US2235506A - Ultra short wave radio system - Google Patents

Description

March 1941 s. A. SCHELKU NOFF ULTRA SHDRT WAVE RADIO, SYSTEM Filed June 8. 1939 4 s t s t 1 FIG] IN EN TOP 5. A SCHEL KUNOFF A TTORNEV s. A. SCHELKUNOFF 235.506

ULTRA SHORT WAVE RADIb SYSTEM March 18, 1941.

Filed June 8 19 4 Sheets-Sheet 2 FIG. 7 I

FIG. 8

- l/E N TOP 8 y S/{ CHEL-KUN OFF A 7'TORNE Y March 18, 1941. a A SCHELKUNOFF 2,235,506

ULTRA SHORT WAVE RADIO SYSTEM Filed June 8, 1 939 4 Sheets-Sheet 4 FIG/.5

RA OIUS FIXED WAVELENGTH VARIABLE lNl/ENTOR By SASCHELKU NOFF Y A TTORNE V Patented Mar. 18, 1 941 PATENT OFFICE 1 ULTRA SHORT WAVE RADIO SYSTEM Sergei A. Schelkunofl, New York, N. Y., assignor to Bell Telephone Laboratories, Incorporated, New York, N. 1., a corporation of New York Application June 8, 1939, Serial No. 278,032 1'? Claims. (Cl. 250-33) This invention relates to radio antennas and apparatus to be associated therewith which are particularly well adapted for use with wavelengths of a few meters or less. More particularly 5 it relates to spherical, spheroidal and similar dipole or doublet antenna systems and combinations of such antenna systems with conical and disctransmission lines.

An .object of this invention is to provide an ultra-short wave antenna system which will retain eiilcient radiating and absorbing properties over a wide range of frequencies.

Another object is to provide an eflicient antenna system having relatively small air resistance for use on aircraft.

Another object is to provide an eflicient impedance transformer for use with the antennas of this invention and other radio apparatus at ultra-high frequencies.

Another object is to provide types of transmission lines readily adaptable for use in ultra short wave systems employing spherical dipole or doublet antennas.

Another object is to provide an ultra short wave antenna system which afiords eiilcient operation and relativelyconstant impedance over a wide range of frequencies and occupies a relatively small space. I

Other and further objects will appear during the course of the following description.

Formobile craft, particularly aircraft, space which may be occupied by antenna systems is frequently very limited. On the other hand, in order that 'a radio system for use on mobile craft 8!! may be of substantial utility it is desirable that it should have a relatively constant impedance and should be capable of emcient. operation over a wide range of frequencies. A primary object of this invention, therefore, is to provide an antenna 4o system, for use in ultra short wave radio systems, which while occupying a relatively small space will, nevertheless, retain a substantially constant impedance and satisfactory operating emciency over a wide range of frequencies.

45 A particular antenna system disclosed in this application which satisfies the above-stated requirements is one of approximately spherical contour, comprising a sphere of highly conductive material, the sphere being divided virtually into 60 hemispheres, insulated from each other, their bases being adjacent, the hemispheres'being employed as the elementsof a dipole or doublet antenna system. Throughout this application and in the appended claims the term "doublet antenna" is employed in the sense imparted to it by the 1931 Standardization Report of the Institute of Radio Engineers. The definition in the report I is A doublet antenna is an antenna consisting of two elevated conductors substantially in the same straight line, of substantially equal length, 5

with the power delivered at the center. Amathematical analysis of such an antenna system is given hereunder. It is demonstrated that such an antenna system, for ultra short wave. radio systems, occupies a relatively small space, has a 10 substantially constant impedance and operates efliciently over a wide range of frequencies.

However, such a spherical dipole or doublet antenna system has a relatively low impedance and it is convenient, in some instances therefore, to 1 provide for use therewith an efflcient ultra high frequency impedance transforming device. To meet this requirement a disc transmission line is suggested and is shown hereinafter to possess the desired electrical properties and in addition to be 20 peculiarly well adapted mechanically for integration with the antennas proposed in this application into novel combinations having substantial merit as antenna systems.

