US2577619A - Metallic structure for delaying unipolarized waves - Google Patents

Description

W. E. KOCK Dec. 4, 1951 METALLIC STRUCTURE AND DELAYING UNIPQLARIZED WAVES Filed May 16, 1947 9 Sheets-Sheet l FIG.

FIG. 2

mans T/ON DEjV/CE lNVENTOR W E. KOCK d)- ATTORNEY W. E. KOCK Dec. 4, 1951 METALLIC STRUCTURE AND DELAYING UNIPOLARIZED WAVES 9 Sheets-Sheet 2 Filed May 16, 1947 Pi P Q TI E an EB wvmron By m5. KOCK Dec. 4, 1951 w. E. KocK 2,577,619

METALLIC STRUCTURE AND DELAYING UNIPOLARIZED WAVES Filed May 16, 1947 9 Sheets-Sheet 3 TRANSLA T/ON DEV/CE FIG. /2

TOP SECTIONAL WEW OFLEWS 70 SIDE VIEW OF LENS 70 e; 3 l 2/\ I 97 s l I 1 H 2 FJTWiWY;

A TTORNEV W. E. KOCK Dec. 4, 1951 METALLIC STRUCTURE AND DELAYING UNIPOLARIZED WAVES 9 Sheets-Shet 4 Filed May 16, 1947 FIELD STRENGTH -DECIBELS MN MW INVENTOR WE. max 0.; ATTORNEY Dec. 4, 1951 w. E. KOCK 2,577,619

METALLIC STRUCTURE AND DELAYING UNIPOLARIZED WAVES Filed May 16, 1947 v z s Sheets-Sheet 5.

FIG. /8

INVENTOR W.'. KOCK ATTORNEY 1951' w. E. KOCK Q 2,577,619

METALLIC STRUCTURE AND DELAYING UNIPOLARIZED WAVES Fil ed May 16, 1947 9 Sheets-Sheet 7 29 FIG. 27

/N 5 N TOR W E. KO C K ATTORNEY W. E. KOCK Dec. 4, 1951 METALLIC STRUCTURE AND DELAYING UNIPOLARI ZED WAVES 9 Sheets-Sheet 8 Filed May 16, 1947 \EN N FIG. 29

neon FOCUS 9a{ FIG. 30

lA/VENTOR W E. KOCK ATTORNEY Patented Dec. 4, 1 951 METALLIC STRUCTURE FOR DELAYING UNIPOLARIZED WAVES Winston E. Kock, Middletown, N. J assignor to Bell Telephone Laboratories, Incorporated, New York, N. Y., a corporation of New York Application May 16, 1947, Serial No. 748,448

11 Claims.

This invention relates to passive devices for changing the phase velocity of electromagnetic Waves and, in particular, to radio refractors designed for use in directive and non-directive antenna systems.

As is known, metallic-dielectric structures comprising iron wires immersed in a solid dielectric substance have been suggested for generating short (-185 meters) radio waves, and an end-on array of passive Wires or directors spaced in the air dielectric medium has been proposed for directively propagating short radio waves. Also solid dielectric delay, or so-called velocitydecrease, refractors of the focussing and nonfocussing types, have been proposed for changing the Wave propagation direction. In addition, fast or metallic-advance, or so-called velocityincrease, wave changers for altering the wave polarization, lenses for focussing the waves and prisms for bending the propagation direction, have been utilized in very short wave (1-10 meters), ultra-short wave (10-100 centimeters) and super-short wave or microwave, (1-10 centimeters) antenna systems. Thus, Patent 706,739, granted to R. A. Fessenden on August 12, 1902, discloses a metallic-dielectric short wave generator; Figs. 1 and 4 of Patent 1,860,123 granted to H. Yagi on May 24, 1932, discloses an end-on array of director wires; Fig. 21 of Patent 2,283,935, granted to A. P. King on May 26, 1942, illustrates a horn antenna having a solid dielectric isotropic delay lens in its mouth aperture, and my copending applications Serial Nos. 642,722 and 642,723, both filed on January 22, 1946, disclose metallic-advance wave changers, lenses and prisms. The refractive index of the isotropic delay lens mentioned above is, of course, greater than unity, whereas the refractive index of the metallic-advance lens is smaller than unity.

in general, the above-mentioned solid dielectric delay lens has a broad bandwidth inasmuch as it is ordinarily utilized for frequency bands in which the variation with frequency of the dielectric constant, and hence of the refractive index, is zero or negligible. On the other hand, the end-on array mentioned above is frequency sensitive since the director wires are only slightly shorter than one-half the operating wavelength. Also, the metallic-advance lens has a fairly narrow band width, inasmuch as the refractive index is dependent primarily on the plate spacing, that is, the channel width as measured in wavelengths, and the plates spacing varies rapidly with frequency.

, The metallic-advance lens has, however, dis.- tinct advantages, not found in the metallic- .dielectric wave generator or in the solid dielectric lens. Thus, the prior art metallic-dielectric and solid dielectric structures are relatively heavy and cumbersome, and the waves passing through these structures are attenuated, the total energy loss being considerable. On the contrary, the fast or metallic-advance lenses described above are lightweight and substantially lossless. Again, the directivity gain of the solid dielectric delay lens is relatively low, whereas the directivity gain of the metallic-advance lens is comparatively high. Accordingly, it now appears desirable to obtain metallic wave changers including polarizers, prisms and lenses, which possess the attributes, but are devoid of the disadvantages, inherent in the above-mentioned prior art structures. In particular, it now appears desirable to secure a lossless, unipolarized, or an isotropic broadband, lightweight metallic lens.

It is one object of this invention to retard very short, ultra-short and super-short radio waves in an efficient and substantially lossless manner.

It is another object of this invention to change a characteristic of an electromagnetic wave, such as the direction or type of polarization, the propagation direction, or the phase velocity, in a more efficient and satisfactory manner than heretofore accomplished.

It is another object of this invention to retract, with or without focussing and with only negligible loss, electromagnetic waves included in an exceedingly large band of wavelengths.

It is another object of this invention to eliminate, in a wave changer such as a lens or prism, reflection loss in a simpler and more satisfactory manner than heretofore accomplished.

It is another object of this invention to secure, in a unipolarized metallic lens, a high gain, broad-band characteristic.

It is still another object of this invention to obtain, in a metallic refractor, an eifective dielectric constant greater than unity and substantlally independent of the permeability of the metallic elements.

It is another object of this invention to obtain a broad band unipolarized refractor which, as compared to broad band unipolarized structures heretofore utilized, is lighter, simpler and less expensive.

