WO1996013943A1

WO1996013943A1 - Hole eliminator for lamp with reflector

Info

Publication number: WO1996013943A1
Application number: PCT/US1995/014109
Authority: WO
Inventors: Philip Jones; Akira Tomita
Original assignee: Raychem Corporation
Priority date: 1994-11-01
Filing date: 1995-10-30
Publication date: 1996-05-09

Abstract

To be useful in image projection systems, white light beams should be well-collimated and have a uniform light intensity. To use a reflector, such as a parabolic reflector, to generate such a beam, a conical prism lens (or other optical system) placed in the beam acts to eliminate a hole in the beam produced by a shield mounted on the light source. To eliminate the hole the relation between the radius r of the shield, the conical prism lens thickness h, the index of refraction n of the conical prism lens, and the angle ζ between the outside surface of the conical prism lens and the radial direction is: r ≈ h*sinζ*[l - cosζ*(n?2 - sin2ζ)-1/2¿]. To correct for a reduction in light intensity with radial distance, the angle ζ may increase with radius. A Fresnel equivalent of the conical prism lens may be used, and when the illumination plane is sufficiently far from the Fresnel lens, dark bands will not be visible. If a non-parabolic reflector is used, the lens system must include a means for collimating the beam.

Description

HOLE ELIMINATOR FOR LAMP WITH REFLECTOR

Technical Field of the Invention

This invention relates generally to a system for eliminating a hole in light projected from a light source which includes a reflecting element, and more particularly to a lens or mirror system for eliminating such hole, particularly when the reflecting element is a parabolic reflector.

Background of the Invention

In a projection display, light from a light source has an image imparted to it and is projected onto a screen for viewing. The display may be a front projection one, in which the projection mechanism is positioned on the same side of the screen as the viewer, or a rear projection one, in which the projection mechanism is positioned on the side of the screen away from the viewer. In either case, a collimated light beam of uniform brightness is required to achieve an image of uniform brightness.

Beams of red, green and blue light may be combined and projected to provide a colored display. The red, green and blue light beams may be generated from separate red, green, and blue light sources. However, using a single white light source whose light is separated into red, green and blue beams is a preferred design, since it is more economical to use a single light source and true white is assured when the beams are combined. The separation (and subsequent recombination) may be effected by dichroic mirrors (e.g., Williams et al., WO 90/05429 (1990); Tanaka et al., US 5,164,821 (1992)).

A conventional light source 20 including parabolic reflector 22 and arc lamp 24 is shown schematically in FIG. la. Arc lamp 24 is powered by power source 34 which is connected by first wire 36 to left lead 32 of arc lamp 24. Power source 34 is also connected via second wire 37, connector socket 38, and internal lamp wire 39 to right lead 33 of arc lamp 24. Right lead 33 is encased in insulator 53, on which is mounted light shield 26, typically about 10 mm in diameter. Light shield 26 intercepts all light that would otherwise exit light source 20 without striking parabolic reflector 22, i.e., a projection from emission gap 30 of arc lamp 24 past an edge of light shield 26 falls on parabolic reflector 22.

Left lead 32 is encased in stiff insulating tube 52 which supports arc lamp 24 such that gap 30 between left lead 32 and right lead 33 is located at the focus of parabolic reflector 22, i.e., at the focus of a parabolic cross-section thereof. Therefore, all light rays emitted from gap 30 and reflected off parabolic reflector 22 travel parallel to axis of symmetry 42 of parabolic reflector 22. Because light source 20 is cylindrically symmetric about axis of symmetry 42 (except for wire 39 which has only a small effect on beam 47), beam 47 produced by light source 20 is also cylindrically symmetric about axis of symmetry 42, as shown in the cross-section thereof in FIG. lb.

