US2290659A - Parasitic oscillation suppressor - Google Patents

Description

Jul 21, 194;. mm 2,290, 59

unsure osczm'nbn surransson =Filed may 15, 1941 ji INDULTT l/E 20 22 Inventor: Donald E. Watts,

His Attorney ance B. A pair of electron Patented July 21, 1%42 Panasrrro osc Donald E. Watts, Schenectady,

'rIoN surlrnnsson N. Y, assi'gnor to General Electric Company, a corporation of New York Application May 15,1941, Serial No. 393,597 5 Claims. (or. 179-171) This invention relates to electron discharge amplifiers, and more'particularly to means for damping parasitic oscillations in such amplifiers.

Electron discharge amplifiers, and particularly those which operate at high power levels and thereby necessarily have conductors of substantial size and length constituting various portions of the amplifier circuit, are frequently subject to self oscillation at undesired frequencies difierent from the resonant frequency of the amplifier circuit. Such self oscillation, usually termed parasitic oscillation, is caused by stray capacities and the like acting in resonance with circuit inductances including the inductance presented by circuit conductors due to their physical length. It is an object of my invention to provide improved means for suppressing such undesired oscillation.

In such amplifiers neutralizing circuits are provided for preventing self oscillation of the amplifier. Parasitic oscillations occur because such neutralizing circuits do not perform properly at certain frequencies, at which parasitic oscillations occur. It is a further object of my invention to provide simplified and improved parasitic oscillation suppressing means placed in such neutralizing circuits of an amplifier in such position that most parasitic oscillations are efiectively suppressed with minimum loss of driving power supplied to the amplifier. I

It is another object of my invention to provide other parasitic oscillation suppressing means cooperating with such means in the neutralizing circuit for effectively suppressing parasitic oscillations not affected by the means in the neutralizing circuit. I

The features of my invention which Ibelieve to be novel are set forth with particularity in the appended claims. My invention itself, both as to its organization and manner of operation, together with further objects and advantages thereof may best be understood .by reference to the following description taken in connection with the accompanying drawing in which the single figure illustrates an amplifier circuit embodying my invention.

In the drawing an inductance Hi, excited from a source of high frequency potential, not shown, is tuned to resonance at. such high frequency by four condensers H, [2, l rially in that order between the terminals of the inductance iii. A point between condensers l2 and i3 is connected to ground through a resistdischarge amplifier devices l5 and I6 are excited from the tuned circuit Ill,

3, and it, connected sell, :2, Is, It, and transmit amplified 5 oscillations to a tuned output circuit including an inductance ll inparallel to two serially connected condensers l8 and I9. A point between. these two condensers l8 and 19 is grounded.

Exciting connections for the device I5 are provided from a point between condensers H and i2 through a resistance 20, shunted by an inductive conductor 2|, to the control electrode 22 of the device l5. Similar exciting connections are provided for the discharge device 86 from a point between the condensers l3 and It through a resistance 23, shunted by an inductive conductor 2 3, to the control electrode 25 of the device I6.

In order to transmit unidirectional current from the control electrodes 22 and 25 to ground, points between the condensers H and I2 and'the' condensers I3 and I t are through a series combination of inductance 26 and resistance 2'! and a similar series combination of inductance 28 and resistance 29 to a point maintained at a suitable negative bias potential with respect to ground. tential is. supplied to such point through a resistance 30 from the negative terminal of a alternating cathode source, not shown. In order to transmit unidi- 'rectional currents from the cathodes 33 and 3 3 to ground, the center tap of the secondaries 35. and 36 of the transformer 37 are grounded. In order to transmit high frequency current from the cathodes 33 and 34 to ground, the four cathode terminals of the devices l5 and I6 are respectively connected to ground through four condensers 38, 39, 40, and M.

Anode current for the devices l5 and i6 flows through connections extending from the respective anodes 42 and 43 of ductance I? andan inductance 44, which is shunted by a seriesresonant circuit including an inductance and a condenser 46 which is conof potential, the negative terminal of which source 'is grounded to the cathodes of the disconnected respectively The negative bias p0- such devices thrcugh inresistances 20 and 23 and charge devices. A point between thecondensers l8 and I9 is alsoconnected to ground.

In order to provide neutralization. for the amplifier, a condenser 48 is connected from a point resistances aspossible.