In accordance with the invention a unipolarized metallic-delay structure comprises a dielectric medium, such as air or polystyrene foam, and an array of flat linear conductive members extending parallel to the magnetic polarization, that is, the H vector, of an incoming electromagnetic wave and hence parallel to the X dimension of the structure. The members are spaced in the medium along the perpendicular Y and Z directions which are parallel, respectively, to the vertical electric (E) vector and the horizontal propagation direction of the wave.

The linear members are of the strip type, that is, each member has a continuous or integral surface and is formed, for example, of sheet metal having a thickness of 0.005 inch. The flat sides of the strips are parallel to the E vector or height Y. The width W or height, that is, the dimension measured parallel to the E vector, of each strip, and the center-to-center spacing of the strips along the Y and Z dimensions are considerably smaller than one-half of the minimum wavelength in the operating band, the width being preferably a quarter of said Wavelength or less.

The dielectric constant of the medium is negligible and the effective dielectric constant of the conductive array, for the assumed E vector, is greater than unity. As explained herein, the eifective dielectric constant is a function of the electric polarizability of a typical member and the number of members dispersed in a unit area, taken in the YZ plane of the array. Hence the refractive index of the array is greater than unity and the aray or structure functions to decrease the phase velocity of a wave or wave component passing through the array and having the assumed vertical E polarization. A horizontal E vector is completely reflected by the horizontal strips and the cut-on orifices between the adjacent strips. Accordingly, only waves having the assumed E polarization are delayed and, as indicated above, the structure is unipolarized. Also, in accordance with the invention, the loss resulting from the reflection, if any, of the vertical E vector at the back and front faces of the array, are substantially eliminated by displacing the lower half of the array, relative to the upper half, a quarter wavelength along the propagation direction. The contours of the front and back faces of the array, or of the structure comprisin the array and the medium, are such that the array or structure constitutes a refractor of the focussing or non-fucussing type. In one embodiment, the structure is shaped so as to constitute a plano-hyperbolic lens. In another embodiment the structure has the shape of a prism.

The linear horizontal members mentioned above may be of the grid type, instead of the strip type, that is, each member may comprise vertical elements or wires spaced horizontally less than a half wavelength and having length dimensions equal to the width dimension W of the member. :3

Assuming grid members are utilized, the vertical E vectors are propagated through the array, as explained above, since the grid member functions, in effect, as a strip member for this orientation of the E polarization; and the phase velocity is decreased. Horizontal E vectors are, however, not reflected but pass through the horizontal grid members, and tln'ough the openings between adjacent grid members, without change in phase velocity. As in the strip array or structure, the grid array or structure is shaped, in one embodiment, so as to constitute a plano-convex lens and, in another embodiment, so as to constitute a prism. In still another embodiment, the array has a constant critical depth or Z-dimension and, assuming the E vector is oblique, the array constitutes a polarization circularizer or a polarization quadrature rotator.

It may be pointed out that the grid array described above should be sharply distinguished from the grid structures illustrated by Fig. 3 of my copending application for Transmission Systems, Serial No. 748,447, filed on May 16, 1947, concurrently with the present application. In the structure of my copending application the spacing between the vertical wires or elements in each horizontal grid member, while less than one Wavelength, may be greater than a half wavelength, whereby the vertical E vector is not reflected by the horizontal grid members and the vector components are not forced to pass, in each vertical panel, between the adjacent horizontal grid members. It may also be added that the thin linear strip or grid members do not, as in the case of the disk elements disclosed in my copending application just mentioned, perturb the magnetic or H field of the incoming wave and, accordingly, the effective dielectric constant of the array is not detrimentally afiected by the H vectors.

The invention will be more fully understood from a perusal of the following specification, taken in conjunction with the drawing on which like reference characters denote elements of similar function and on which:

Fig. 1 is a perspective view of a phase velocity changer constructed in accordance with the invention and comprising linear strip members, and Fig. 2 is a diagram used in explaining Fig. 1;

Fig. 3 is a perspective view of another phase velocity change reconstructed in accordance with the invention and comprising linear grid members, and Fig. 4 is a diagram used in explaining s- Fig. 5 is a perspective view of a polarization rotator or circular polarizer constructed in accordance with the invention and comprising several tiers and Fig. 6 is a partial perspective view of one of the tiers included in the structure of Fig. 5;

Figs. 7 and 8 are, respectively, perspective and end views of a strip prism constructed in accordance with the invention;

Figs. 9 and 10 are, respectively, front and end views of a grid prism constructed in accordance with the invention;

Fig. 11 is a perspective view of a point-beam antenna system comprising a plano-convex, circularly symmetrical, strip lens constructed in accordance with the invention, and Figs. l2, l3 and 14, are respectively top sectional, side and front views of the lens included in the system of Fig. 11;

Figs. 15 and 16 are diagrams used for explaining the shape or contour of the lens of Fig. 11;

Fig. 17 is a set of curves illustrating the directive band width characteristics of the system of Fig. 11;

Fig. 18 is a perspective view of a fan-beam antenna system comprising a plano-convex, cylindrically symmetrical, strip lens constructed in accordance with the invention;

Fig. 19 is a perspective view of a point-beam antenna system comprising a different planoconvex, circularly symmetrical, strip lens constructed in accordance with the invention, and Fig. 20 is a front or back view of the lens of Fig. 19;

Fig. 21 is a sectional side View of a point-beam antenna system comprising a plano-convex, circularly symmetrical, grid lens constructed in accordance with the invention, and Figs. 22 and 23 are, respectively, front and exploded perspective views of the lens of Fig. 21

Fig. 24 is a perspective view of a point-beam antenna system comprising a pyramidal horn equipped with a stepped, plane-convex, circularly symmetrical strip lens constructed in accordance with the invention; and Figs. 25, 26, 27, 28,

29 and 30 are, respectively, perspective, front, vertical sectional, diagonal sectional, top, and partial end views of the stepped lens included in the system of Fig. 24;

Figs. 31 and 32 are directive patterns of the lens of Fig. 24 taken, respectively, with the horn sides orshields removed and with the horn sides in position;

Figs. 33 and 34 illustrate the directivity bandwidth characteristic, and Fig. 35 illustrates the reflection-frequency characteristic, of the system of Fig. 24:.

