US1795397A - Directionally-selective radio receiving system - Google Patents

Description

March 10, 1931. R. s. HOYT 1,795, 7

DIRECTIONALLY SELECTIVE RADIO RECEIVING SYSTEI Filed Dec. 29. 1927 s Sheets-Sheet '1 INVTOR z. sip w BY W Y ATTORNEY March 10', 1931. R, s. HOYT 1,795,397

DIRECTIONALLY SELECTIVE RADIO RECEIVING SYSTEM Filed Dec. 29. 1927 3 Sheets-Sheet 2 INVENTOR 1 R. S. floyt fiwpn ATTORNEY March 10, 1931. R. s. HOYT DIRECTIONA LLY SELECTIVE RADIO RECEIVING SYSTEM Filed Dec. 29. 1927 3 Sheets-Sheet 3 INVENTOR 5.50

ATTORNEY Fatentecl Mar. 10, 1 931 UNITED STATES PATENT OFFICE RAY S. HOYT, OF RIVER EDGE, NEW JERSEY, ASSIGNOR T AMERICAN TELEPHONE AND TELEGRAPH COMPANY, A CORPORATION OF NEW YORK DIRECTIONALLY-SELECTIVE RADIO RECEIVING SYSTEM Application filed December 29 1927. Serial No. 243,380.

It is among the objects of my invention to provide new and improved apparatus and a corresponding method for receiving radio signals from a desired direction with avoidpance of interference from other directions.

Another object is to provide a radio receiving system that shall be adjustable or flexible in design so as to receive efficiently from one particular direction with substantially no interference from another direction, or directions, Another object is to adjust the design so as to reduce the effect of random interference over a considerable range of directions other than the desired directionof retem of receiving wave antennae adjustable or flexible in design for one or more of the purposes stated above. Still another object is to provide a. system of elements comprising two parallel wave antennae in staggered and spaced relation, with the dimensions and other determining factors of the system so assigned that there will be good reception in one desired direction and exclusion of interferencein another desired direction or di eetions. All these objectsand other objects of my invention will become apparent on consideration of a limited number of specific embodiments of the invention which are disclosed in the following specification taken with the accompanying drawings. It will be understood'that the following disclosure relates to these examples of the invention and that the scope of the invention will be indicated in the appended claims.

Referring to the drawings, Figure l is a diagrammatic plan view of .a system of two parallel staggered wave antennae embodying my invention; Fig. 2 is a corresponding diagrammatic elevation of one of these two wave antennae; Fig. 8 is a diagrammatic plan view with certain dimensions and other data indicated as a basis for discussion; Fig. 4 is a diagrammatic plan showing one way of combining the currents received over the several antennae for the operation of a single receiver; Fig. 5 corresponds with Fig. 4: but has the receiver at the front end instead of at the back end; Fig. 6 is a diagram ception. Another object is to provide a sys-.

pensation currents taken from opposite ends of each wave antenna; Fig. 6a is a diagram showing compensation for F 1g. 5 in the mannor of Fig. 6; Fig. '7 1s a d agram showing another manner of effecting compensation as applied to a system like that of Fig. 4:;

Fig. 7 a shows the same manner of effecting compensation as applied to the system of Fig.

5; Fig. 8 showsstill another manner of effecting compensation in a system such as that of Fig; a; Fig. 8a shows the kind of compensation of Fig. 8 as applied to a system like that of Fig. 5; Fig. 9 is a curve diagram giving the measure of advantage forcertain particular examples of my invention as f xed for example by certain definitely assigned parameters, and Fig. 10 is a similar curve diagram in which the parameters are the same except that the direction for null reception is different.

A wave antenna may be a-comparatively long horizontal conductor extending in a straight line at a moderate height and in a direction nearly the same as that in which maximum intensity of receiving is desired.

The elementary theory of such a wave antenna is presented in the papers by H. H.-

Beverage et 'al., in the Journal of the A. I. E. E. for March, 1923' (page 258), and succeeding issues. My invention may be embodied in a pair of wave antennae inspacecl parallel relation and staggered; that is, one of them displaced along the direction of their length. Two such wave antennae are shown in Fig. 1, adapted for reception of radio waves incomingon the left. These two antennae are indicated as 1 and 2, and corre sponding parts and data associated therewith are distinguished by the subscriptsl and 2.