An analysis of conical lines is also given and 25 advantageous uses of conical lines in connection with ultra-high frequency antenna systems of the types disclosed in this invention are pointed out.

A new form of conical line in which the component cones are inverted with respect to each other 30 is disclosed. This type of line has been designated and is referred to throughout the application and claims as an inverted conical line and is particularly wellsuited for use with the dipole or doublet antenna systems of this invention as will become apparent hereinafter.

The results obtainedwith truly spherical systerns may be approximated by spheroidal or ellipsoidal systems in which the system is divided as for the spherical system into halves, the halves 40 forming a dipole or doublet system.

Antennas of this invention may in general be either of solid construction, shell construction or the desired surfaces may be simulated to any desired degree of approximation by groups of con- 5 ducting wires or rods appropriately shaped and joined together.

For more effective use on aircraft, the system may be shaped to approximate a.streafiilined outline to offer minimum air resistance. In this 60 particular application non-conducting portions whose sole function is to complete the streamlining" of the system may be added to the radiating portions of the antenna system.

The principles of the invention will be more readily understood in connection with the following description and the accompanying drawings, in which:

Fig. 1 shows in a partly sectional view, a spherical dipole or doublet antenna coupled by a conical transmission line to a simple concentric transmission line;

Fig. -2 shows in a partly sectional view a spherical dipole or doublet antenna coupled by a disc transmission line to a simple concentric transmission line; 1

Fig. 3 shows in a partly sectional view a spherical antenna coupled to a simple concentric transmission line by an inverted? conical line;

.Fig. 4 illustrates a combination of conical and disc transmission lines providing an impedance transformation the combination providing convenient means of joining two simple concentric line circuits of substantially different impedances;

Fig. 5 ilustrates the coordinate system employed in the mathematical analysis of the spherical antenna;

Fig. 6 illustrates the coordinate system employed in the mathematical analysis of the usual type of conical transmission line;

Figs. 1 and 8 are illustrative of the coordinate,

system employed in connection with the mathematical analysis of inverted conical lines;

Fig. 9 shows the relation between the angle it between the boundaries of the conductors of an inverted conical line and its characteristic impedance Z0;

Fig. 10 shows the relation between the angles 0' and 0" of the conical conductors of the usual type of conical line with respect to their common axis and the characteristic impedance Z0 of the line;

Fig. 11 illustrates the coordinate system employed in the mathematical analysis of the disc lined tail of non-conductive material;

' Fig. 1'7 shows an ellipsoidal dipole or doublet antenna of this invention which approximates the properties of the spherical dipole antenna and offers less wind resistance and reduces icing difficulties when employed on aircraft.

The mathematical theory of the spherical di pole or doublet antenna is quite simple. This antenna comprises a conducting sphere with an equatorial segment removed. The two sections of the sphere remaining-form a dipole or doublet antenna system, each member of the dipole or doublet being energized from the periphery of its base. The coordinate system most convenient for the mathematical analysis of such a system is that shown in Fig. 5. As shown in Fig. 5 the coordinates of a typical point P are the distance r from the center 0 of the system, that is the radius of the sphere, the colatitude or vertical angle 0 and the longitude or horizontal angle In view of the physical configuration of the system, as indicated above, in the present analysis,

attention will be directed to axially symmetric electric waves, in which the lines of magnetomotive intensity H are coaxial circles and the electromotive intensity has two components, the

. radial component Eland the meridian component E,. The horizontal component E of the electromotive intensity and the radial component H! and meridian component H, of the magnetomotive intensity are all zero for the physical configuration assumed.

The plan of analysis is as follows:

Starting with the general set of Maxwell's equations expressed in terms of spherical coordinates, an axially symmetric fleld is assumed so that only partial derivatives with respect to vanish. As a consequence 01' this assumption, the general set of equations breaks up into two independent subsets. One of these subsets represents e1ectric" waves and the other magnetic waves.