Referring to Fig. 1 reference numeral 1 denotes a wave delay structure comprising a dielectric medium 2, such as air, and nine linear conductive strip members 3, as for example, sheet-metal strips, spaced along the Y and Z dimensions of the assumed volume X, Y, Z of the medium. The nine strip members 3 form an array or metallic structure it, or so-called artificial dielectric material which, as will be more fully explained hereinafter, functions to delay the phase velocity of electromagnetic waves. Numerals 5 and 8 denote arrows representing, respectively, the electric polarization E and the direction of propagation of the incoming wave; and numerals l and 8 designate, respectively, the plane of the incoming Wave front and the vertical plane of wave propagation. The width W, parallel to the E vector 5, of each strip 9 is made small relative to one-half the shortest operating wavelength, for example, one quarter wavelength, in order to avoid resonant effects which may occur when the width is in the vicinity of one-half wavelength. The center-to-center spacings Sy and Sz along the Y and Z. dimensions are each less than one wavelength and preferably less than a half wavelength. Aswill be explained, the spacings are dependent upon the selected number N of strips per square unit area, observed looking at the end of the array, that is, the numberper square unit area, of strips intersecting the YZ plane of the array. The spacings between adjacent strips, which in effect constitute electric dipoles, should be greater than the so-called breakdown value or, in other words, sufficiently great to prevent short-circuiting of the strips. The nine strips are arranged in three vertical panels 9, It, H or three strips each or, considered differently, they are arranged in three horizontal tiers l2, l3 and it. As is apparent from the drawing, the corresponding strips in the three panels are horizontally aligned. Numerals I5 and iii denote respectively the front and back faces of the structure i.

Assuming an electric field E having a polarizatlon 5 and a propagation direction 8 is impressed on the array d, the field produces a redistribution of the charges on the conductor strips 3. Each strip in effect comprises an infinite number of vertical linear elements ['1 of infinitesimal thickness, and the impressed field causes these elements to act like small electric dipoles. Each of these dipoles possesses a certain electric dipole moment which is related to the impressed electric vector 5 and the electric polarizability of the element, and hence of the strip, by the equation where M is the electric dipole moment Assuming there are N stri s 3 per unit area, taken in the YZ plane, of the dielectric medium 2 F=Nai (2) where F is the total polarization of all the strips 3, that is, of the array 4 which is immersed in the dielectric medium 2. But, as explained in my copending concurrently filed application, Serial No. 748,447,

n l2 n v 7t (3) r and where,

n is the refractive index of the array 4 e is the effective dielectric constant of the array 4;

s0 is the dielectric constant of free space;

E1 is the relative dielectric constant;

c is the phase velocity of the Waves in the array 4 on is the phase velocity of the waves in free space;

A is the wavelength as measured in the array 4,

and

k0 is the wavelength in free space.

Hence, if the electric polarizability, am, of the typical strip member 3 is known, the refractive index n of the array 4, may be ascertained. Since the dielectric constant of the air medium 2 is unity, and therefore, negligible, the refractive index of the structure I is the same substantially,

as the refractive index of the array 4. As shown below, the polarizability of the strip is directly proportional to the square of the width, W, of the strip. Also, as shown below, the electric polarizability of the strips 3 is positive, so that from Equation 5, we have that is, the refractive index of the array 4, or of the structure I, is greater than unity. Hence U0 and A0 are, respectively greater than 12 and A; and the array l constitutes a metallic-delay wave changer for decreasing the phase velocity of waves having the E polarization 5.

Considered from a difierent viewpoint, the strips, or more accurately the infinitesimal segments ll, may be regarded as capacitative conductors which load free space. Analogously, shunt capacitors along a transmission line function to reduce the wave velocity. To continue the analogue, in the case of a charged parallel plate air condenser, the capacity may be increased by inserting between the plates either solid dielectric material or insulated conducting objects, pro vided the objects or elements each have an appreciable length in the direction of the electrostatic lines of force, that is, in a direction perpendicular to the plates. Assuming solid dielectric material is inserted, the increase in capacity is caused by the shift, produced by the applied field, of the oppositely charged particles comprising the molecules of the solid material. Assuming spaced conductive strips 3 are inserted between the plates, the width dimension W being perpendicular to the plates, the strips 3 or segments l1 cause a rearrangement of the lines of force, and a consequent. increasein their number, similar to the rearrangement caused by the shift,

mentioned above, of the oppositely charged particles. Hence, the strips 3 or segments ll, Fig. 1, may be considered as segments of individual condensers or as objects which, under the action of the applied field, function as electric dipoles and produce a dielectric polarization comparable to that resulting from the rearrangement of the charged particles comprising a nonpolar dielectric. Both the polarizability theory and the capacity loading theory explain satisfactorily the delay characteristic of the metallic wave change of Fig. 1, and of the metallic delay prisms and lenses.

The electric polarizability of the strip 3 may be determined from Equation 15 on page 97 of the textbook Static and Dynamic Electricity by Smythe, for the torque of an elliptic dielectric cylinder. Thus,

where Hence, substituting the value of am as given in Equation 7 for a. in Equation 5 we have for the refractive index n of the array 4,

where, as above, N is the number of strips per square unit area taken in the YZ plane.

It is apparent from Equations 7 and 8 that the electric polarizability am and the refractive index n are each directly proportional to the square of the width W of the strip. The refractive index n is also directly proportional to the number N of strips per square unit area in the YZ plane. The number N is dependent on the center-to-center vertical spacing Sy and the center-to-center horizontal spacing Sz of the strips; and these spacings may or may not be equal, as is desired. With the desired index n and hence the desired phase velocity change vov, selected, and with W given, the number N of strips, and hence the spacings Sy and Sz, may be determined. Conversely, with W and N selected, the theoretical value of the refractive index as, for example, 1.36, may be ascertained.

Equations 5 and 8 are based on the assumptions that the strip width W is small relative to a half wavelength of the wave and that the spacings By and S2 between strips are sufficiently large to avoid appreciable mutual coupling. For optimum operation W should be less than a quarter wavelength.

Thus far only the electric (E) field of the incoming wave has been considered. The magnetic (H) field of the way may be disregarded, that is, the relative permeability ,ur of the array may be assumed to be unity, the same as air, inasmuch as the strips 3 have a negligible thickness or Z dimension. More specifically, by virtue of the negligible strip thickness eddy currents are not produced on the strips by the attenuated H field and, accordingly, the H field is not perturbed. Since the H field is not perturbed it does not affect the refractive index. In this respect the strips correspond to the disk elements disclosed in my copending, concurrently filed, application, Serial No. 748,447, rather than to the spherical elements disclosed therein.

If, instead of air, a. substance, such as hard rubber, having a dielectric constant Em substan tially different from unit is utilized in place of the air medium 2, and as a filler, the efiective dielectric constant 6c of the modified structure I may be determined from the following equation 10g ec=ks log 6+k1n. log 6m (9) where ks and km are the volume proportions, in per cent, occupied by the conductive structure and the medium, respectively, in a unit volume of the structure. By substituting for s, Equation 4, the value of so, as determined from Equation 9, the refractive index of the modified structure may be ascertained.

