US1800901A - Vacuum-tube circuits - Google Patents

Description

April 14, 1931. E PETERSON ET AL V 1,800,901

` VACUUM TUBEl CIRCUITS Filed Nov. 50. 192'7 4 Sheets-Sheet l 4 {(lgll llllll 4a' 57 39 mlm' 30 sa@ 34 n 3, 35 t HG2.

40 60 B0 10o 12a uo 150150 20o 220 40 Vous Fmrf PorfA//AL 'g ff=z2AMPEREs Arm/MY Arm/Mfr prl 14, 1931. E. PETERSON ET AL VACUUM TUBE CIRCUITS Filed Nov.

, 1927 4 sheets-sheetl 2 o om 0 98 j ma fume-NE Pfff/95M HERBERT fl/A N5 pril 14, 1931. E PETERSON ET AL 1,800,901

VACUUM TUBE (II'CUI'I'S` Filed Nov. 30. 1927- 4 Sheets-Sheet 5 Har?.

Afm/MY April 14, 1931. E. PETERSON ET AL VACUUM TUBE CIRCUITS Filed Nov. 30. 1927 4 Sheets-Sheet 4 /GENE Pfff/@50N Hmmm/.3 [VA N5 Pound Apr. 14,1931.`

, lJNrrsl) sTATEsfPATENr OFFICE :vom farmer. or nw roux. u. r aim maaar r. avants, or miaou. wxscozmr, wenn 'ro nu. rammron Luonroans. mooaronrxp, or m man ir. r., a oonron'rmr or nw You Y y vacunar-rm mourra annua. ma :maar ao, im. um: n. mm.

i This invent-ion relates to space discharge vacuum tubes, articularl to methods of operation thereo wherein t e relation existin 'between4 the fundamental frequencies an the products of modulation in the output circuit of a vacuum tube can be regulated.

Heretofore in modulatin systems employthe productionl ing vacuum tubes in whic of the modulation component was considered lo as occurring in the plate circuit, the modulating action has been attributed to a nonlinear relation existing between either the plate volta volta an late current. In such cases the mod ation been considered as due entirely to the resistance variations in the s ace path of the tube, conveniently terme R modulation.

When a vacuum tube is used as an amplia, er its operation is usually restricted to the so-called straight portion of the plate current-grid voltage characteristic curve. If the curve were actually strai ht it is apparent that distortionless ampli cation free from the eects of modulation would result. Such an ideal condition has never been produced in a. single vacuum tube amplifier although in s ecial circuits it has been found possible to e iminate certain of the modulation products.

ln dealing with the operation of a Vacuum tube both as an amplifier and as a modulator, the amplification factor p. has generally been considered constant although in ractice a variation has been found to exist opending on changes in the values of 'd and plate yoltage which, however, it has en the practice to disregard. V

This variation in the amplification factor has been found to be responsible for additional products of modulation in the vacuum tube output circuit independent of the R0 modulation. modulation.

In accordance with an object of the present invention the modulation current present in the output of a vacuum tube is controlled by regulating the amplitude and phase of the R0 modulation and the p modulation.

I0 Another object of this invention is to conand plate current or thegrid This action will be termed ,L

trol the ratio ofthe modulation products to the' fundamental product in the out ut of the space dischar vacuum tube y the proper selection o values ofgrid biasing potential, plate potential, plate circuit resistance and alternating grid potential.

Another object of this invention is to determine for any Aspace discharge vacuum tube the conditions of operation under which the ratio of second orden modulation currents to fundamental currentwill attain its smallest possible value.

Still another object of the invention is to operate a vacuum tube so that the second and third order modulation products are reduced or eliminated.

` Another object is to facilitate the selection of a vacuum tube together with itsconditions of operation which will operate at prescribed power and modulation levels with a minimum expenditure of direct current power.

Another ob'ject of the invention is to approach an ideal amplifier.

Still another object of the invention is to increase the ower output of a vacuum tube am lifier wit out increasing distortion.

feature of the invention involves regulating the value and phase of the components 3f-modulation in a vacuum tube due to the a modulation and Ro modulation with respect to each other whereby a desired ratio of modulation products to fundamental product is obtained.