The paths of four light rays 40a-40d are traced in FIG. la. Their positions within beam 47 are shown in FIG. lb. (Light rays within the beam are generally referenced with the numeral "40," the hole in the beam has the reference numeral "45," and the entirety of the beam, i.e., the light rays 40 and the hole 45, has the reference numeral "47.") A first ray 40a passes close to the upper edge of light shield 26, strikes parabolic reflector 22, and exits parallel to axis of symmetry 42. Ray 40a lies at the outer boundary of beam 47, as shown in FIG. lb, since light rays cannot strike parabolic reflector 22 any farther from apex 23 due to light shield 26. Another ray 40d strikes parabolic reflector 22 closer to apex 23, is reflected parallel to axis of symmetry 42, and passes just outside of light shield 26. As shown in FIG. lb, ray 40d lies at the outer boundary of a hole 45 in beam 47. Rays which strike parabolic reflector 22 closer to apex 23 than ray 40d are reflected back parallel to axis of symmetry 42, and strike arc lamp 24 and light shield 26, thereby creating hole 45 in the beam. Beam 47 from light source 20 is therefore a column of light with a central hole 45 having a radius equal to that of light shield 26, with innermost ray 40d located at the outside of the hole 45. So, while this type of light source 20 provides a highly collimated beam 47, hole 45 therein makes it less desirable for uses requiring uniform illumination. In displays which utilize electronically-modulated light scattering materials, the contrast ratio increases with an increase in the f number, i.e., a decrease in the diameter of beam 47. Unfortunately, decreasing the diameter of beam 47 to increase the contrast of the display increases relative size of hole 45. (Also shown in FIGS, la and lb are light rays 40b and 40c which exit gap 30 at angles between that of previously discussed rays 40a and 40d. They too strike parabolic reflector 22 and are reflected parallel to the axis of symmetry 42.) Summary of the Invention

An object of the present invention is therefore to provide a light source with a beam having a uniform intensity. Another object of the present invention is to provide a light source with a highly collimated beam. Another object of the present invention is to provide a light source with a large f number. Another object of the present invention is to provide an electronically modulated light scattering display with high contrast. Another object of the present invention is to provide a light source which uses a reflector, particularly a parabolic reflector. Another object of the present invention is to provide a light source which uses a prism lens for elminating the central hole. Another object of the present invention is to provide a light source which uses a mirror system for eliminating the central hole. Additional objects and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objects and advantages of the invention may be realized and obtained by means of the instrumentalities and combinations particularly pointed out in the claims.

The present invention is directed to a lamp for providing a collimated beam. The light source is mounted on a reflector, and has a shield to intercept light which would otherwise escape the lamp without reflecting from the reflector. The shield produces a hole in the reflected light, and a light displacement system (which can be a lens or mirror system) is used to displace light towards the center of the hole by a distance equal to the width of the hole, thereby eliminating the hole in the beam. In a preferred embodiment, the light is displaced by a distance approximatly equal to the radius of the hole.

Brief Description of the Drawing^

FIG. la is a cross-sectional view of a conventional lamp. FIG. lb is a cross-section of the beam produced by such lamp, showing a hole in the beam.

FIG. 2a is a cross-sectional view of a lamp according to the present invention, which includes a conical prism lens. FIG. 2b is a cross-section of the beam produced by the lamp of FIG. 2a. FIG. 2c is a cross-sectional view of a lens which is the Fresnel equivalent of the conical prism lens of FIG. 2a. FIG. 2d is a cross-section of the beam produced by the lens of FIG. 2c.

FIG. 3 shows a perspective view of the conical prism lens of FIG. 2a. FIG. 4 is a plot of the ratio of the radial displacement to the thickness of the cone versus the angle of the cone apex to the radial direction for indices of refraction n=1.5 and n=1.7. FIG. 5 is a schematic view of a transmissive liquid crystal projector using the conical prism of the present invention.

FIG. 6 is a schematic view of a reflective liquid crystal projector using the conical prism of the present invention.

FIG. 7a shows path traces of two pairs of light rays which emanate from a light source with a separation angle of δφ, reflect from the parabolic reflector, and pass through a prism lens having a corrective curvature for providing a uniform light intensity. FIG. 7b is a view from the front of the parabolic reflector of path traces of two pairs of light rays, the rays in each pair emanating at an azimuthal separation angle δμ. FIG. 7c is a plot of light intensity versus radius for the light reflected from the parabolic reflector of FIG. 7a. FIG. 7d is a plot of light intensity versus radius for light transmitted through the prism lens with a corrective curvature shown in FIG. 7a. FIG. 7e is a diagram illustrating the effective thickness h' of a prism lens with corrective curvature.