I shall employ the term transducer inthis specification in its usual meaning of any assembly of apparatus having a set of input terminals to which input electromotive forces are applied and a set of output terminals for output currents, these currents being a function of the said electromotive forces. Under this broad term may be mentioned as 'examples a transformer, a rotary converter, an amplifier, an attenuator, a transmission line,

for a system such as that of Fig. 4, with coma phase shifter. Some authorities-use the term quadripole instead of transducer.

In this specification I use certain characters, such as T, sometimes to represent a piece of apparatus, say T for a transducer, and sometimes to represent an associated numerical measure such as the complex numbcr value for the transfer factor of the transducer. In each instance the sense intended will be apparent from the context.

As shown in Fig. 1, combining transducers T and T are situated at the back ends of the antennae 1 and 2. Each transducer enables the ratio j/I of its output current to its input current I to be adjusted to any value as regards magnitude and phase. The antenna output currents I and 1 are evidently the transducer input currents. The transducer output currents j and 7' after having been adjusted to any desired values by means of the respective transducers T and T are finally added together in some such manner, for instance, as indicated by Fig. 4:. It will be found that either one (but not both) of the transducers can be proportioned arbitrarily as regards its transfer factor T or can even be omitted or be combined with the other; but for symmetry both are retained in all of the figures.

Referring to Fig. 3, 8 denotes the length of each wave-antenna, p, the staggering of the wave-antennae (that is, the amount by which antenna 2 is displaced longitudinally with respect to antenna 1), g the spacing (transverse separation) of the wave-antennae, and r the direct distance between the two front ends (and hence between the two back ends). VVave-antenna 1 is taken as the reference axis, its front end 0 as origin of coordinates, and the direction from front to back as positive direction. Angles are measured counter-clockwise from the reference axis (antenna 1). 8 denotes the angle from the reference axis to the line 7 In Fig. 3 there is shown a train of plane electromagnetic waves (radio waves, or space waves) incident at any angle 0, measured from the reference axis to the direction of propagation of the radio waves along the earths surface. f denotes the horizontal component of the electric force of the radio waves at the front end of antenna 1, and f the corresponding simultaneous value at the front end of antenna 2; these electric forces are along the direction of incidence (that is, the direction of propagation in space).

As explicitly indicated by Fig. 1, each antenna is terminated at its front end in an arbitrary impedance 1V, and at its back end in a transducer which presents to the waveantenna an arbitrary impedance Z. For simplicity, the front-end terminal iinpedances are not shown in Fig. 3, but they are to be regarded as present there. (In most practical applications, Z and V would be made equal to the characteristic impedance K of each wave-antenna, for simplicity and also for securing desirable directional characteristics for the individual wave, antennae. I denotes the output current of antenna 1 and hence the input current of transducer T and denotes the corresponding output current of transducer T and similarly for T and j Evidently these currents are functions of 6; this fact will usually be explicity indicated by employing the functional symbols I (6),l (6]),j (6),j (0) ;thus the symbols 1 ,1 j y' are to be regarded as abbreviations of the corresponding functional symbols. The fact that these currents depend on the incidence angle 6 of the radio waves, may be indicated by saying that these currents iave directional properties or characteristics.

The transfer factors T and T of the combining transducers are to be regarded as unlike, in general. They are merely the current ratios defined by the equations Z 1 .7: a 172:.72(0)/I2(6) 7 which, of course, are independent of 0.

Finally, the received currents j (49) and i (0) are regarded as being directly combined by simple addition so that, if J (6) denotes the resultant current, then U =7'1( +7 0 tions 2w) an/ a Since the two antennae are alike and are terminated alike (as regards the impedances W and Z), it is evident that y denoting the propagation constant of the radio waves along their direction of propagation over the earths surface (per unit length) F 3 shows that Hence, by Equations (8), (7), (8), (9), the resultant received current To establish a a 1 11 a The function G(6) will be termed the group factor (or, more fully, the group directional factor). This term is adopted because G(6) does not depend on the wave-antennae themselves, but only on their grouping as represented by the relative positions of their front ends (or of their backs ends), as specified by r and 8, and on the ratio T /T of the transfer factors of the combining transducers.