' In the usual amplifier, which does not have the the inductive conductors 2| .and, 24, parasitic oscillations generally occur because the neutralizing circuit is not a perfectly balanced bridge circuit at all frequencies by reason of the fact that the conductors in the bridge circuit have inductance due to their physical length. By placing the resistances 20 and 23, shunted respectively by the inductive conductors 2| and 24, in the circuit common to the grid cathode circuit and the neutralizing circuit of each discharge device, it is possible to suppress most parasitic oscillations which can arise in the amplifier, by producing critical damping in the amplifier circuits in which such oscillations arise, that is, by producing just enough damping in the circuits of the amplifier to prevent oscillations therein, while suitably proportionin'g the values of the resistances, and the inductance of the conductors in shunt thereto, to I require minimum drive volt-age for the amplifier.

known, the correct values for the.

So far as is resistances 20 and 23 and the inductive conductors 2| and 24 cannot be predetermined for a given amplifier. The values are most easily determined by trial, and may conveniently be determined by the following procedure. The resistances 20 and 23 are each given a value in the order of 20 to 100 ohms, the value preferably being 100 ohmsin order to avoid any danger of overheating the resistor. Each resistance is preferably a short non-inductive resistor, such as may be formed by a stick of carbon. Each inductive conductor 2| and 24 is preferably formed by a conducting holder for the accompanying resistance. This holder may have, for example, as is shown in the drawing, an elongated U-shaped form, between the ends of which the resistance is supported. For the first trial it is preferred .to make this conductor as short and straight as possible.

adjustment of the conductors 2| and 24 is satisfactory. If, however, a substantially increased driving voltage is necessary under such conditions of adjustment to maintain rated output for the amplifier still further adjustment must be made. This further adjustment involves decreasing the inductance values of the conductors 2| and 24 in order to obtain suflicient excitation for the amplifier,. with the minimum driving voltage mentioned above.

Such further adjustment comprises decreasing the value'of resistances 20 and 23, and repeating the trials with successively increasing values of inductance for the conductors 2| and 24, until a point is reached at which parasitic oscillations cease. If rated output for the amplifier is then obtained with the minimum driving voltage, the adjustment is satisfactory. If not, the value of resistance 20 and 23 is further decreased in steps, the inductances 2| and 24 being roperly adjusted at each step, until a pc is reached at which parasitic oscillations do not exist and the amplifier operates emciently at its rated output-with minimum driving voltage.

During ment of the resistances tors 2| and 24, the resistances 20 and heat dissipation thereof. If the rated heat disthe above described process of adjust- 20 and 23 and conducz sipation of these resistances 20 and 23 is exv ably two separate resistances dissipation rating may be connected in parallel The amplifier is then excited with a driving voltage of the prpper operating frequency, which is the same as the resonant frequencies of the tuned circuits IO, N, |2, |3, l4, and l1,|8, Is.

This driving voltage is adjusted to the minimum value at which efficient operation of the discharge devices 5 and [6 may be attained. If parasitic oscillations appear after such adjustment of the resistances 20and 23 and conductors 2| and 24, further adjustments must be made. The first step in such further adjustments is to increase the length, and correspondingly the inductance, of the conductors 2| and 2'4. By increasing the inductances of these conductors in successive steps a point can be reached at which parasitic oscillations cease. If at such adjustment the amplifier provides its rated output with the above-mentioned minimum driving voltage, the

ceeded, the final adjustment to obtain efllcient operation of the amplifier is unobtainable with resistances of such rating. Either single resistances of higher rating without substantially greater inductance may be utilized, or preferof the same heat to each other to form each of the resistances 23 and 23. The use of a plurality of such resistances connected in parallel not only multiplies the heat dissipation of the resistance element formed thereby, but also decreases the inductance of the resistance element,-thereby decreasing the inductance necessary in the conductors 2| and 24 and correspondingly making it easier to utilize the minimum driving voltage to obtain rated output for the amplifier.

After such adjustments have been completed, it may be found that parasitic oscillations still exist which cannot be suppressed by such means. If such be the case, the resistances or the series resonant circuit 45, 46 may be utilized to suppress the remaining parasitic oscillations.