FIG. 8a is a cross-sectional view of a lamp where a bi-concave lens is used to collimate the light from an elliptical reflector prior to incidence on a conical prism. FIG. 8b is a cross-sectional view of a lamp where the order of the conical prism lens and the elliptical reflector in FIG. 8a is reversed.

FIG. 9a shows an embodiment of the invention in which a mirror system is used to eliminate the central hole. FIG. 9b shows a construction for the mirror system of Fig. 9a.

FIG. 10 shows another embodiment in which a mirror system is used to eliminate the central hole.

Description of the Preferred Embodiments

FIG. 2a shows a lamp 120 for projection of a highly collimated beam with no central hole. Lamp 120 includes parabolic reflector 122 and arc lamp 124. Arc lamp 124 is powered by power source 134 which is connected by first wire 136 to left lead 132 of arc lamp 124. Power source 134 is also connected via second wire 137, connector socket 138, and internal lamp wire 139 to right lead 133 of arc lamp 124. Right lead 133 is encased in insulator 153, on which is mounted light shield 126 which prevents light from exiting lamp 120 without striking parabolic reflector 122, i.e., a projection from emission gap 130 of the arc lamp 124 past an edge of light shield 126 falls on parabolic reflector 122. Left lead 132 is encased in stiff insulating tube 152 which supports arc lamp 124 such that gap 130 between left lead 132 and right lead 133 is located at the focus of parabolic reflector 122, i.e., at the focus of the parabolic cross-sections thereof. Therefore, all light rays leaving gap 130 and reflecting off parabolic reflector 122 travel parallel to axis of symmetry 142 of parabolic reflector 122. For optimum collimation, gap 130 should be as small as practicable, e.g., approximately 1 mm in length. Because all optical com¬ ponents of lamp 120 are cylindrically symmetric about axis of symmetry 142 (except for wire 139 which has only a small effect), beam 140 produced by lamp 120 is also cylindrically symmetric about axis of symmetry 142.

The paths of four light rays 140a 140aV 140a", 140b/l 40b'/ 140b", 140c/ 140cV 140c", and 140d 140d'/140d" are traced in FIG. 2a. The portion of each ray between parabolic reflector 122 and conical prism lens 150 is assigned an unprimed reference numeral; the portion of each ray passing through conical prism lens 150 is assigned a reference numeral with a single prime ('); and the portion of each ray subsequent to conical prism lens 150 is assigned a reference numeral with a double prime ("). (Herein, "lens" refers to any transparent optical element used to manipulate a beam of light, and a "lens system" may be a single lens or a combination of lenses.)

After reflecting off parabolic reflector 122, rays 140 travel parallel to axis of symmetry 142 until they strike conical prism lens 150 (which is shown in a perspective view in FIG. 3). Conical prism lens 150 has a conical exterior surface 151 and a conical interior surface 152. The angle ø of conical exterior surface 151 from the radial direction (i.e., ø is the angle between the normal vector to the surface of the cone and axis of symmetry 142, and interior angle of apex 154 of lens 150 is 180° - 2 * φ) is equal to that of interior surface 152 from radial so that conical prism lens 150 has a constant thickness h. As shown in FIG. 2a, each ray 140 strikes lens conical prism 150 at an angle φ from the normal of conical exterior surface 151. Rays 140 are refracted by the larger index of refraction n₂ of conical prism lens 150 to produce rays 140' which pass therethrough at an angle θ from the normal, where according to Snell's law: n_| sin φ = n₂ sin θ , and n[ is the index of refraction of the surrounding medium, which will usually be air and thus is close to unity. On exiting conical prism lens 150 light rays 1407140" are again refracted according to Snell's law, and since conical interior and exterior surfaces 152 and 151 are parallel, exiting rays 140" again travel parallel to the axis of symmetry 142. Solving for the inward radial displacement d of light upon passing through conical prism lens 150, it is determined that: d = h * sin φ * [1 - nj cos φ * (n₂ - n^ sin φ)^~ ] or, when the index of the surrounding medium is unity and setting n₂ = n: d = h * sin φ * [1 - cos φ * (n² - sin² φ )^"1/2] .