The functional notation G(6) denotes that 6 is regarded as the independent variable and hence that the other quantities are regarded. as parameters. It may be observed, however, that 68 could be taken as the independentvariable, which would correspond to taking the line along 1" as reference axis and would apparently reduce the number of independent parameters by one; but this reduction is only apparent, because, when the line along 1" is taken as reference axis, the parameter 8 then appears in the formula for the directional characteristic of each wave-antenna alone, whereas it is absent from that formula when the direction of the wave-antenna is taken as the reference axis. On the whole it seems best to take the wave-antenna as the reference axis, so that 8 then occurs as a parameter in the formula for G(6).

If Y (6) denotes the resultant directional admittance of the array of the two waveantennae, combined as aforesaid by means of the transducersT and T and referred 'to the electric force f so that Y (6) is defined by the equation we ==Z(6) fb then, by (10),

Y. (6)=T Y(6)G(6). (14:) (Subscript a denotes array)..

The directional selectivity of a single wave-antenna, sa antenna 1, with respect to any arbitrary re 'erence value 6 of 6, is represented by the directional ratio p 6) defined by the equation P l '")|a a polar graph of which is the so-called polar diagram representing the directional characteristic of a single wave antenna. (Usually 6,=O). From (15), (4), (5), (6) it is seen that P1( P2( )=P( )?l )i- Similarly the directional selectivity of the array of the two wave-antennae is represented by the resultant directional'ratio ,,(6) defined by the equation pa e/ e l; I whence by (10) and (16) successively, it is seen that q 1 (0) 0 0 6( Pa(6)" W P( )m i where 6) =p (6) =p (6) is the directional ratio of each wave-antenna. By introducing the group factor ratio 0(6) defined by the equation l )b (18) can be written .as they are sufficiently known in the art; reference may be made to the paper by Beverage et al. mentioned heretofore, where it is shown that the directional ratio (6) of a single wave antenna can be made decidedly directional by suitable choice of its length; an alternative exposition of the fundamental theory and formulas for a single wave antenna Willalso be found in the article by Carson and Hoyt, Propagation of periodic currents over'a system of parallel wires, in the Bell System Technical Journal for July,

. 1927, pages 532 to 535. The system disclosed in the present specification in accordance with my invention, affords improvement in that the resultant directional ratio p (6) ,of the spaced staggered array of the two waveantennae can be made decidedly more directional than the directional ratio 0(6) of each wave-antenna alone; this improvement results from the presence in Equation (20) of the group factor ratio 0(6) which, as will be shown in the present specification can be given particularly desirable directional prop erties by suitably proportioning the array of the two wave-antennae as regards the staggering 10, spacing g, and ratio U=T /'I characterizing. the combining transducers, so that the product 0(6),)(6) will be decidedly more directional than is p( 6) alone.

By reference to Equation (11) it is seen that the group factor ratio 0(6) defined by (19) represents the dependence of the directional selectivity of the system constituting my invention, on the staggering 10 and spacing 9 of the array of the two wave-antennae (or, in other words, on 1' and 8), and on the ratio T /T of the transfer factors of the-combining transducers. Therefore the properties of the group factor ratio 0(6) will now be set forth in some detail. This will be accomplished by studying the group factor GM) since (TM) is proportional to GM)[ by Equation (19); it is seen that aM)=l 5 when 0=6 Vith a View to writing Equation 11 for GM) in more useful and more signi cant forms we introduce the ratios P, Q, R, defined by the following equations, in which 19 denotes the wave length of the radio waves in the direction of propagation along the earths surface:

and from Fig. 3 we note that R=P/cos 8=Q/sin S. (24:)

so Also, we let I denote the propagation constant of the radio waves per wave length, so that a therefore denoting the attenuation con- 25 stant of the radio waves per wave length. Furthermore, we let U denote the ratio of T to T so that T /T ==U=[ U] 6 (26) "u, therefore denoting the phase-angle of U.

By means of the substitutions (23), (25), (26) Equation (11) for GM) becomes ag denoting angle of or, more fully 45 phase angle of.

Since cos M8)=cos (8-6), it follows from (27) that Similarly Equation (29) shows that these lines, but the major lobe of its polar diagram may be roughly symmetrical with respect to some intermediate line. It may be advantageous to so choose the direction of the array of wave-antennae that the desired signal will come in approximately along this intermediate line, so that the major lobe of the polar diagram will be at least roughly symmetrical with respect to the direction of the desired signal. But, in many cases, considerable dissymmetry may be desirable (to minimize the effects of static and other inter ference) such desired dissymmetry can be at least partially secured by suitably choosing the direction of the wave-antennae.