Attention is then concentrated on the electric waves which are of most practical interest in connection with this invention. The most general axially symmetric wave can be resolved into zonal components. Having obtained the most essential properties of zonal waves, the actual resolution of a spherical wave caused by an alternating electromotive force across two circles of latitude equidistant from the equator is e1- fected. Thus, it is possible to determine the current passing through the source of the electromotive force and, hence, the radiation impedance. I

Maxwell's equations for electromagnetic waves in perfect dielectrics are For the sake of symmetry Er is measured, as illustrated in Fig.5, toward the center of the "frame of reference in Equations 2. ,Similarly, H. is measured from the south" pole toward the 6 north pole in Equations 3. Magnetic lines of I force in the first case and electric lines in the second case are circles coaxial with the frame of reference. Y

For the purposes of this analysis the electric waves only need be considered. The results can be readily extended, however, to magnetic waves by an interchange of letters.

In order to simplify the solution of Equations 2 the following function is introduced 15 =r sin Hi. (4) j and the electromagnetic intensities are expressed as follows:

' 'ipr sin or where The quantity A is the wave-length in centimeters 35 and 1;, the characteristic impedance of space. Introducing the function #1 into the last equation of theset 3, there results i It can be shown that this equation possesses m) fie n. 5-

65 and TKcos 0) is the Harmonic of the first order. Remembering the implied factor, e it is apparent 70 that Equation '7 represents progressive diverging waves andEquation 8, progressive converging waves. The field along the axis of. the frame of referenoeis'flnite, and n is, therefore, a positive integer. 7 Substituting iI/n from Equation '1 into Equation 5, the following expressions for the intensities of diverging "Mary zonal waves" 5 E.=A, Times (De- 1o n n+1 t ipr Ef- A, I P,.(c0s 0)e i) I 10 An ideal Hertzlan. dipole or doublet is the source of primary zonal waves. The radial specific impedance is defined by Em mm Z,, 1

m 4 mm) and the meridian specific impedance by E n(n+1)P(cos a 5 H iprT(cos 0) (12) The radial specific impedance, governing transmission-of energy in the radial direction, is of especial interest in connection with this inven tion. Similarly, the intensities of converging zonal waves are I H L IT 'L Z w aponry so E l mon!) (13 Er: fl" 'l :'l .p g) av(r+el) r and the corresponding impedances become E. fix- 2o] w f-( r E, n(n+ 1) P,.(cos 0) 4o sphere will be termed a free spherical oscillation." If the original distribution of charge is symmetric about an axis and varies as P; ;(cos 0) from "north to south,then the spherical oscillation will be called zonal.

For a perfectly conducting sphere the free periods of zonal oscillations can be found from the following equation:

1.0a) -r.(ip 5) stating the fact that the meridian component of the electromotive intensity vanishes on the surface of the oscillator of radius a.

In the case of primary waves, (15) becomes By previous implication p has a positive real part; therefore, the lower sign in (17) should be 7 rejected. Hence, the time factor in the Expressions 10 for the intensities is 6 2; 20: Thus, the attenuation equals i er 2a nep 5 sec so that the disappearance of the field in the neighborhood of the oscillator is practically instantaneous. The wave-length is in which the coeflicients are given by" By (11) the magnetomotive intensity on the sur face of the oscillator is n( o Hence, the magnetomotive force acting along the typical parallel (the electric current crossing the parallel) is 210 I I a 0T; 0 v 22 The complex flux of energy can nowbe caicu- 1 lated h-aMn-i- 1f (Thestar signifies the conjugate of.) The real part of Q represents'twice the energy radiated by the oscillator.

It may be noted that the total flux of energy is the sum of the quantities of energy carried by component zonal waves. Letting h there results Ian-21.0), mowsin one) (2 n-l I I a-Lacum and and since it can be proved that r I I LE,,,(0)I,,(0)d0=0,ifm;-n (27 A case of especial practical importance will now be considered. Let the electromotive force V (the time factor e is of course understood) be 81110: I (2n+1)E 1 m f P -sin a It can be shown that B2m=0 and that the odd numbered coefllcients are given by With a degree of accuracy suflicient for most practical applications therefore, we may write,

pedance is tabulated below for four frequencies.