Another feature .is the use of a four-element vacuum tube with the circuit potentials adjustedto such values that the p. modulation and R0 modulation are equal and of opposite hase so that an am lified wave substantialy free from distortlon by modulation products, results.

Heretofore characteristic vacuum tube equations expressed the output current of a vacuum tube as a function of grid and late grid circuit. When only one variable is considered, namely, grid potential, the tube characteristic equation expressing plate current as a function of the variable grid potential is a power series involving this variable.

For many practical purposes analysis by the above method is fairly satisfactory but where the requirements with respect to modulation are particularly severe as, for instance in multi-channel repeaters, the assumption of the constant ,a may not be sufliciently accurate. In fact it has been shown that in the normal operation of the vacuum tube ampliiier where the grid potential is always negative and where the greatest applied negative potential does not exceed the most negative end of the grid potential plate current characteristic curve, Variation in ,i has been variations in the amplification factor a. The

accuracy of the theory is substantiated by the close agreement of the calculated with the experimental results.

In the practice of the invention it is first necessary to consider the expressions affecting the values' of the various roducts i of modulation. As indicated herelnbefore,

products of modulation refer to the currents of combination frequencies which are present in the output circuit of a'vacuum tube due to variations in amplification factor a (a modulation) and in the internal plate resistance vof the tube R0 (Ro modulation). The new expressions developed by applicants giving coefiicients of the modulation products are derived by representing the characteristic tube equation as a double power series and have been developed in terms of the amplification factor a, the internal resistance of the tube R0, and their differential parameters.

Where 01 01 M LRO J2- m-cuR-Raw, R0013EA (2) In these equations E denotes peak value of the E. M. F. impressed upon the grid, J1 and J2 the peak values of the fundamental and second harmonic currents, p. the am lication factor at the operating point, o the corresponding value of the internal resistance, and R the external resistance in the plate circuit.

-The derivation of the above relations involves the consideration of two variables, namely p. and R0. The same treatment may be extended to tubes having characteristics depending upon a larger number of variables. Thus a four-element double grid tube characteristic may be expressed by a triple power series, etc.

If the potential of one of the rids of a four-element tube be held constantt e relation for three-element tubes may be used so that the expression given for C1 and C2 will apply; Four-element tubes will be considered in a later part of this specification.

From the relations given in Equations '(1) and (2) the ratio of second harmonic to fundamental current in vacuum tubes and the behavior of this ratio as a function ofthe four independent variables, namely, (a) alternating grid potential or grid swing, (b) plate circuit resistance, c) grid potentie, (d) plate potential, canbe etermined. From these results an amplifier tube which is required to meet prescribed conditions in a. given circuit may be selected.

A better understanding of the invention maybe had from the following descri tion inconjunction with the accompanying rawings of which Fia. 1 shows a three-element vacuum tube used in an ordinary amplifier circuit. Fig. 2 shows a special four-element vacuum tube used as an am lifier in accordance with the invention. igs. 3 to 14 are curves showing tube characteristics and experimental results which will be referred to as the description of the invention proceeds.

The relations givenby theEquations (l) and (2) will rst be considered in connection with three-element vacuum tubes connected in the simple amplifier circuit such as that shown in Fig. 1. Herein the three-element vacuum tube 10 has an output circuit associated therewith which can be traced from the plate l, to a source of plate potential which is here represented by battery 6, through the resistance 5, the cathode 3, back to the plate` 1 through the space path of the tube. The input circuit can be traced from the grid 2 to the cathode 3 through a suitable means for impressing an alternatin wave upon the input circuit such as tran ormer 4. Some means such as a battery 9 is provided for maintaining a biasing potential upon the grid. An alterynating wave is provided in any-well known manner from a source 8.

Since the various modulation products of -a given order are related in a simple manner, the present discussion as regards modulation tetona: v 3` products, will be limited to the harmonics of a single impressed frequenc 4. It is assumed that the plate circuit impe ance R is a pure resistance which is the same at all fr uencies, that the late current is never re need to zero and t at the grid current is always ne ligihle.

order that the invention may be more readily understood the equations which illustrate the operation of the vacuum tubes will first be discussed for various conditions of operation after which the methods of selecting the correct circuit`parameters for ob- 'talnin the desired cond1tions will be illustrated y means of e rimental data. Conditions of operation or which the ratio of the second order modulation product to fundamental attains its smallest possible value will be discussed rst. These conditions have been determined for two distinct cases. InA

the first case the load resistance is restricted to equal the tube resistance whereas in the second case no restriction is placed on the load resistance.