The effect of changing 8 to B-t is to rotate the polar diagram of through the angle 5. This follows from Equation (27), since (98 is unchanged when 0 and 8 are each increased by 5. In particular, changing 8 S rotates the polar diagram through the angle 25 (since 8=8-28).

Thus far we have dealt with the group factor GM) in a general manner. 'We now proceed to show how the system can be so proportioned that the absolute value GM)] of the 'roup factor will have desirable directional properties. (We are not concerned with the phase angle of the received current JM), and hence not with the phase angle of GM).)

On referring back to Equations (28) and (29), it is seen that jGM)[ depends on no less than five parameters, namely the set a, U I, u, R, 8 or the equivalent but more convenient set or, IUI, w, P, Q, (since P=R cos 8 and Q=R sin 8). The parameter 0:, namely the attenuation constant (per wave length) of the radio waves in space, is of course supposed to be known, at least approximately. The four remaining parameters U a P, Q are at our disposal and, in fact, are to beso chosen as to yield so far as possible the desired directional properties for |GM)[. The considerations leading up to the evaluation of these parameters will now be set forth.

In most practical applications it is desirable that be Zero, or at least very small), at some specified value of 6, say 6. Therefore the first condition to be imposed will be that GM)1=O. It will be found that this condition su'iiices to determine both 1U) and u, that is, the absolute value and the phase angle of U=T /T For securing approximately maximum sensitivity at a particular value of 0, say 6 (usually 0 0), it is evidently necessary that the output currents j M) and shall not be greatly out of phase for 6 6. Of course the sensitivity is a maximum when thesetwo currents are exactly in phase; moreover the practical operation of the system is considerably facilitated if they are in phase. From Equation (32), it is seen that the condition that they be in phase for 6; 6 is that i (6) O, (6) being the angle by which j (6) leads 7' (6). However, a value of somewhat different from Zero may in some applications lead to appreciably better directional characteristics for Consequently (6) will be regarded as specified, but not necessarily equal to zero. Specification of the values of 6 and (6) enables either P or Q. to be eliminated; it will be found advantageous to eliminate P rather than Q, for a practical reason appearing later.

As a result of applying the foregoing considerations it will be found that G(6)i is expressed in terms of the parameters Q, 6, 6, (6). For any particular application, the most suitable values of these parameters can best be determined by plotting sets of curves of r(6)=| G(6)/G(6 )l for various sets of values of these parameters, as described ne'ar the endof this specification and illustrated by Figs. 9 and 10.

Having thus outlined the proposed steps in theevaluation of the parameters fixingthe system, we shall now indicate the details of' those steps, and the resulting design-procedure and design-formulas.

Before applying the condition I G(6) ]O, it may be noted from Equation (33) that if G(6)-iszero for 6 6 it is also zero for 6 6 such that see =28, (37) p but (as shown in connection with Equation (50)), as long as o: is not zero, there are no other values of 6 for which G is zero. In general, 6' and 6 are distinct; but they are coincident when 6 is chosen equal to 8, for then 67 =8 by Equation (37).

From Equation (27), the necessary and sufiicient condition that G (6) be zero for any specified value 6 of 6 (and hence for 6 286), is that U have the value U such that I U/ rR cos (a -5) (38) =e ,(:l:b=1, 3, 5, (39) since i l -1 =0. (40) s, Hence, by aid of Equations and (24) I 1: 0:12 cos (r-a) (41) 6:24P cos 0'+Q sin 6) I Qt =21rR COS (6'-5) +611" (43) H =27r(P cos 6 Q sin 6') (Mr. (44) These will be employed as design-formulas for U I and u. v

A few comments regarding the quantity 6, first occurring in Equation (39), are perhapsdesirable: From it is seen that 791 is the angleof the number -11 regarded as a complex numberyhence bean be assigned any odd positive or negative integral:-

value, as represented in. (39). Thusthe exponent 6% in (39), or the term In in (43) and (44), corresponds to?) reversals of phase in the transducers. It will be found that 5 occurs in most of the following equations, and hence that the form and properties of the physical system depend on the choice of b; it will be sufficiently illustrative of my invention to assume 6 1 for the applications to be hereinafter described.