In operation, Figs. 1 and 2, waves having an E polarization 5 and a propagation direction 6 pass through the structure l and, as explained above, the phase velocity of the waves is decreased from Do to 1). Since the strip or dimension (W) in a direction parallel to the electric vector, is small relative to a half wavelength, the E vector is not short-circuited and reflection of the E vector 5 is negligible. According to one plausible theory, the vertically spaced horizontal strips 3 in each vertical panel in effect divide the E vector into colinear components and force these components to pass into the openings [8 between the ad acent strips. The horizontal component, if any, of the electric vector of the incoming wave is reflected in part by the strips 3 and in part by the cut-off dielectric guide spacings a. between the ad acent strips, the spacing a or width of the openings l8 being less than a half wavelength. Thus, assuming the wave has a tilted or oblique E vector i9, Fig. l, the horizontal E component 20 is reflected by the strips 3 and the openings l8, as shown in Fig. 2 by the dotted arrows 22, and the vertical E component 21 passes through the array and is retarded, as explained above and as shown in Fig. 2 by the dot-dash arrows 23. Accordingly, the array 4 or structure I, is unipolarized.

By way of comparison, it may be pointed out that the quasi-isotropic structure, disclosed in my copending concurrently filed application, Serial No. 748,447, and comprising painted disks, may be rendered unipolarized by reducing, in each vertical panel, the horizontal spacing S1 between disks to zero and, in this manner, securing strips of contiguous disks which simulate or resemble the rectangular metallic strips 3 of Fig. 1.

Referring to Fig. 3, reference numeral 24 denotes a wave changer comprising an air dielectric medium 2 and a metallic array 25 immersed therein. The array 25 is the same as the array 4, Fig. 1, except that the nine linear conductive members 26 are of the grid, instead of the strip, type. Each horizontal grid comprises a plurality of vertical elements 21, wires or rods, spaced apart a distance Sr smaller than one-half the minimum operating wavelength and each having a length equal to the width W, of the member. As in the array 4 of Fig. 1 the center-to-center spacings S and 82, along the Y and Z dimensions, of the members 26 are each less than the minimum wavelength and preferably less than one-half of the aforesaid wavelength. Also, as in the structure 4 of Fig. 1, the nine linear members 26 are arranged in three vertical panels 9, 10, ll of three grids each or, considered differently, they are arranged in three horizontal tiers l2, l3 and H. The corresponding strips in the three panels are horizontally aligned. The electric polarizability am of the typical grid member 26., Fig 3, is the same. as that of the typical strip member 3, Fig. 1. With N, and hence Sy and Sz selected, and W given, the refractive index n, which is greater than unity of the array 25 or changer 24 may be determined. Assuming the changer is to be used as a polarization circularizer, the Z dimension or thickness of the changer, as measure-d in wavelengths in the array is one-quarter wavelength, or an odd multiple thereof, larger than the same thickness Z, as measured in fre space. If it is to be utilized as a 90-degree polarization rotator, the thickness Z, as measured in Wavelengths in the array is one-half wavelength, or a multiple thereof, greater than the same thickness Z as measured in air.

In operation, Fig. 3, assuming an oblique E vector is titled at 45 degrees relative to the rods 2! and having the proper wavelength is incoming to the changer 24 along the path 6, and that the Wave changer is employed as a polarization circularizer, the horizontal E component 20 of the vector |9 passes through the grids 26, since this component is perpendicular to the rods, as shown by the arrows 29, Fig. 4. The phase velocity of the horizontal component 29 is not decreased by the array 25, but the phase velocity of the vertical vector component 2| is delayed 90 degrees relative to the horizontal component. Since the spacing Sr between the adjacent rods 21 is each grid 26 is smaller than a half wavelength, the vertical E component 2| of wave |9 does not pass between the rods and accordingly, for this component, each grid member 26 functions like a strip member. As shown by the arrows 28, Fig. 4, the vertical component is propagated through the orifices l8 between the grids 2B in each panel and little, if any, reflection of this component occurs. Hence, upon emerging from the changer, the two components 2|! and 2| are in time quadrature and, since they are in space quadrature, the emergent wave is, circularly polarized.

Assuming now that the wave changer 24 is utilized as a polarization rotator, the horizontal component 20 is retarded 180 degrees relative to the vertical component 2|, and the polarization of the resultant outgoing vector 39 is perpendicular to the polarization of the incoming vector l9. In other words, the polarization of vector I9 is rotated 90 degrees. It may be observed here that, since the circularizer and rotator just described are of the velocity-decrease type, they are distinguishable from the velocityincrease circularizer and rotator disc osed in my copending application Serial No. 642,722, filed January 22, 1946.

Referring to Figs. and 6, the wave changer 3| is basically the same as the changer 2! except that the dielectric medium 32 is polystyrene foam instead of air. The polystyrene foam medium has a dielectric constant of 1.014 and a refractive index of 1.007 and, in this respect, is substantially the same as the air medium which has a dielectric constant equal to unity and a refractive index equal to unity. More specifically, the wave changer 3| comprises an array 25 of 960 metallic rods arranged in ten vertical panels 33-42 each comprising eight horizontal rows of rods 2'! spaced less than a half wavelength apart. Stated differently, the 960 rods are arranged in eight tiers 43 to 59, each comprising ten horizontal linear grid members 26 and the members 26 each comprising twelve vertical rods 21. As

shown in Fig. 6, the ten grid members of each :5

10 tierare mounted or embedded in a horizontal polystyrene foam slab 5|. The eight slabs 5| for the eight tiers are vertically stacked so that the grid members 26 are spaced along the Y and Z dimension of the structure. The operation of the wave changer 3| as a polarization circularizer or as a polarization rotator is the same as that of the wave changer 2|, Fig. 3.

Referring to Figs. 7 and 8, reference numeral 52 denotes a metallic delay prism comprising an array 53 immersed in a polystyrene foam medium 32. The array 53 comprises 21 conductive strip members 3 arranged in six horizontal tiers 54 to 59 of 1, 2, 3, 4, 5 and 6 strips, respectively. The six tiers are mounted in individual polystyrene foam slabs 5|, each strip being fitted into a linear slot in the foam; and the Z dimension of the slabs are stepped, as illustrated, so as to form the prism. The corresponding strips in each tier are horizontally aligned. As in the structure 01" Fig. 1, the electric polarizability and the refractive index of the prism 52 may be determined from Equations '7 and 8, respectively.

In operation, assuming the wave incoming to the prism travels along path 6 and has an E vector parallel to the width W of the strip, the wave front portion passing through the thicker or bottom section of the prism is retarded a greater amount than the wave front portion passing through the thinner or upper section of the prism. Hence the wave direction 5 is bent or refracted as shown by arrow 60, the amount of refraction being dependent upon the refractive index n of the prism. In each of Figs. 1 and 8 the dotted arrow 9| denotes the outgoing propagation direction which the waves would have if the wave direction were not bent by the prism. Since only waves having an E vector, or an E vector component, parallel to the width W of the strips are retarded, the prism is a unipolarized delay structure.