.Such unsuppressed parasitic oscillations may be caused by the discharge devices l5 and I4 operating either in push-pull, or balanced relation, or in parallel with each other. If the parasitic oscillations appear in substantially constant intensity all along the length of the in-' du'ctance II, they are caused by the devices and I3 operating in parallel. If, however, the parasitic oscillations are most intense at the ends of the inductance l1, and are a minimum at its frequency current at such frequency, and conthe heat dissipation required of 23 may exceed the rated 21, 29 and 50,

'sequently parasitic oscillations of such frequency, caused by parallel operation of the devices l and I6, cannot occur.

If the unsuppressed parasitic oscillations are caused by push-pull, or balanced, operation of the devices l5 and IS, the resistances 27 and 29, connected in series with the inductances 26 and 28, preferably at the grounded ends'thereof, and the resistance 50, may be adjusted in value until such parasitic oscillations cease. As is the case with the resistances 20 and 23 and the inductive conductors 2| and 24, the proper values for the resistances Hand 29 cannot be predetermined, so far as is now known. It is preferred to determine the values of the resistances 21 and 29 by trying initially resistances of some small value,- such, for example, as 50 ohms, andincreasing the value thereof in small steps until such parasitic oscillations cease. It has been found in typical amplifiers that resistances 21 and 29 may have values, for example, of the order of 300 ohms, while resistance 50 may be of the order of 20 ohms. Such typical amplifiers were operated at frequencies of about four megacycles, and parasitic oscillations of the order of forty megacycles were suppressed by the use of applicants teachings.

While I have shown and described a particular embodiment of my invention, it will be obvious to those skilled in the art that changes and modifications may be made without departing from-my invention in its broader aspects, and I, therefore, aim in the appended claims to cover all such changes and modifications as fall with-' in the true spirit and scope of my invention.

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

1. In combination; an electron discharge device having ananode, a cathode, and a control electrode, an input circuit extending between said cathode and control electrode, an output circuit including a source of operating potential and a conductive impedance extending between said anode and cathode, said circuits being arranged for operation at a predetermined frequency and having undesired feedback therebetween which tends to produce self-oscillation at another frequency through said device, and

means for by-passing currents of said other frequency around said conductive impedance through an impedance sufficiently low that said self-oscillation is suppressed at said other frequency, said conductive impedance and said bypassing means having substantial impedance at said, predetermined frequency.

2. In combination, an electron. discharge device having an anode, a cathode, and a control electrode, an input circuit extending between said a cathode and control electrode, an output circuit including a source of operating potential and a conductive impedance extending between saidanode and cathode, said circuits being ar- 1 ranged for operation at a predetermined frequency and having undesired feedback therebetween which tends to produce self-oscillation at.another frequency through said device, and means comprising a circuit series resonant at said other frequency for by-passing currents of said. other frequency around said conductive impedance.

3. In combination, an electron discharge device having ananode, a cathode, and a control electrode, an input circuit extending between said cathode and control electrode, an output circuit extending between said cathode and anode, means for applying a voltage from said anode through a portion of said output circuit to said control-electrode in such phase as to suppress self-oscillation of said device at an undesired frequency, said means being ineffective to suppress such self-oscillation at another frequency, and means in the remaining portion of said output circuit to reduce the impedance thereof at said other frequency to a value sufliciently low that self-oscillation at said other frequency is suppressed. I i

4. An amplifier comprising a pair of electron discharge devices each having an anode, a cathode, and a control electrode, input and output circuits respectively connecting said control elec-- tion of said input circuit in such phase as tosuppress self-oscillation of said amplifier, said means being ineffective to suppress such selfoscillation at undesired frequencies, damping means in each of said portions of said input circuit to produce critical damping in the circuits of said amplifier for one of said other frequencies, and means to reduce the impedance of said common branch of said output circuit at another of said other frequencies to a value so low that self-oscillation of said devices at such other frequency is suppressed.

5. A balanced amplifier having an output circuit and two input terminals, and a grounded terminal of intermediate potential, a tuned circuit connected between said input terminals and having an intermediate potential point, a resistance connected from said point to said grounded terminal, said amplifier having a tendency to self-oscillation by reason of feed-back from said output circuit to said input terminals at a certainfrequency, and an impedance including a resistance connected from-each of said input terminals to said grounded, terminal, said three resistances being of-such size and so related to each other as to produce critical damping in the circuits of said amplifier at said certain frequency, thereby to suppress such self-oscillation.

DONALD E. wA'rrs.

Images