Letting A denote the ratio of second order modulation roduct to fundamental it follows from nations 1 and 2 given above 5R., E ,j (3) For the case in whichY the two impedances match the Equations 2 and 3 become:

. una in n C aE, 4R., aulne., (4) 5# y. 5R() bauma u 5) From Equation 1 the output fundamental power is seento be y Referring lto Equation (2) it is seen that the determination of the modulation coeicient C; requires the evaluation of three partial derivatives in the general case, for which RfRo. These derivatives may be obtained graphically from the tube characteristics, practical examples of which are shown for one type of tube in Figs. 3 to 5. From lation to fundamental coeicient increases rapidl as the negative grid potential is increa 'This e ect is less pronounced at the higher plate potentials, due to the greater constancy of that component of modulation which results from the variation of the internal resistance of the tube. This is evident from Fig. 5 for if the variation in the slope of these curves at 240 and 120 volts late potential is compared; at 120 volts the s opes are more variable with grid potential than at 240 volts. It is also seen from the curves of Fig. 7 that for a constant value of funda- `mental the modulation products will decrease as the plate potential yis increased. For example, ifthe tube is being operated at a negat1ve grid potential of 24 volts and plate otentia of 120 volts, the curves show that i the late potential is increased to 24() volts and t e negative gridpotential to 65 volts, the fundamental output (for the same alternatin grid' voltage) will be left unchanged, ut the modulation roducts will be reduced b about 7 T. U. he curves of Fig. 6 and ig. 8 show similar results for two other values of load resistance. Similar conclusions apply to other tubes-as may be ascertained by determinin the tube characteristics andY computinug t e values of the fundamental and mod ation coeicients Cl and C2 as in the case herein described.

It is interesting to note that in the case -of high external resistance iii the circuit of the ,tube under consideration the second order product of modulation will pass through aminimum value as the negative id potential is increased. This is shown 1n Fig. 6. The explanation is obtained by an examination of the terms in the relation for C, as given in Equation (2) From Fig. 4 it is seen that the rst term of the equation .a V aE.

is practicall constant as the id'I potential is varied. igs. 3 and 5 show t at 2 I both increase as the ne ative grid potential increases. The effect o these characteristics upon the modulation products is seen in the relation -for C2, :for as the negative Agrid potential is increased lthe component of modulation due to the a variation will become smaller while the component due to Ro lll@ Alt ough 'an examination of a large variety of tubes disclosed that the ratio of second order modulation products to fundap families of curves such as those of Fig., 6, it-

is possible to select a tube, together with its operating potentials and load resistance, which will yield a given power output with a minimum amount of modulation. The problem is somewhat simplified if the alternating grid otential is supposedly fixed. This may be illustrated by means'of the following simple exam le:

Suppose that wit a given grid swing of 40 volts (peak value) it is required to select a tube, together with its grid potential, plate potential and load resistance which will yield an output level of 0.2 watt with a minimum amountof second harmonic. It is first necessary to examine the output characteristics of a number of tubes to determine which one for the given grid swing must "be selected in order to confine operation to the characteristic and prevent the grid from becoming positive. For the sake of convenience, the re uiiements of the problem are such as will a apt themselves to the characteristics of the tube already given. It is to be noted however, that the method is adaptable to any type of tube.