Referring to Figs. 9 and 10, the metallic delay prism 62 comprises an array 63 which is immersed in the polystyrene foam medium 32. The array 53 comprises, as in Fig. 8, twenty-one conductive grid members 21 arranged in six horizontal tiers 64 to B9 of 1, 2, 3, 4, 5 and 6 grids, respectively, the six tiers being mounted in individual foam slabs 5|. The grids in adjacent tiers are staggered so that, as compared to the array 53, Fig. 8, the array 63 contains a large number N of metallic members 26 per unit area in the Z2 plane. Accordingly, assuming the polarizability of the grid members 26, Fig. 10, is the same as that of the strip members 3, Fig. 3, the refractive index of the array 62, Fig. 10, is greater than that of the array 53, Fig. 8. The staggered arrangement permits a more compact construction. As in the prism 52 of Figs. 7 and 8, and as shown by arrow 69, in operation, the prism 62 is unipolarized and functions to refract waves having an E vector 5 and a propagation direction 6.

Referring to Figs. 11, 12, 13 and 14, reference numeral 10 denotes a unipolarized, circularly symmetrical, metallic delay lens comprising an array 1| of forty-nine conductive strip members 3 spaced apart a distance Sy along the Y dimension, and a distance Sz along the Z dimension, of the polystyrene foam medium 32. The fortynine strips 3 are arranged in seven vertical panels 12 to E8 or, stated differently, in eight horizontal tiers '19 to $6. The foam medium 32 comprises eight vertically stacked slabs 81 to 94 which contain vertical slots for retaining the strips 3 and support, respectively, the eight tiers 19 to 86. Numerals 95, 96, 9'! and 98 denote, respectively, the front plane face, the back convex face, the optical axis and the point focus of the lens I0. As shown on the drawing, the lengths of the slots and the strips 3, and the depths of certain outer slots and the widths of the corresponding outer strips, are selected so as to conform to the convex contour of the back face 96 of the lens. Accordingly, while the number N of strips per unit area of the YZ section is constant throughout the array I I, the array includes fractional, as well as whole strips. Numeral 99 denotes a conical, point-type horn having its throat orifice positioned at the focal point 98 and connected by the dielectric guide I to a translation device IOI.

As in the polarization circularizer, polarization rotator and prisms heretofore described, with W and N and hence Sy and Sz, selected, the polarizability a and the refractive index 71., greater than unity, may be ascertained. Conversely, with n and W selected the value of N and hence the values Sy and S2, may be determied.

Referring to Fig. 15, the equation for the contour of the convex face 99, which faces towards the focus 98 and the horn 99, will now be determined. In Fig. 15, the reference letter A denotes the phase length of the path traversed by a wavelet or ray emitted at the focus 98 and propagated through the thick portion of the lens and along the axis 97 of the fiat front face 95. Reference letter B denotes the phase length of the path of a ray emanating from the focus 98 and propagated to the periphery so as to avoid the lens and reach the front face 95. In order to convert the spherical wave front I02 originating at the focus 99 into a plane wave front I03 at the front face 95, A and B must be equal. Now

i 2 A v and where f is the focal length of the lens.

a is the lens thickness along the axis 91.

y is the radius or half-aperture of the lens 10. v is the phase velocity in the lens.

110 is the phase velocity in free space.

Hence f fi (f+ Z/ 7 v0 (1 or, since where n is the refractive index, we have (n 1):r +fm(n1)y =0 (14) which is the equation for a hyperbola, the origin 0, 0 being at the vertex of the curve or convex face 95. The hyperbolically convex-plano lens shape is disclosed and claimed in my concurrently filed application, Serial No. 748,447.

In operation, Fig. 11, assuming the device IN is a transmitter, energy is supplied by the transmitter IIII over guide I00 to the horn 99 and a wave having a vertical polarization 2| and a spherical wave front I02, is propagated towards the lens 50. The phase of the wavelets passing through the thick central or vertex portion of the lens are retarded a greater amount than the phase of the Wavelets propagated through the outer thinner lens portion, and the wavelets arriving at the flat front face are cophasal. Stated differently, the outgoing spherical wave front I02 is converted by the lens I0 to a plane wave front I03 extending perpendicular to the axis 91. In reception, the converse operation is obtained, and an incoming plane wave front I03 having a propagation direction parallel to the axis 97 is transformed by the positive planoconvey lens I0 into a spherical wave front I02 converging on the focus 98. In other words. the incoming parallel rays I04 and I05 are bent or refraction by the lens, as shown by the rays I06 and I01, and focussed on the primary antenna 99. Inasmuch as the lens I0 is circularly symmetrical, focussing action is obtained in all planes containing the axis 91. As in the prism of Fig. 7, only waves or wave components electrically polarized parallel to the width dimension W of the strips are refracted, the waves polarized perpendicular to the aforesaid dimension being reflected.

If desired, the flat surface of the lens I0, Fig. 11, instead of the convex surface, may face the focus 98 and conical horn 99. In this alternative arrangement, illustrated by Fig. 16, the curvature of the convex face differs from that of the convex face illustrated by Figs. 11 and 15. More particularly, and as disclosed and claimed in the copending application of A. M. Skellett Serial No. 717,214, filed December 19, 1946, now United States Patent 2,547,416, granted April 3, 1951, with the flat lens surface 98 facing the horn, the curvature of the convex surface 95 is obtainable from the following equations in parametric form where n is the refractive index and r is a parameter, and the terms Rmax, R, D, d, r and ,f represent the distances indicated in Fig. 16.

Referring to Fig. 17, reference numerals H0, III and H2 denote respectively the E-plane directive patterns taken over a 12 per cent band and, more specifically, at 3860, 4120 and 4360 megacycles corresponding, respectively, to wavelengths of 7.76, 7.28 and 6.89 centimeters, for a system similar to that illustrated by Figs. 11 and 15. In the tested system, a lens 10 having a diameter or aperture of three feet and comprising conductive sheet metal strips having a thickness of 0.005 of an inch was utilized. The strip width W and the spacings Sz and S were, respectively, three-quarters of an inch, threeeighths of an inch and one and one-half inches. As shown in Fig. 17, the half-power widths II3 of the major lobes II4 of the three patterns I I0, III and H2 are the same, namely, 4.5 degrees, so that the directive action is uniform over a very wide band of frequencies. In addition, the minor lobes H5 in the three patterns are down from or below the major lobe peak about the same amount, namely, about 15 decibels. It may be added that, for wavelengths of 7 and 8 cen imeters, the measured refr ctiv indices of the 13 tested lens were about 1.43 and 1.41, respectively, and the index n calculated in accordance with Equation 8 was 1.45. Accordingly, the measured and calculated values of n are substantially equal or at least comparable.