, a I (4m+3)E V 2m+1 o uni-H00 Introducing into Equation 29, and remembering that there results 7 V Bzm+i= z-+1 a (31) Hence, by Equation 22 we have I a dnH-l 1 v: v

I (0) +12% zh+1( a) T m+1( I Thus, the current flowing through the north" edge of the generator is I I I Q 03 4-1 I( cos aw g T},,, (s1n a) 33 The ratio of the electromotive' force V to this current is the radiation impedance. This is z== Q ohms 34 005 EO ZM-H Znfi-I a 1 where the quantities Azm+1 and 'Wzm+1 are defined by m+i= m+i m+M a fl-( fl flier-m i (36) a is the ratio of the circumference of the oscillator to the wave-length, and sin a is taken as 0.1.

The radiation The radiation The radiation impedance resistance R reactanco X m in ohms in ohms in ohms 3. 7 -44 1 11. 3 --l8 2 v 14. s -14. 95 21. 0 12.3 l4. so 9- a generator or from a transmission line ter-- minating at the center or along the "northsouth axis of the spherical system, to'the edges of the component hemispheres. The first method which will beinvestigated consists in employing a cone transmission line, i. e., the transmission line formed by two coaxial conducting members having conical surfaces which if projected would have coincident vertices. In its usual form the conical transmission line comprises one conical conducting member enclosed by-a second conical conducting member, the two being arranged coaxially, asmall portion of each member near the common apex being removed to avoid short-circuiting the line at the common apex point and to afford access for making connections to that end of the line. It is in this form obviously a special type of coaxial or concentric transmission line and it is commonly employed to provide a uniformly tapered medium for interconnecting two tubular concentric lines,

the conductors of one of which are of larger 1 diameter than those of the other; If the two tubular concentric lines are ofsubstantially the same impedance the section of conical line joining them may be proportioned to have this impedance so that no electrical irregularities o'c'cur at the junction. A novel form of conlcaltransmission line in which the two conical conducting members forming it are not assembled one within the other but, being coaxially arranged, are extended in opposite directions from a common vertex point, is. described hereinunder and" advantageous uses of the arrangement are explained. The term inverted conical line hard as is mentioned above-been coined to designate this type of line and, as in the case of the llfiilfll conical line, it is necessary to remove a small portion of one or both component cones adjacent to the common apex point to avoid short-circuiting the line at this point and to afford access for making connections to that end of the line.

As long as the vertices of the two conical conducting members are coincident no impedance transformation-is introduced by the conical line of either the usual or the inverted types discussed herein.

The usual form of conical line is, as mentioned above, that based upon the system of coordinates shown in Fig. 6 where one cone enclous the other, 1/ being the angle between the-cones, 0' being the angle made by the inner cone and 0" approaches being that made by the outer cone with their common axis.

The alternative form, referred to above and hereinafter as the inverted conical line, is that based upon the system of coordinates shown in Figs"?- and 8, the essential distinction :being that the two cones are inverted with respect to each other. In the precise arrangement illustrated by the coordinate system of Fig. l the two cones are of identical dimensions and p is theangle b tween them,.0' being defined as 90--a, where and 0" being defined as 90+a. 0 could equally well be defined as the angle made by one of the cones with the common axis of the .cones and 0'. would then be 0'+. The latter definitions might be found more convenient in a non-sym-' metrical system, that is,.a system in which the two component cones had difierentapex angles.-

For the present purposes-however, a symmetrical system is'convenie'nt and may be employed .as shown in'Flg. 3 to connect a concentric line with a spherical dipole antenna system. The

underlying principles'are, of course, substantially identical and will become apparent to those skilled in the art during the. course of the following description.