First, suppose the tube to work into a re' sistance of 2500 ohms. In this case it is evident on solving Equation (6) for C1 that this coefficient must be 0.316 X 103. Referrin to Fig. 7, it is seen at once for this value o C1 the ratio of C2 to C1 will be a minimum and e ual to 5.7 10a when a plate potential o l240 volts and a negative Ygrid potential of volts are used. In a similar way it is found from Fig. 8 that when anexternal resistance of 6,000 ohms is used a plate potential of 240 volts and a negative grid potential of volts should be used, in which case the ratio of C2 to C1 is 3.68X10". Finally, if an external resistance of 15,000 ohms is used, it is found from Equation (6) and Fig. 6 that a plate potential o 240 volts and a negative grid used, in which case the ratio of C, to C, is 0.8X 10. 4I-Ience, in order to obtain an output level of 0.2 watts with a grid swin of 40 volts, a tube of the type discussed shoul be used, the plate potential being 240 volts, the negative grid potential 66 volts, and the load resistance 15,000 ohms. For these values the reduction of the second harmonic below the 'fundamental is found from Equation (l) to be 36 T. U. This ratio could probably be potential of 66 volts should be.

improved still more by increasing the load resistance and decreasing the negative 'd otential, as may be determined by calcu atin the modulation coefficient for greater va ues of load resistance.

The foregoing relations will now be considered with respect to four-element (double gip-id) tubes. It has already been pointed out t at if one of the grids is maintained at a constant potential the tube characteristic equation can be expressed by a double ower series and the expressions for the modu ation coefficients C, and C: as given above will apply. A g

t may be shown that the modulation products may be reduced to a minimum and in fact may be made to vanish for all practical urposes in a tube having two grids, one of w ich is maintained at a constant positive potential with respect to the cathode, the three remaining elements being used as an ordinary vacuum tube.

Such a tube is shown in Fig. 2, which illustrates the tube as it would be used in an amplilier circuit. Therein tube 30 is a four-element vacuum tube having an anode 31, a cathode 34 and two grids 32 and^33s One grid 33 is maintained at a constant positive potential with respect to the cathode 34 b means of a source of potential 38. In al other res ects the circuit is similar to that shown in ig. 1. In order to avoid confusion the constant otential rid will be referred to as the fixe grid an tlie other one as the var ing Grid. The voltage u on the fixed gri will e denoted by E whi e alternating voltage impressed upon the varying grid will be called E.

In a manner similar to that used in contential being held constant. At other values of plate potential the amplification factor is slightly lower but the variations with varying grid and fixed grid potential is very nearly the same.

Fi s. 10 and 11 show the internal resistance Ro otted as a functionof platepotential for various values of negative bias upon the var ing grid, the potential upon the fixed gri Ef, bein maintained at 40 volts. These curvesarea o a typical grou the corresponding groups for ower fixe grid potentials bein very similar especially as to the changejko slope from positive to negative as the varying grid potential is increased.

From these results fundamental and second order coefficients ma be calculated by means of Equation 2 in which for double grid tubes E, is given as Eu. If the load resistance is calculated second order coefficient as a function of the varying grid voltage; for various values of plate voltage, the ed grid -potential being the same for all cases is shown b Fig. 12. These curves show that for eachV p ate voltage there is at least one value of varying grid voltage for which the second order coefficient vanishes and in one case there are three such points. It can be shown that the curves for other values of fixed grid voltage, other conditions remaining the same are all similar to these. strate that the second order currents should reverse in phase as the varying grid voltage is varied through one of the vanishing oints. This is due to the variation in the actors expressing the p. modulation and the R0 modulation since it is evident from Equation (2) that, as the smaller component becomes eqlilial and then exceeds the other, the sign of t e coeiiicient C, changes denoting a change in phase of the resultant modulation current.

As pointed out above Fi 12 shows that for one set of circuit con itions there are three values of grid potential at which the second order coeicient becomes zero. The curve of Fig. 13 shows the calculated relation of the first harmonic current to the varyin grid swing as determined from Equation (1 for the same circuit conditions. Likewise for the same conditions a curve showing the measured relation of the third harmonic and varying grid .swin is shown in Fig. 14. A comparison of these t ree curves representinc fundamental and the second and third o er roducts of modulation reveals that when t e second order product is zero at one point the third order product is practically zero while the fundamental is almost at its maximum value. It can be shown by comparison that the calculated and measured values of current for the above relations are in close accord..

To summarize it can be shown from the expressions for the coefficients of the modulation products that their value can be regulated by properly selecting the circuit parameters, that having determined the tube characteristic curves, the operating points which give a minimum value of second and third order modulation products can be selected, and in some type of vacuum tubes the higher order modu ation product-s can be made to vanish and in one t pe of tube used, at the point of maximum undamental output.