Since 1//\=f/e and since o, the velocity of phase propagation of the radio waves along the earths surface, is approximately independent of the frequency f, it is seen from Equations 44), (21), (22) that 'M' b7r is approximately independent of the frequency. It is advantageous that the requisite value of ubnis thus (approximately) independent of the frequency, since that is a necessary condition for preserving the wave form of a composite wave, such for. instance as a carrier wave modulated by speech.

If '=(6) denotes the value of (6) when U=U, then by Equation (29),

''=u21rR cos (6 3), (45) whence, by aid of (43) and (44), I

v are (sin 6'sin n+5 (47) #4128111; (0-6) an; (0+0'2a) a a 1 swea This equationshows that, so long as a is not zero, G"(6) is zero when and only when 11/ is zero; and hence, from Equation (49) together with (46), it is seen that G (6) is zero for 6 6 and also for 6=286=6 but for no other values of 6, so long as a is'not zero.

Although, so long as a is not zero, there are two and only two values of 6 at which G 6) is zero, designated by 6' and 6" and such that 6' 6 =28, it will now appear that when a 0 there are additional values of 6 at which G (6) =0. For, when (1 0, Equation (50) reduces to I I each of which shows that G (6) is zero for those values of 6 such that 1//=n(2w), where in=0, 1, 2, 3, If any one of these critical values of 6 is denoted by 6,,, then from (49) and (46) it is seen that cos(6'8) cos(6,,,8)=

n/R,(in (),1, 2, 3, (53) Since the value of the left side of this equation necessarily lies between i2, the applicable values of n lie in the range 2R$n$2R (54) Hence, when the only applicable value of n is O; whence, by (53), there are then only two values of 6 namely 6 and 6" such that For values of there may be additional values of 6 in general, the complete set of values of 6 includes 6 and 6", these being the values of 6 for n i 0.

It is desirable, particularly for engineering applications, to have some knowledge regarding the nature and shape of the graph of I G (6) I in the neighborhood of the values of 6, which it will be recalled are the values of 6 where I G (6) I is zero. For this purpose a formula for the slope of I G"(6) I is useful, namely a formula for d IG'(6) I /d6; however, as will appear below, the absolute value of the slope will suffice, namely I 03G (6) ,"d6 I For the case =0, it is found from Equations (52), (49), (46) that In particular, when 6 6 IoZG(6)/cZ6I =27rRIS1I1(6 3)I, (57) because, when 6 6 sin (IV/2) =0 and therefore cos (1///2) =1, by Equation (52). Since, in general, the right side of (57) is not zero, it is seen that the graph of IG(6)I has at 6 6 a cusp minimum, not a stationary minimum; but when 6 =8+n1r (where in 0,1,2,3, then the graph of I G(6)I has a stationary minimum, not a cusp minimum. It is geometrically evident that at each 6 where I G has a cusp minimumthe slope of IG(6)I must be negative at 6=6 and positive-at 6=6 and the absolute values of these two slopes must be equal. It is for these reasons that a knowledge of the absolute value of the slope suifices.

For engineering applications it is also desirable to know the values of 6, say 6 for which I G (6)I has stationary values (that is, a horizontal tangent)-thus not including cusp extrema. Equation (56), by its factor sin (6- 8), shows that one set of values of 6 is given by the equation 6s=8+m (in=0, 1, 2, 3, (58),

From Equations (49) and (46) it is seen that the factor cos ('/2) in (56) contributes the additional set of values of 6 given implicity by the equation b/2R,(ib=1, 3, 5, (59) or, explicity,

6 =8+cos- [cos(68)-b/2R], (60

which gives two values of 6 for each applicable value of Z). Equation (59) shows that the applicable values of 5 lie in the range which will be employed as a design-formula for P. (This formula shows that by choosing a sufliciently large value for Z), the staggering ratio P 29/)t, could be made so large that interaction between the two waveantennae would be practically nil, even for small values of the transverse spacing ratio Q=g/,\; but the directional characteristics would be less desirable for most applications).

Since R =P +Q (64) '(6) can now be calculated by means of the formula obtained from Equations (49) and (46). A direct but less'simple formula for 1/1 (6) could be obtained by substituting (62) into (47) and employing (49).