For any type of conical. line in which the vertices of the two component conical members are coincident the characteristic impedance is tan 0" tan (as n 2 1 tan 0' 9 tan 0 If the angle between the boundaries of the type, that is,- the type in which. one conical conductor encloses the other, the impedance closely I zo= 60 log ohms (40 In Fig. 9 the impedance of an inverted conical line of' the symmetrical type contemplated in connection with Fig.7, isplotted with respect to the angle 0 between thetwo cones.

Since the angle 1p should be small for low impedances and the impedance of spherical, sphe-- roidal and the like dipole antenna systems is inherently low the combination illustrated in Fig.

3 is readily proportioned to afford apreclse impedance match between the dipole antenna system and the inverted conical line. 1

In the following mathematical analysis use is made of a spherical frame of reference coaxial with the cone transmission line. For the present, attention is confined to the principal mode of propagation in which the lines of electromotlve intensity extend fromone cone to the other. The assumption is also made th'atthe cones are perfectly conducting. This introduces no appreciable error in the expression for the characteristic impedance (above one megacycle) at the fre- 'quencies we are contemplating. As the result of these assumptions, the radial'component of e. m. i. vanishes identically and the-field equa-.

tions assume a particularly simple form. Thefollowlng are the equations for the fundamental mode of propagation along conical transmission lines where fir) is some function of 1' only.

It is convenient to introduce into these equations the magnetomotive force 1(1') acting along any parallel at distance r from the vertex of the transmission line,

I (r) =21rr sin 0H, amperes (42) This magnetomotive force equals, of course, the radial current at distance r from the vertex.

Multiplying the first equation of (41) by 21mm sin 0, we obtain v where 10) is also a function of r only. The electromotive force V(r) along the typical meridian is related to f(r) as follows:

where 6' and 0" are the colatitudes of the metallic boundaries of the conical transmission line. Hence,

tan $0 Taking into account Equations 42), 43) and (45) the set of Equations (41) can be rewritten as follows:

Thus, the conical transmission line, the members of which have a common vertex, turns out to be an ordinary line with uniformly distributed constants. The characteristic impedance of this transmission line is tan 0' tan g0 z,=-"- log =60 log ohms 48) r tan %0' t tan $8 If the metallic boundaries of the conical transmission line of the inverted type make an angle a with the equatorial plane, then as above stated Substituting this in Equation (48), the followlng expression for the characteristic impedance results 1+tan a Z =l20 log (51) 1 tan a If a is small, then Z =120 log (1+2 tan %a =240 138.11%0:

' 120a ohms (52) Thus, for example, if the angle between the metallic boundaries equals V radian, the characteristic impedance is approximately 12 ohms. The inverted conical transmission line may therefore, as stated above, be employed in systems requiring a low impedance line. v

The disc transmission line will next be investigated. It is, by definition, two parallel metallic discs, energized at a, pair of points lying on the perpendicular, which includes the axes of the two discs. One possible use for such a a generator, or from the end of a coaxial transmission line, at the center of a spherical dipole or doublet antenna'to its edges as shown in Fig. 2. I

The characteristic impedance of a disc transmission iine is given by the relation n is the intrinsic impedance of the dielectric between the discs, A is the .wave-length, h is the separation between the discs and p is the distance from the axes. of the discs. For air =37? ohms.

For very short waves or at greatdistances from the axes, this becomes so that the impedance'is seen to be essentially a pure resistance varying inversely as the distance.

If, on the other hand, the wave-length is very long by comparison with the circumference 21g,

, Equation 54 becomes substantially The resistance part is then seen to be practically independent of the distance from the axis of the "disc ton line." Y

The curves ll, 48 and 50 of Fig. 12 illustrate 35 transmission line is to convey the energy from v variation of resistance, reactance and impedance, respectively, with the ratio of circumference to wave-length for a disc transmission line havin a maximum radius of one wave-length. The

Equation 53 shows that in order to calculate the characteristic impedance of a disc transmission line, we have merely to multiply the ordinates of the curves of Fig. 12 by or those of the remaining curves 52 El and 56, Figs. 13 to 15, inclusive, respectively, by

21p 7 In the following mathematical analysis, a cylindrical frame of reference coaxial with the disc transmission line, as shown in Fig. 11, is employed. For the present, attention is confined to the principal mode of propagation in which the lines of electromotive intensity extend from one disc to the other; No substantial error is introduced in the calculation of the characteristic impedance if the discs are assumed to be perfectly V;Eh volts, I =21rpH amperes Thus, the series impedance of the disc transmiss sion line is seen to be inversely proportional to the distance from the axis and the shunt admittance directly proportional to this distance.