(The above equalitites need hold only to the extent necessary to make the distance of the innermost rays from the axis of symmetry 142 small in comparison to the radius of beam 140.)

Plots of radial displacement d divided by thickness h of conical prism lens 150 versus angle φ of exterior surface 151 from the radial direction for indices of refraction n] = 1 and n₂ = n = 1.5 and 1.7 are shown in FIG. 4. As expected, radial displacement d increases monotonically with angle φ for constant index of refraction n and thickness h, and when angle φ is zero, radial displacement d is zero. As also expected, a larger radial displacement d is achieved with an increase in the index of refraction n, for constant thickness h and angle φ. It should be noted that losses from reflection increase with increasing angle φ, so the optical efficiency of conical prism lens 150 decreases with increasing angle φ. An anti-reflection coating on conical prism lens 150 is therefore preferred, especially when the angle φ is large.

As shown in FIG. 2a, light ray 140d strikes parabolic reflector 122 near apex 123, is reflected parallel to axis of symmetry 142, passes just outside of light shield 126, and is refracted as it enters and exits conical prism lens 150 such that exiting ray 140d" is displaced radially inwards by a distance r from incident ray 140d. Thickness h and index of refraction n of lens 150 are chosen such that the radial displacement d of the ray 140αV140d" is equal to the radius r of the projection of light shield 126 along axis of symmetry 142. Therefore, ray 140d" travels along axis of symmetry 142, as shown in FIG. 2a, and is located at the center of output beam 140, as shown in FIG. 2b.

Another light ray 140a is emitted from gap 130 of arc lamp 124, passes near the edge of light shield 126, is reflected from parabolic reflector 122, and travels parallel to axis of symmetry 142. Ray 140a is refracted by conical prism lens 150 to produce ray 140a' which travels therethrough. On exiting, ray 140a' is again refracted to produce ray 140a" which travels parallel to axis of symmetry 142 and is also radially displaced inwards by a distance r. As shown in FIG. 2b, ray 140a' lies at the outer boundary of beam 140, since light rays cannot strike parabolic reflector 122 any farther from apex 123 of parabolic reflector 122 due to light shield 126.

Also shown in FIGS. 2a and 2b are light rays 140b and 140c which exit gap 130 at angles between that of previously discussed rays 140a and 140d. They too strike parabolic reflector 122, are reflected parallel to axis of symmetry 142, are refracted on entering and leaving conical prism lens 150, and exit traveling parallel to axis of symmetry 142. Beam 140 is therefore a column of light with no central hole, suitable for uses requiring a uniform illumination. It should be noted that conical prism lens 150 produces a non-conformal mapping since a multiplicity of points on the edge of the hole of FIG. lb are mapped to the central point 140d"/142 of FIG. 2b.

Fresnel lens 250 shown in cross-section in FIG. 2c is the Fresnel equivalent of conical prism lens 150 of FIG. 2a. Fresnel lens 250 is formed by slicing conical prism lens 150 of FIG. 2a into annular sections and aligning these sections substantially in a single plane. FIG. 2d shows the top portions of five complete sections 261-265 and one partial section 266 in cross-section. Fresnel lens 250 has cylindrical symmetry about optical axis 242, and front and back faces 251 and 252, respectively, of each section 261-266 is sloped at an angle ø from the radial direction.

The paths of a first pair of light rays 241 and 242 and a second pair of light rays 243 and 244 (referred to collectively by reference numeral 240) through Fresnel lens 250 are traced in FIG. 2c. Each ray 240 strikes the conical element at an angle φ from the normal of the front conical surface 251. The rays 240 are refracted by the larger index of refraction n of Fresnel lens 250 to produce rays 241724272437244' which pass through Fresnel lens 250 at an angle θ from the normal, where according to Snell's law: sin φ = n sin θ . On exiting Fresnel lens 250 the rays are again refracted according to Snell's law, and since front and back surfaces 251 and 252 are parallel, exiting rays 241"/242"/243"/244" again travel parallel to axis of symmetry 242. As before, upon solving for the inward radial displacement d of the light upon passing through a single section Fresnel lens 250, it is determined that d = h * sin φ * [1 - cos φ * (n² - sin² φ)^{"1 2}] .

