frequencies and a wide assortment of the kinds of vacuum tubes used in said amplifiers. It is further evident that regeneration is not a necessary part of the invention; one or more vacuum tubes can be used in accordance with the claims of my invention to provide decreased distortion in almost any type of amplifier.

It is also within the spirit of the invention, when a vacuum tube having more than one grid is used as a stabilizing tube, toassociate the screen grid of said stabilizing tube, instead of its anode, with the output circuit of the amplifier. In case this arrangement is made, the screen grid may be considered as an anode that allows part of the tube current to pass through tobe collected by another electrode. Whether an anode collects all the tube current or allows part to pass through to another electrode, there will be an amplification factor for said anode and the control electrode taken in combination; and if the alternating component of anode voltage is reversed in polarity with respect to the alternating component of control electrode voltage, if both alternating components have identical wave forms, and if the amplitude of the anode alternating voltage is equal to the amplitude of the control electrode alternating voltage multiplied by said amplification factor, then the alternating component of said tube current will be zero. Therefore, said tube current will be a function of the distortion in said anode alternating voltage and can be used to neutralize said distortion.

What I claim as new and desire to secure by Letters Patent of the United States is:

1. In combination, a regenerative amplifier consisting in part of an input circuit and an output circuit, and a vacuum tube consisting of at least a cathode, a control grid, and an anode, said control grid being associated with said input circuit and said anode being associated with said output circuit in a manner such that when fluctuating current flows in the leading-in conductor of said anode, said fluctuating current will tend to stabilize the phase, magnitude, and wave form of the voltage across said output circuit to values that will enable the alternating component of voltage between said anode and said cathode to become substantially 180 electrical degrees out of phase with the alternating component of voltage between said control grid and said cathode and that will also enable said alternating component of anode voltage to become substantially equal in magnitude to the product of said alternating component of grid voltage times the amplification factor of said vacuum tube.

2. In combination, a regenerative amplifier consisting in part of an input circuit and an output circuit; a vacuum tube consisting of at least a cathode, an anode, and a control grid, said control grid being associated with said input circuit and said anode being associated with said output circuit in a manner such that when fluctuating current flows in the leading-in conductor of said anode, said fluctuating current will cause a stabilizing voltage to exist; and means whereby said stabilizing voltage may be utilized to cause the phase, magnitude, and wave form of the voltage across said output circuit closely to approach values that will enable the alternating component of voltage between said anode and said cathode to become substantially 180 electrical degrees out of phase with the alternating component of voltage between said control grid and said cathode and that will also enable said alternating component of anode voltage to become substantially equal in magnitude to the product of said alternating component of grid voltage times the amplification factor of said vacuum tube.

3. In combination, an amplifier consisting in part of an input circuit and an output circuit, and a vacuum tube consisting of at least a cathode, a control grid, and an anode, said control grid being associated with said input circuit and said anode being associated with said output circuit in a manner such that when fluctuating current flows in the leading-in conductor of said anode, said fluctuating current will tend to stabilize the phase, magnitude, and wave form of the voltage across said output circuit to values that will enable the alternating component of voltage between said anode and said cathode to become substantially 180 electrical degrees out of phase with the alternating component of voltage between said control grid and said cathode and that will also enable said alternating component of anode voltage to become substantially equal in magnitude to the product of said alternating component of grid voltage times the amplification factor of said vacuum tube.

4. In combination, an amplifier consisting in part of an input circuit and an output circuit; a vacuum tube consisting of at least a cathode, an anode, and a control grid, said control grid being associated with said input circuit and said anode being associated with said output circuit in a manner such that when fluctuating current flows in the leadingein conductor of said anode, said fluctuating current will cause a stabilizing voltageto exist; andmeans whereby said stabilizing voltage may be utilized to cause the phase, magnitude, and wave form of the voltage across said output circuit closely to approach values that will enable the alternating component of voltagebetween said anode and said cathode to become substantially 180 electrical degrees out of phase with the alternating component of voltage between said control grid and said cathode and that will also enable said alternating component of anode voltage to become substantially equal in magnitude to the product of said alternating component of grid voltage times the amplification factor of said vacuum tube.

5. In an amplifier, a vacuum tube consisting of at least a cathode, an anode, and a control grid, said vacuumtube being connected in the circuit'of said amplifier in a manner such that a stabilizing influence will be caused to exist Within said amplifier whenever the alternating component of voltage between said anode and said cathode differs in wave form or magnitude from the product of the alternating component of voltage between said control grid and said cathode times the amplification factor of said vacuum-tube, or whenever the phase of said alternating component of anode voltage diifers from a value 180 electrical degrees leading or lagging said alternating component of control grid voltage, or whenever both of said differences occur simultaneously, said stabilizing influence being proper to cause a reduction of said'differences.

JOHN MARTIN CAGE.

Images