By means of Equation (65) together with (52) and (19), the curves in Figs. 9 and 10 were computed. The ordinates of these curves represent the values of the group factor ratio 0' (0) IG (0)/G (6 )I when 6, (the reference value of 6) is taken as Zero, and b=+1. 0"(9) denotes, of course, the

and

value of the function (6) when U is so evalution to 6 1 and 0. On any one figure, 6',

0 and 0 (6) have the fixed values indicated, while the various curves thereon correspond to the various values of Q afi'iXed to the curves. -The curves are all for the limiting case of 42 0; for illustrative purposes and even for most applications, this limiting case represents the actual case with sufficient closeness, because a is very small. It will be noted that 'the shape of these curves depends very considerably on the spacing of the wave-antennae I as fixed by the parameter Q=g//\. This fact influences the choice of Q; but Q must be chosen large enough so that the spacing 20 =AQ is sufficient to prevent troublesome interaction between the two wave-antennae. It is for this reason that the curves have been plotted with Q, instead of P, as parameter; for P would seldom be subject to any practical restriction. Inspection of Figs. 9 and shows that certain of the curves (particularly for Q=O.15 and Q=0.20 in Fig; 9, and $7 0.10 in Fig. 10), have very small values of a (6) over a considerable range of 6, namely a range extending approximately from 180 to 0. For certain practical applications, this is a valuable property and is an improvement accomplished by my invention it is obtained by a proper choice of the spacing ratio 02 07); For smaller values of the values of 0(6) over this range of 6 are on the wholemuch largerr I I [Of course each curve in Figs. 9 and 10 corresponds to a definite design to a spaced staggered array of two parallel Wav'enntennte, so far as the spacing and thestaggering and the combining transducers are c The following table gives the values of the parameters pertaining to the ten designs corresponding to the ten curves in Figs. 9 and 10.

The values. of, Z), 0, 5 (5), 6, 6} were pre assigned; the resulting values of P, R, 8, 6', 0 III I, t0b1r were computed by means of Equations (62), (63), (64), (58), (42),

(44) respectively.

0, 0 1, 0 0, a (0) =0, U 1 1, 0 0 and 8+ 180 0' Q P R a 0" 0'-04 240 0 .333 .333 0. 00 120. 00 00. 240 .05 .305 .300 0. 32 13s. 03 70. 5 240 .10 .270 .204 10. 03 150. s7 s0. 0 240 15 .247 .200 31. 102. 00 01.

0 240 20 .213 42. 57 20513 101. 0 220 0 .283 .283 0. 00 140. 00 78. 0 220 .05 .205 .270 10. 70 101. at 0 220 1O 247 267 22. 07 184.10" 91. 2 220 l5 229 274 -,3 23 2.05. 57 919 220 20 .210 .200 43. 07 227.13 104. 4

oncerned.

fairly comprehensive, are furnished primarily for illustrative purposes in this patent specificatlon. If needed in practical appllcations, similar sets of curves for other values of the parameters can be readily computed by means of the formulas furnished in this patent specification.

From the remarks in the paragraph following Equation (36), it is seen that any set of curves such as those in Figs. 9 and 10 remain valid when the sign of 8 is changed provided also the sign of 6 is changed; this-fact renders it unnecessary to construct separatesets of curves for the case of 8. 7

With a supposed known (at least approximately), and with b chosen, and 9, P, Q, evaluated, the corresponding requisite values for I U I and u can be calculated by means of Equations (42) and (44) respectively; and p and q by means of (21) and (22), a being SIlPPOSQCl lZDOWIl, of course, since it is the wave length of the radio waves.

It is seen that the foregoing procedure constitutes a systematic system-method for proportioning the system as regards the stagger ing 32 and the separation q of the two waveantennae, and the ratio U=T /T pertaining i to the terminal transducers, with the object of securing desirable directional characteris tics.

tem that will embody my invention in a particular 1nstance,the procedure may be summarized as follows: Two parallel spaced and staggered wave antenna will be planned, with the spac ng g and-the stagger distance ;0 and the ratio T /T as yet undetermined. These desired to be received; also the selectivity for.