If the electromotive intensity is eliminated from Equation 56, there results where A is the wave-length. This is Bessels equation of order one. Its solution, appropriate to the waves diverging from the axis, and the corresponding expression for the electromotive intensity can be written as follows:

H=AK,(ik E=A1 K ('ikp) (60) where is the specific impedance of the dielectric between the metallic boundaries of the disc. For air or empty space, as above stated, n is 377 ohms.

From Equations 58 and 60, it follows that V: AnhKgGkp), I=2rApK (ikp) (61) Hence, the characteristic impedance H of a disc transmission line is given by the following expression (P) v o( p) Thus, where the wave-length is relatively short, the characteristic impedance of the disc transmission line is seen to depend upon the distance from the axis of the line.

On the other hand, if the wave-length is very long, as compared to the length of the circumierence 21 the following approximate expression for the impedance is obtained,

[592.2-592.2i(g ;"+o.11s)]% (63) In this case the resistance part of the impedance practically independent of the distance from the axis of the line.

For very short waves or at great distances, another approximate expression results Here th impedance is practically a pure resistance nd as was observed in connection with Equation 62, it is inversely proportional to the distance from the axis. I

.By means of the following identities, where Nv is Neumann's function representing the imaginary part of Hankels function H}(:c) the real and the imaginary parts of Z0 may be separated.

In Fig. 4 a combination of a disc transmission line 34 and a conical transmission 38 is shown which may be employed to join two coaxial lines 36 and 40 of different characteristic impedances, the structure of Fig. 4 providing the appropriate impedance transformation required to, in effect, match the impedances of the lines. Such a. struc-' ture could, for example, be inserted between the radio receiver or transmitter and any of the antenna systems of Figs. 1 to 3, inclusive, to provide additional impedance transformation, if desired.

' In Fig. 16 a further possible practical application of a combination of structures of this invention is illustrated, 60 being the lower surface of an aircraft. The two fhemispheres 24 forming a spherical dipole or doublet antenna, discs 30 antenna through the disc transmission line.

forming a disc transmission line and concentric line comprising conductors 63 and 64 serving to connect radio apparatus in the aircraft to the Extension 66 of non-conducting material is added to complete the streamlining of the antenna system and streamlined support 62 serves to mechanically support the antenna system in a suitable position beneath the aircraft. Obviously, concentric line 63, 64 may be enclosed in members 66 and 82 so as to present no additional air resistance and the space between the hemispheres 24 may be enclosed or filled with non-conducting material to reduce air resistance and also'conceivably to alter the characteristics of the disc transmission line composed of discs 30. As is well known in the art, non-conducting material employed in the manners above described should have very low dielectric losses. For this reason in most instances a thin shell providing the desired contour for streamlining will be found preferable to substantially solid members.

It would obviously require only reasonable mechanical skill to devise means for rotating the spherical dipole or doublet antenna (leaving member 66 stationary) in one, or several, planes so that its directional properties can be employed to best advantage for horizontal, vertical or oblique transmission and reception, as desired. As for any dipole or doublet antenna system, maximum radiation efliciency of the spherical diplole or doublet antenna system obtains in the plane normal to the longitudinal center of the system.

Fig. 17 shows a modification of the spherical dipole or doublet antenna into an ellipsoidal antenna system which, while retaining to a substantial degree the desirable impedance-frequency characteristics of the former, is more readily adaptable to streamlining, or to shaping for the purpose of avoiding or reducing accumulation of ice upon the exterior under some conditions encountered in flying.