Referring to Fig. 18, reference numeral I denotes a unipolarized metallic-delay lens which is similar to the lens 13 of Fig. 11, the primary difference between the two lenses being that the lens I20 is cylindrically symmetrical and has a line focus I2I, whereas the lens 10 is circularly symmetrical and has a point focus 98. Also, the lens I29 has an optical axial plane I22, whereas the lens 16 has an optical axis 91. The front face I23 of the lens I20 is flat and the back face I24 is cylindrically convex, the convex curvature being determined in accordance with Equation 14. The lens I20 comprises an array I25 of conductive strips 3 spaced in the medium 32 along the Y and Z dimensions and arranged in tiers 19 to 86, the tiers being supported or retained in slots in individual foam slabs 81 to 94. The strips 3 have equal lengths and are mounted in slots having equal lengths corresponding to the strip length. Numeral I 26 denotes a sectoral horn having its mouth orifice I21 aligned with the focal line I2I, the mouth orifice being equipped with the flares I28. The horn I26 is connected by guide I30 to the translation device MI. The polarizabi-lity am and the refractive index n, of the lens I20, for waves having an E-vector 2I parallel to the strip width 3, are the same, respectively, as the polarizability and index of the lens 10, Fig. 11.

In operation, Fig. 18, assuming the E-vector 2I is vertical and the focal line I2I is horizontal, waves supplied by device IOI over guide I30 to the horn I26 are projected toward the reflector I20 in the form of a vertical fan-beam, the beam being wide in the vertical plane and narrow in the horizontal plane. The lens I20 functions to focus the waves in the vertical plane, but does not focus the waves in the horizontal plane, that is, the lens I20 has a horizontal fan-beam characteristic. Hence, the outgoing beam from the system comprising the horn or primary antenna I26 and the lens or secondary antenna I20 is narrow in both the vertical plane and the magnetic plane, or, in other words, the system has a pointbeam characteristic.

Referring to Figs. 19 and 20, reference numeral I30 denotes a unipolarized, circularly symmetrical, strip lens of simple lightweight construction. The lens I30 comprises six cellophane panel sheets I3I, I33, I35, I31, I39 and MI, and five solid dielectric, cellophane or polystyrene foam, spacer sheets I32, I34, I35, I38 and I40, a spacer sheet being positioned between adjacent panel sheets. Each panel sheet contains a circular lens panel I42 comprising a plurality of conductive tin foil strips 3 fastened to the front surface of the panel sheet. The panels have graded diameters, and hence different pluralities of strips, in conformity with the convex optical back face 95 of the lens I33. As in the grid prism of Fig. 10, the corresponding strips in adjacent panels are not aligned, but are staggered. Thus, the eight strips 3 of panel sheet I33 are staggered relative to the nine strips 3 on panel sheet I3I, and are therefore opposite the eight spaces I43 of panel sheet I3 I. Also, the lens axis 91 passes through the central strip 3 of panel sheet I3I and the through spacing I43 between the two central strips 3 on panel sheet !33. The panel and spacer sheets are held together by means of the wooden ring I44, wooden ring I45 and the boltand-nut assemblies I46; and the lens is supported by the stand I41. As in the structures previously described, the refractive index n is greater than unity for waves having the E-vector H and is dependent upon the values selected for the factors W and N. The lens I30 has a point focus 98 and its convex face is toward the horn 99.

The operation of the system I30, 99, Fig. 19, is the same as that of the system 10, 99 of Fig. 11. Thus for waves electrically polarized parallel to the strip width W, the lens focusses the waves in all planes containing the optical axis 91 and the system I39, 99 has a point-beam characteristic.

Referring to Figs. 21, 22 and 23, reference numeral I50 denotes a unipolarized, circularly symmetrical, grid lens of simple lightweight construction. The len I50 comprises an array I5I of 39 horizontal grid members 26 each comprising a plurality of vertical metallic rod elements 21. The rods 21 are spaced a distance SX smaller than a half wavelength apart and have lengths corresponding to the grid width W. The grids 26 are arranged in seven circular vertical panels I52, I53, I54, I55, I56, I51 and I58 extending parallel to the XY plane. The panels I52 to I58 have graded diameters and the component rod 21 are disposed and arranged so that the outermost surfaces of the outermost rods lie in and substantially define the contour of a predetermined conventional shape of wave refractor, such as, for the present example, that of a plane-convex lens.

Alternatively the structure of Figs. 21, 22 and 23 can be considered as one in which the grids are arranged in 11 plano-convex vertical curtains I59 to I69 extending parallel to the YZ plane. In the XY plane the grids are staggered and arranged in 15 horizontal tiers I10. More specifically, panels I52, I53, I54, I55, I56, I51 and I58 have, respectively, eight, seven, six, seven, six, five and two tiers, the grids in adjacent tiers or in adjacent panels being staggered. The rods 21 are mounted in vertical polystyrene foam slabs I1I, one for each of curtains I59 to I69, as shown in the expanded perspective view of Fig. 23, and the slabs extend parallel to the YZ plane and are spaced the same horizontal distance Sx, as the rods. The Sy and Sz spacings of the grids 26 are each smaller than a Wavelength and, as in the structures of Figs. 3 and 5, preferably less than a half wavelength. The slabs I1I are held in position by means of the single wooden ring I48. As before, the refractive index n of the lens I50 is greater than unity for waves having the vertical E-vector 2 I and is dependent upon the values selected for W and N. The lens I50 has a point focus 98 and is unipolarized; and its convex face 96 is toward the horn 99.

The operation of the system, Fig. 21, comprising the lens I 50 and the horn 99 is the same, for waves having the E vector 2!, as that of the system I30, 99, Fig. 19, and as that of the system 10, 99, Fig. 11. As regards the H-vector components, if any, these components pass through the grid lens I53, Fig. 21, whereas they are reflected by the lens 10 and the lens I30. If desired, matching sections of tapered dielectric constant may be utilized on each face of the lens I30 and I50 in order to minimize the reflection losses, if

any.

The type of construction utilized in the strip lens I30 Fig. 19 and the rid lens I50, Fig. 21, permits a relatively close spacing S2. By reason of the close spacing, and also by virtue of the staggered arrangement, a high efiective dielectric constant, and therefore a high delay, are obtainable. The measured effective dielectric constants of a strip array and a grid array, constructed in accordance with Figs. 19 and 21, were 225 and 20, respectively. The strip type artificial material having a dielectric constant of 225 is especially suitable for use in a traveling wave tube, since the corresponding refractive index is 15 and a tube filled with the aforesaid material would have a phase velocity v of that is,%,-, the velocity in free space.