that frequency range may be enhanced by interposing appropriate filters between the antennae and the associated receiving apparatus. Appropriate values will be chosen and assigned at the outset for 6, 6, 1) (0), b and on Thereupon, one can proceed readily to construct a diagram such as in Fig. 9 or 10, making a family of curves by varying the value of Q, the ratio of the lateral spacing q to the wave length By means of such a diagram the best value for 9 will be chosen;

Next, the stagger distance p=aP is determined by Equation (62), and thereupon the value of U=T /T becomes definite and ascertainable, as expressed in Equation ('39) The value chosen for the angle 6? will be one particular direction of especially. bad interference for which it is desired to makethe reception null. Incidentally, there will result another direction-of null reception, which to gether with the assigned direction, will be To design and adjust a wave antenna sys--- symmetrical about the axis passing through the front ends of the antennae.

Moreover, a low intensity of interference over a considerable range of directions may be had by suitable choice of Q as, for example, in Fig. 9, where it will be seen that from 180 to 240 the ordinates are very much less for Q=O.15 than for Q=0.05.

Furthermore, still another arbitrary direction of null interfering reception may be assigned and embodied in the design by practicing compensation, as disclosed in connection with Figs. 6 to Sc, as will be pointed out presently. Also, in this connection it will be seen that this assignment will fix still another direction of null interfering reception, which together with the assigned directions, will be symmetrical about an axis parallel to the wave antennae.

Referring to Fig. i, this is the same as Fig. 1 except that in Fig. 4 the transducers T and the T and the receiver are situated at a convenient common point more or less remote from the back end of the wave-antennw, to which they are connected by means of metallic transmission lines'L and L (each suitably transposed) and transformers M and M Any dififerences in these transmission lines can be made up in one of the transducers, or by means of a supplementary artificial line inserted in series with the shorter transmission line. In setting forth the fundamental theory of such systems, it is convenient to regard the transmission lines L and L (including any supplementary artificial line) and the transformers M and M as constituting parts of the transducers T and T Fig. 5 represents a system which does not differ from that of Fig.4 or of Fig. 1 in fundamental principles, but possesses the practical advantage of enabling the combining apparatus to be'located near the front end instead of at the back end. This is accomplished by employing two wires in multiple for each wave-antenna and terminating the back enc of each wave-antenna with a three winding transformer (N and N respectively) so con nected that the wave-antenna currents coming in over the two wires in multiple are propagated from the back end to the front end over the metallic circuit composed of the two wires. Of course the same results could be obtained by means of the system of Fig. l employing separate metallic transmission lines for propagating the wave-antenna output currents I and I fromthe back end to the combining transducers and the receiver all situated at a common point near the front end, but the method of Fig. 5 is simpler and more economical and is fr e from possible interaction between the metallic transmission lines and the wave-antenna It is known that a single wave-antenna can be compensated to secure null reception from any specified angle of incidence of the radio waves, by suitably combining a small portion of the front-end current with the back-end output current. (See, for instance, Fig. 19 of the above-cited paper by Beverage et al.). This feature of construction and operation is involved in the systems of Figs. 6 to 8a.

Figs. 6, 7, 8 show three ways for compeneating the system of Fig. 4 so as to secure null reception from any specified angle of incidence of the radio waves, by combining a small portion of the front-end currents with the back-end output currents, by means of metallic transmission lines (L and L transformers (M and M and supplementary transducers (T and T and T In Fig. 6 the individual wave-antennae are first separately compensated (by adjusting the transducers T and T respectively), and then the so-compensated wave-antennae are suitably combined by adjusting the transducers T and T In Fig. 7 the front-end currents are combined with each other by means of T and T and the back-end currents with each other by means of T and T and finally the resultant front-end current is combined with the resultant back-end current by means of T Briefly stated, in Fig. 6 the two antennae are first individually compensated, and then are combined; whereas in Fig. 7 the two antennae are first combined (front end with front end, back end with back end), and then the combination is compensated as though it were a single wave-antenna. In Fig. 8 the four currents, consisting of the two output currents and the two compensating currents, are added directly together after having been adjusted to suitable values by means of the transducers T T T T re-- spectively. The systems in Figs. 6, 7 8 are effectively equivalent; but economically and operatively they differ somewhat from each other.

Figs. 6a, 7a, 8a show the same three different ways of compensation applied to the system of Fig. 5. Thus Figs. 6a, 7 (1, 8a correspond to Figs. 6, 7, 8 respectively in the same manner as Fig. 5 corresponds to Fig. 4.