Obviously, an additional non-conducting member may be added to the combination of Fig. 16 which would extend the contour of the left hemisphere 24 so as to reduce difllculties resulting from icing. Alternatively, a non-conducting shell of any desired contour could be provided to enclose the entire antenna structure.

Obviously, too, the principles embodied in the structures of Figs. 1 and 3 may be readily incorporated by modification of the structures of Figs. 16 and 17 in these latter structures whenever the circumstances indicate that it would be advantageous to do so.

Numerous additional applications of the abovedisclosed principles of the invention will occur to those skilled in the art and no attempt has here been made to exhaust such possibilities. The scope of the invention is defined in the following claims.

What is claimed is:

1. In a high frequency transmission system including two circuits of different impedance, means for obtaining a change in impedance level comprising a pair of conducting discs placed coaxially,

. the distance between said discs being less than 10 per cent of their diameters, and means for connecting the said two circuits of the system electrically to the centers and. the peripheries of said pair of discs respectively.

2. In a high frequency transmission system including two portions of like impedance but of different physical dimensions, an inverted coniaasasoe cal transmission line comprising a pair of coaxial cones having a common apex point but extending in opposite directions along their common axis from said apex point, a small portion of at least one of said cones adjacent its apex being removed, the apex angles of said cones being proportioned to make the impedance of said conical line substantially the same as that of the said two portions of said system, the circuit portion of smaller dimensions connecting to said cones near their respective apices and the circuit portion of larger dimensions connecting to said cones at their base peripheries.

3. In combination, a doublet antenna of spherical contour comprising two spherical segments each of one base and of like dimensions, the altitude of each segment being aproximately .9 the spherical radius of the segments, the segments being insulated from each other and placed coaxially with bases paralleland separated by a distance of approximately .2 the spherical radius, and a non-conducting tail therefor, the nonconducting tail being disposed closely adjacent to the rear side of said doublet antenna, its adjacent surface fitting the contour of said antenna, its other surfaces being streamlined whereby the drag of the air on said antenna is effectively reduced irrespective of the direction in which the said antenna is oriented for maximum response.

4. In combination, a doublet antenna system for radio system comprising a pair of hemiellipsoidal conducting members of like dimensions placed coaxially with bases adjacent but spaced approximately .1 the base diameter apart, and insulated from each other and a disc transmission line comprising two conducting discs placed parallel and coaxially, their peripheries connecting to the base peripheries of the said two hemiellipsoidal members, respectively, and means for making electrical connection to said discs near the respective center points thereof.

5. A doublet antenna for radio systems comprising a pair of conducting spherical segments each of one base and of like dimensions, the altitude of each segment being approximately .9 the spherical radius of the segments, the segments being placed coaxially with their bases parallel and spaced less than 25 per cent of their spherical radii apart, the spherical radius of said segments being approximately 16 per cent of the wavelength of the lowest frequency to be employed, and means for severally and symmetrically connecting electrically to the respective peripheries of the bases of said two segments.

6. A doublet antenna for radio systems comprising a pair of hemispheroidal conducting members placed coaxially, their bases adjacent, said members being separated by a distance of approximately 10 per cent of their base diameters and electrically insulated from each other.

7. A doublet antenna for radio systems comprising a pair of h'emiellipsoidal conducting members placed coaxially, their bases adjacent and electrically insulated from each other, the radii of the bases of said conducting members being approximately one-sixth the wave-length of the lowest frequency to be employed, the separation between the bases being less than 25 per cent of the radius of the bases, and means for severally connecting electrically to the respective peripheries of the bases of said two hemiellipsoidal members.

8. A doublet antenna for aircraft as in claim '7, the conducting members thereof being shaped to reduce air resistance and icing difficulties.