Referring to Figs. 24 to 30, reference numeral I80 denotes a large unipolarized, circularly symmetrical strip lens having a square periphery I8I and positioned at the mouth orifice I82 of a square pyramidal metallic horn I33. The lens has an optical axis 97, a focal point 98 and, as explained below, an unstepped multiple-zone inner or central section I84, an unstepped multiple-zone outer section I85, a plane ofiset front face I86 and a stepped convex offset back face I8'I. The center of the horn throat orifice I88 is coincident with the focal point 98 of the lens, and the throat orifice is connected by the rectangular guide I89 to a translation device IOI, such as a super-short wave transmitter or receiver. The flare angles of the horn sides, in the E and H planes, are such that, if the sides were extended to the front face I86, as shown by the dotted lines I90, Fig. 27, the four horn sides would just touch the circular periphery of the inner or central zone I84. In order to permit utilization of the lens I80 and the horn I83 with five, six or seven-centimeter waves corresponding to guides I00, and throat orifices I83 of diiferent size, the small throat section I9I of the horn is made removable. Hence, a throat section I9I containing a throat orifice I88 of given size may be removed, and another throat section I9I containing a throat orifice I83 of different size may be inserted in its place.

The lens I80 includes an array comprising a large plurality, say 1200, assuming the lens is six foot square, of horizontal sheet metal strip membars 3 spaced a distance Sy along the Y dimension, and a distance Sz along the Z dimensions of the lens. The strips are arranged in a large number, say 40, of vertical panels I92 (XY plane), Fig. 30, or stated differently, in a large number, say 60, of horizontal tiers I93 (XZ plane). As in the lens of Fig. 11, the tiers are supported in horizontal polystyrene foam rectangular slabs I94, one for each tier, the strip members 3 being mounted in individual retaining slots in the slabs. The vertically stacked slabs I94 are held in position by means of the square wooden frame I95.

The diameter or aperture of the circular central section I84 is ordinarily, but not necessarily, sufficiently large so that the unstepped central section I84 includes several wavelength zones or steps, whereby this section may be stepped, if desired, for the purpose of reducing the thickness or Z dimension of this central section I 84. Similarly, the diagonal dimension of the square lens is ordinarily sufficiently large so that the unstepped outer section I85 includes several wavelength zones, whereby this section may also be stepped, if desired, for the purpose of reducing the thickness of the outer section I85. While, as shown, the outer section is stepped relative to the central section, each of these sections taken by itself is not stepped inasmuch as stepping a lens having a refractive index independent of frequency, such as lens I80, reduces the band width, as in the case of a parabolic reflector. By way of contrast, stepping a lens, such as a fast metallic lens, having an index which varies rapidly with frequency increases the bandwidth. The step I96 between the two sections is of the intermediate type disclosed in my copending application, Serial No. 642,723 mentioned above. If desired, however, the step I96 may be of the horizontal type or of the focal type, both also disclosed in the last-mentioned application. As will now be explained, the convex curvature for the outer section I differs from the convex curvature of the central section I84.

Referring to Fig. 30, it is assumed that the ray I91 from the focus 98 travels a distance T through free space and arrives at the front face I 86 of the lens I80, whereas the ray I98 from the focus 98 travels the same distance T through the artificial delay material or metallic array and arrives at the front face I89. Now

and

where i1 is the time the ray I91 takes to travel the distance T in free space,

i2 is the time the ray I98 takes to travel the distance T in the metallic delay array,

2:0 is the phase velocity in free space, and

o is the phase velocity in the lens I80.

Since,

vo o (20) we have,

But, in order for rays I91 and I98 to arrive in phase at the lens face I86,

t2-t1 must equal where K is any integer and A0 is the mid-band wavelength in free space.

As already indicated, referring to Fig. 15, Equation 14 defines the convex contour I 99 of the central section I84, the origin 0, 0 being taken at the vertex of the curve. Consequently, by adding T as given by Equation 27 to the value of X, and subtracting it from f, in Equation 14,

7 17 we obtain the following equation for the convex contour of the back face of the outer section I85.

In practice, the particular value for K is utilized which brings the convex contour of the outer section nearest to the corners of the horn.

In order to eliminate the reflection, if any, of energy from the front and back faces of the lens into the guide I89, the upper half of the lens is offset, horizontally, a distance D equal to one.- quarter of the mid-band wavelength, or an odd multiple thereof, relative to the bottom half of the lens.

The operation of the lens I80 is basically the same as that of the. lens .of Fig. 11. Thus, waves having a vertical E vector 2I parallel to the width dimension of the strips are delayed and refracted, and focussing in all planes containing the axis 91 is obtained. Waves having a horizontal E vector are reflected at the faces of the lens. The horn sides function to shield the lens whereby, as discussed in more detail below, the minor lobes are reduced and a highly satisfactory directive characteristic is obtained. By reason of the quarter wave offset D the wavelets reflected from the to half of the lens are displaced 180 degrees in time phase rel tive to t s refl ted y the wer ha f and sub tantia ly no reflected ene gy en e s the uid 189.-

A sh l d en s stem const ucted accord: ance with Figs. 24 to 30 was tested at frequen es included in the 3900 to 4300 megacycle band. In the tested system, the X and Y dimensions of the lens were each six feet, the lens thickness or Z dimension was sixteen inches and the focal length ,7 was about five feet. Ihe Width W of the strips 3 and the strip spacings S): and Sewers, respectively, about 0.75, 1.31 and 0.875 inch, corresponding to 0.25, 0.46 and 0.13 of the mid-band or intermediate wavelength of 7.2 centimeters; and the strip thickness was 0,005 inch. The measured index of refraction .n of the lens I80, with the strip width and strip spacings mentioned above, Was about 1.50. .As discussed below, ,Figs. 31 to 35 illustrate the measuredcha acteristics of the tested system I80, I83, just described.