In some of the systems represented schematically in these figures, transformers would evidently be necessary at certain points in order to secure a state of balance with respect to earth, as is well known in the communication art. These have been omitted from the figures for the sake of simplicity of description and exposition, but they may be regarded as implied. In some instances, the transducers would have to be unilateral in order to secure the contemplated results.

It will readily be understood from the foregoing exposition that the diagrams of Figs. 9 and 10 relate not to the design of the individual antennae but to a factor based on the relative arrangement and combination of two antennae with a receiving system. If each of these two antennae is directionally selective, then they should be placed so that their polar diagrams will be oriented alike, and preferably so that the maximum radii of the polar diagrams will be pointed approximately toward the direction fromwhich good signal reception is desired.

I claim:

1.- The method of directionally selective radio receiving which consists in building up from the radio waves separate, parallel, horizontal waves, spaced and staggered, and combining them with cumulative efiect for a desired direction of reception and with low effect for an undesired range of directions other than the said desired direction and with null effect for a plurality of particular directions other than said desired direction.

2. The method of securing maximum reception in one direction and minimum inter ference in another direction with a pair of parallel wave antennae, which consists in adj usting them at a certain spacing g and a certain stagger distance p, and also adjusting the ratio of the transfer factors from said antennae to a common receiver at a certain value U, the values of g, p, and U being optimum for the said character of reception.

3. The method of securing maximum reception in one direction and minimum interference in some other directions with a pair of parallel wave antennae, which consists in making them of the best length for receiving at the desired frequency in the direction of their length individually, and also adjusting them at a certain spacing and a certain stagger distance p, and also adjusting the ratio of the transfer factors from said antennae to a common receiver at a certain value U, and also drawing 05 compensating currents through filters from the opposite ends of the respective antennae; thereby getting the desired character of reception with low intensity over a considerable range of directions other than the said one direction, and null reception in two assigned directions other than said one direction, and null reception in two other directions dependent on said assigned directions.

4. The method of directionally selective radio receiving which consists in building up from the radio waves separate, parallel, horizontal wire waves, spaced and staggered each to the magnitude of best individual directional selectivity at the desired frequency,

and combining them with cumulative effect for a desired direction of reception and with low effectfor a range of undesired directions, and with null effect for at least one particular undesired direction.

5. In a radio receiving system, in combination, two parallel wave antennae spaced apart a certain distance andstaggereda certain distance 19, .a common receiver, and transducers between each antenna and the receiver with a certain ratio U between their transfer factors T and T thevalues g, p, and Ubeing optimum forI'maximum reception in one certain direction and also optimum for min; imum interference in another certain direction. j

6. In a radio receiving system, in combination two parallel wave antennae spaced apart a certain distance g and staggered a certain distance p, a common receiver, transducers between each end of each antenna and the receiver with their transfer factors and the values 9 and p at optimum for good reception in one certain direction and for minimum interference in a range of directions different therefrom and also for null reception in two other assigned directions and two additional directions, dependent on the assigned directions.

7. In a radio receiving system, in combination two parallel wave antennae, a receiver, and transducers between each end of each antenna and the receiver, said antennae being spaced apart and staggered by optimum distances, and said transducers having their transfer factors related at optimum values for good reception in one assigned direction and minimum interference in a range of other directions and null interference in several particular other directions, of which at least one is assigned.

8. In a radio receiving system, in combina tion two parallel wave antennae each of best length for directional selectivity at the desired frequency, a receiver, and transducers between each antenna and the receiver, said transducers comprising filters appropriate to currents of said frequency, and said antennae being spaced apart and staggered by distances optimum for maximum reception in one assigned direction and minimum interference in an assigned range of other directions and null interference in at least one assigned other direction and at least one other direction dependent on the last mentioned assigned direction. r

9. In a radio receiving system, in combination, two parallel wave antennae, a receiver, and transducers between: each antenna and the receiver,said antennae being spaced apart and staggered by optimum distances and the antennae being spaced and staggered at proper distances and said transducers having the proper ratio between their transfer factors to make reception maximum in one assigned direction and interference minimum in another assignedl direction.

In testimony whereof, I have signed my name tothis specification this 28th clay of December, 1927.

RAY S. HOYT.

Images