, 9. In combination, a spheroidal doublet antenna comprising two spherical segments, each of one base, said segments being of like spherical radius, and both having altitudes of approximately .9 said radius, the segments being placed coaxially with bases adjacent but separated by approximately .2 of the spherical radius, a conical transmission line comprising two coaxial cones of different apex angles having a common apex point, the cone of larger apex angle enclosing that of smaller apex angle, the former cone having a small portion at its apex end removed, the apex angles of said cones being proportioned to impart an impedance to the conical line substantially equal to that of the said doublet antenna, the conical line being assembled within one radiating member of said antenna, the peripheries of the larger ends of the cones connecting symmetrically to the peripheries of the bases of the two radiating members of said antenna, respectively, and means connecting to the conical line near the apex point thereof, said means comprising a cylindrical twoconductor concentric line passing through the radiating member enclosing the conical -line,-the outer conductor of the concentric line connecting to the outer member of the conical line and the inner conductor of the concentric line connecting to the inner member of the conical line.

10. In an ultra-short wave radio system operating over a broad range of frequencies, an antenna system comprising two radiating elements of substantially hemispherical shape having equal radii, the said radii being approximately one-sixth the wave-length of the lowest frequency of said broad range of frequencies, said elements being arranged coaxially with bases adjacent but spaced approximately 20 per cent of their radius apart and insulated from each other. a

11. In a radio system the combination of an the bases of the two conducting members respectively.

12. In a high frequency transmission system, a disc transmission lin comprising two conducting coaxial discs spaced and insulated from each other, the diametersof and the spacing between said discs being proportioned to provide a predetermined impedance transformation between the centers and the peripheries, respectively, of the discs. i

' 13. In a radio system the combination of an antenna system comprising two conducting members of substantially hemispherical exterior, the two members being placed coaxially with bases adjacent but spaced approximately 20 per cent of their radius apart and-insulated from each other and an inverted conical line comprising two "coaxial conical members having a common apex point, said members. extending in opposite dipheries of the" bases of the two conducting meinbers respectively, the apex angles of the cones being chosen so that the impedance of said line is substantially the same as that of said antenna system.

14. In a radio system for aircrait, an antenna system of spheroidal contour, comprising two radiating elements each consisting of a hemispheroidal conducting member, the base radii of said respective conducting members each being approximately one-sixth of the maximum wavelength of the frequency range in which said system is to operate, the radiating elements being placed coaxially with bases adjacent but separated by approximately 20 per cent of their radius and being insulated from each other.

15. In a radio system, a transmission line consisting of two conducting members of conical shape having a small portion near the apex of each removed, said members being coaxially arranged and extending in opposite directions from a common apex point, the angle between said members being less than thirty degrees, so that the line formed by them will have a predetermined low characteristic impedance.

16. In a radio system, the combination of a two-element antenna system of spheroidal contour the elements of said antenna system comprising identical spherical segments of one base each having an altitude of approximately .9 the spherical radius of the segment, the segments being placed coaxially with bases adjacent but separated by approximately .2 the spherical radius of the segments, the segments being insulated from each other, and an inverted conical line comprising two coaxial'cones having a common apex point and extending-in opposite directions therefrom a small section of one cone near its apex being removed, the outer peripheries of the two conical members of said line connecting to the base peripheries of the two radiating members of said antenna system respectively, the angle between said two conical members of said line being such that the characteristic impedance of said line substantially matches that of said antenna system.

l 17. In a radio system, an antenna system comprising two conducting members of hemispheroidal shape and of like radii, said members being placed coaxially with their bases toward each other. the bases being separated bya distance of approximately .2 the radius of the bases the base periphery of" each member joining the base periphery of a conical conducting member, the latter members each having a small portion near their respective apices removed, the latter members being so positioned that their surfaces if projected would have a common'vertex point, the angle between said latter members being such that the impedance of the conical line formed by them is substantially equal to the impedance of said antenna system, and means connecting to placed coaxially with respect to the said antenna members and extending through one of said antennamembers to said conical members, the

outer conductor of said cylindrical line pair con necting to the nearer conical member and the inner conductor to the farther conical member.

SERGEI A. scrimxoisorr'.

Images