These curves demon--l in connection with particular types of tubes, it is ,nevertheless capable of broad application and is to be limited only by the scope of the appended claims. l

What is claimed is:

1. In a space discharge circuit, a space discharge device having electrodes, an input and an output circuit associated with said electrodes, a resistance lin said output circuit, means for applyin a'wave to the input circuitto be repente means for appl g operating .potentials to the input an out ut electrodes, the magnitudes of the operating potentials being proportioned (for a ven peak amplitude of the input wave) wit respect to each other to control the ratio of output modulation component to output fun- Although this invention has been described damental component and both` said magnitudes being proportionedwith respect to the external output resistance to control the output of usefu components.

2. In a space discharge circuit, a space discharge device having electrodes, an input and an output circuit associated with said electrodes, a resistance `in said output circuit, means for applyingpotentials to said electrodes, the values of said resistance and said potentials being proportioned so that the ratio of the modulation products to the fundamental products is a minimum.

.3. In a space dischar e circuit, a space discharge device having e ectrodes, an input and an output circuit associated with said electrodes, a resistance in said output circuit, means for ap lying potentials to said electrodes, the va ues of said resistance and said lpotentials beinulproportioned so that the ratio of the mo ation products to the fundamental products is zero.

.4. In a space discharge circuit, a space discharge device having electrodes, an input and an output circuit associated with said electrodes, means for regulating otentials applied to said electrodes, means or appl ing a waveof alternating potential to said input circuit to be repeated, means for repeating said wave in said output circuit togather with components of modulation-resulting from the changing amplification factor and the changing internal resistance of said discharge device, said regulatingmeans causing a predetermined relation to exist between said varying ampliiication modulation component and said varying internal resistance modulation component.

5. In a space discharge circuit, a space discharge device having four electrodes, means for applying potentials to said electrodes, said potentials being proportioned with respect to each other at such values that modulation due to the varying amplification factor of said space discharge device substantially balances the modulation resulting from izo the varying plate resistance of said space disother in'controllable amount.

I7. In a space discharge circuit, a space discharge device havin an anode, a cathode, a grid electrode an a control electrode or net, an output circuit including a resistance associated with said anode, means for applying potentials to said electrodes, means for applying a wave of alternating potential to said control electrode to be repeated, means for repeating said wave in said'output circuit together with components of u modula-V tion and Ro modulation, the values of said resistance and said potentials being regulated to predetermine the value and phase relation of the n and R0 modulation.

8. The method of operating aspace discharge device which comprises proportionng its circuit parameters to eect a predetermined relation between the p. modulation components of current and the R0 modulation components of current.

9. The method of operating a space discharge amplifier to prevent distortion which comprises generating a component of modulation due to the variable ampliiication factor p and another component of modulation due to the variable space resistance R0, and regulating the value and phase of the even order modulation components so that they ar substantially equal and opposed to each Qt er.

Y charge amplifier which has electrodes and input and output circuits associated with said electrodes which comprises applying a wave of alternating` potential to said input circuit to be amplified, roportioning the values of operating potentials applied to the input and output electrodes relative to each other to give substantiall minimum modulation products relative to undamental products and at the same time proportioning the values of both of said operating potentials with respect to the external output resistance to give the required power output.

- 11. The method of operating a space discharge device having electrodes, input and output circuits associated therewith, said output circuits havin resistance which conirises controlling t e ratio of output moduation iroducts to output fundamental component y proportioning the input yand output 10. The method of operating a space disl operatin potentials with respect to each other an vat.the same time controlling the useful power output by proportioning both ofsaid operating potentials with respect to the external output resistance.

12. the operation of a double grid tube in which a fixed potential is applied to one grid and a wave to be reeated is applied to the other grid, the metho of reducin in the o ut oing circuit both even and od order mo ulation products which comprises roportioning t e operating potentials a p d to the tube electrodes with respect to the asttached output impedance to values corresponding su stantially to the minimum points on the even and odd order modulation characteristics and at the same time to a high value of fundamental out ut.

In witness whereof, we ereunto subscribe our names this 22nd and 26th days of November, A. D. 1927.

EUGENE PETERSON. HERBERT P. EVANS.

4 ios

Images