Referring to Figs. 31 and 32, reference n nnerals 200 and 20I denote the H or horizontal plane directive patterns, both measured at a wavelength of about 7.2 centimeters, for the tested lens I80 taken by itself, that is, without the shields or horn sides, and for the tested lens I80 with the horn sides, respectively. Numerals 202 and 203 designated the major lobes, and numerals 204 and 205 denote the minor lobes, of patterns 200 and 201, respectively. Each pattern, it will be observed, is highly symmetrical about the axis 91. The symmetry indicates a high degree of homogeneity of the artificial dielectric material or array, a quality or property not always found in solid dielectric substances, such as polystyrene. As shown in Fig. 31, the half-power width 205 of the major lobe 202 is about 2.6 degrees and the half-power Width 201 of the major lobe 203 is approximately 2.68 degrees. The beam or halfpower width is, of course, a function of the lens aperture and for a larger aperture as, for example, 10 feet, the beam would be considerably more narrow. The minor lobes are exceedingly low and are. generally smaller, as is desired, than the minor lobes in the patterns of com arabl prior art structures. Thus. even'in t e p t 200- for the SiXr'fQOt unshi l d ens. the minor obes 0.4 a e a out 28 decibels own. a bel w the p a o t maj r lobe 200-. I th pattern. 20 for h si o hi e e h minor lobes 208 are more than and almost 40 decibels do n Un bt y. if he aperture were 10 feet the minor lobes would be still lower. By way of comparison, the minor lobes fo the shielded dielectric channel fast lens dised in lny oopending application Serial No. 172;), are about 2 5 to 30 decibels down. While t 00 and 2M are horizontal plane x h, pattern represents fairly accurate? the pattern for any other plans containing the anis 0?. In practice, the major lobe of the'vertical plane pattern is usually slightly more narthanfmajor lobe 2122 or 203, and the minor lobes are more pronounced inasmuch as the waves are polarized in the vertical plane and, accordingly, the illumination of the lens by the throat orifice is tapered less in this plane than in the horizontal plane.

Referring to Figs. 33 and 34, the curve 2) denotes the absolute gain of a standard horn over the wide frequency band 3920 to 4300 megacyoles and numeral 2iI designates the measured gain, over t e standard rn. of the ac ual y Cons ru ed shielded lens I80, I83. Since, during the test, the field strength varied along the horn mouth orifice or lens aperture, but did not vary in the orifice of the standard horn, a correction is applied to the curve 2I I. Reference numeral 222 denotes the corrected gain curve for the shielded lens I80. The gain for the ideal or optimum isotropic antenna having an area corresponding to that oflens I80 is represented by the curve H3. The difference in the gain curves 2I2 and 2I5 is a measure of the eficiency of the shielded lens; and the curve 214, Fig. 34, represents this measured difference. For comparison purposes, the measured efficiency curve 2 i 5 of a shielded metallic-advance channel lens is shown in Fig. 34. The mid-band gain or efliciency, taken at 4220 megacycles, of the shielded metallic-delay lens of the invention is about one decibel better than that of the metallic-advance lens. Also, by reason of the fact that the refractive index n of the lens of the invention is constant with frequency, the gain curve 2M is fiat, that is, the gain is uniform, over the 500 (4420 minus 3920) megacyole band. On the other hand, the index n of the metallic-advance lens varies with frequency and the gain is not uniform over the band.

Referring to Fig. 35, reference numeral 2 I S denotes the measured frequency-matching characteristic for the horn I03, Fig. 24 and numeral 2 I l designates the frequency-matching curve of the lens I as attached to the horn I83. As shown by the curves the standing wave ratios or characteristics 2I6 and 2ll, as measured in the guide I86, do not vary substantially over the band, as is desired. Thus, the ratio or curve 2i? for the horn varies less than 0.4 decibel and the ratio 2? l for the shielded lens varies about 0.3 decibel.

Although the invention has been described in connection with certain embodiments, it is to be understood that it is not to be limited to the embodiments described inasmuch as other apparatus may be successfully employed in practicing the invention.

What is claimed is:

l. A metallic electromagnetic phase velocity changer comprising fiat rectangular conductive members having equal width dimensions extending in a direction parallel to the electric polarization of an incoming electromagnetic wave and spaced apart in a direction coincident with said width dimensions, said structure having an effective dielectric constant greater than unity and dependent upon the aforesaid width dimensions of said members.

2. A metallic array for retarding an electromagnetic wave having a given wavelength, at given propagation path and given electric and magnetic polarizations, said array comprising conductive members spaced in a dielectric medium along directions parallel to said propaga tion path and said electric polarization of said wave, the length and width of said members being parallel to said magnetic and electric wave polarizations, respectively, the spacing between adjacent members and the width of each member being dependent upon and smaller than onehalf of said Wavelength.

3. An array in accordance with claim 2, said array having an effective dielectric constant greater than unity and dependent upon the spacing and width of said members.

4. A piano-convex metallic lens for refractlng an electromagnetic wave, said lens comprising a polystyrene foam medium and flat conductive strip members spaced in saidmedium along directions parallel to the electric polarization and the propagation path of said wave.

5. A lens in accordance with claim 4, said lens being circularly symmetrical and having a point focus.

6. A lens in accordance with claim 4, said lens being cylindrically symmetrical and having a line focus.

7. In combination, a lens for refracting electromagnetic Waves included in a given band of wavelengths, said lens having an axis and a focus, a primary antenna at said focus, one-half of said lens being displaced relative to the other half, and along said axis, one-quarter of the mean Wavelength in said band, or an odd multiple thereof.

8. A plano-convex, circularly symmetrical, stepped radio lens comprising an array of conductive strips extending parallel to the magnetic vector of an incoming plane polarized electromagnetic Wave, the outermost surfaces of the outermost strips of said array being positioned to substantially define the exterior contours of said plane-convex, circularly symmetrical, stepped radio lens, the widths of said strips and the spacings therebetween being proportioned to produce 20 a predetermined eflfective dielectric constant for said leris.

9. A hyperbolically convex-plano, circularly symmetrical, lens for refracting a super-short radio wave included in a given band, said lens having a refractive index greater than unity and an optical axis, said lens comprising an inner zone and an outer stepped zone, each zone comprising a plurality of flat conductive strip members having longitudinal dimensions extending parallel to the magnetic polarization of said waves; the width dimension of said members being smaller than one-half the minimum wavelength in said band and extending parallel to the electric polarization of said waves, said members being arranged in tiers extending parallel to said axis, each tier being mounted in a polystyrene foam slab, the' spacing between tiers and the spacing in each tier between the adjacent members being smaller than one wavelength.

10. A lens in accordance with claim 9, the focal length of one-half of said lens beinga quarter of the mean wavelength in said band, or an odd multiple thereof, greater than the focal length of the other half of said lens.

11. In combination, a pyramidal horn having square throat and mouth orifices, a translation device connected to the throat orifice of said horn, a stepped plano-convex square metallic lens positioned at the mouth orifice of said horn and having its convex surface facing said throat orifice, said lens having an optical axis coincident with the axis of the horn and a point focus coincident with the center of said throat orifice, said lens comprising a plurality of conductive strips extending perpendicular to said optical lens axis.

WINSTON E. KOCK.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 2,288,735 OConnell July 7, 1942 2,415,807 Barrow et a1. Feb. 18, 1947 2,464,269 Smith Mar. 15, 1949 OTHER REFERENCES Electronic Industries, March 1946, page 66.

Electronics, March 1946, page 101.

Bell Telephone System Publication, Monograph 11-1423.

Images