As with conical prism lens 150 of FIG. 2a, the thickness h and the index of refraction n of Fresnel lens 250 is chosen such that the radial displacement d of rays passing through central section 261 is equal to the radius r of the projection of light shield 126 of FIG. 2a along the axis of symmetry 242, thereby removing the hole from the beam 240. However, this relationship for d does not hold when a ray enters Fresnel lens 250 in one section and exits in another section. For instance, ray 242 enters in the third section 263 and exits in the second section 262 as ray 242". Therefore, ray 242 effectively experiences a lens of a greater thickness than h, and is displaced by a greater distance than that given in the expression above. However, adjacent ray 241 enters the lens in third section 263 and also exits from the third section 263 as ray 241". Therefore, as shown in FIG. 2d, from the uniform brightness beam 240 incident on the Fresnel lens 250, Fresnel lens 250 produces a beam 240" having concentric circular dark bands 270, 271, 272, etc. Each dark band 270, 271, 272, etc., has an outer radius equal to the radius of the boundary between sections 261, 262, 263, etc., of Fresnel lens 250. However, such dark bands 270, 271, 272, etc., still allow a uniform brightness beam to be attained. Since, as indicated in FIG. 2c, the divergence angle ψ of rays 240" leaving Fresnel lens 250 is generally on the order of 5_ due to the finite size of emission gap 130, a blurring effect will occur when p » q/sin ψ , where p is the distance of the illumination plane of beam 240" from Fresnel lens 250, and q is the thickness of the dark bands 270, 271, 272, etc., so that the beam again becomes of uniform brightness.

A transmissive projector system 380 using lamp 120 is shown schematically in FIG.

5. Arc lamp 124 is located at the focus of parabolic reflector 122, so that light emitted therefrom is reflected to form a beam 140 traveling parallel to axis of symmetry 142. Light shield 126 intercepts light which would otherwise not be reflected by parabolic reflector 122, thereby insuring that all light 140 emanating from the arc lamp 124/reflector 122 system is collimated. However, collimated light 140 from parabolic reflector 122 has a central hole, as shown in FIG. lb, since light shield 126 also intercepts light traveling along a path close to axis of symmetry 142. Conical prism lens 150 acts to shift all rays of reflected beam 140 closer to axis of symmetry 142 to produce a beam 140" that does not have a central hole.

Arc lamp 124, parabolic reflector 122 and conical prism lens 150 produce collimated white light having red, blue and green components r, g and b, respectively. The light is directed to a top left dichroic filter mirror 385 which reflects red component r to- wards a lower left mirror 388, but allows green component g and blue component b to pass through to a top center dichroic filter mirror 386. Top center dichroic filter mirror 386 allows blue component b to pass through to a blue-component image forming element 393, and reflects green component g towards a green-component image forming element 392. Red component r is reflected from a mirror 388 to a red-component image forming element 391. Image forming elements 391, 392 and 393 may be made of encapsulated liquid crystal material, as is described in U.S. Patent Nos. 4,435,047 (1984), 4,606,61 1 (1986), 4,616,903 (1986), and 4,707,080 (1987), all to Fergason; U.S. Patent Nos. 5,075,789 (1991), 5,136,403 (1992), 5,138,472 (1992), and 5,175,637 (1992), all to the present inventor as the sole inventor or a coinventor; U.S. Patent Nos. 4,671,618 (1987), 4,673,255 (1987), 4,685,771 (1987), 4,688,900 (1987), all assigned to Kent State University; U.S. Patent Nos. 4,834,509 (1989), 4,818,070 (1989), 5,162,934 (1992), all assigned to Asahi Glass; and European Patent Application No. EP 313,053 (1989), assigned to Dainippon Ink and Chemicals; and each of these patents is hereby incorporated herein by reference. During operation, image forming elements 391, 392 and 393 have pixels (representative pixels being identified in the figure by reference numerals 391a, 392a, and 393a) which are independently switchable between light transmitting and light reflecting states by computerized control system 399 connected to the image forming elements 391, 392 and 393 via lines 396, 397 and 398, respectively. Image-forming elements 391, 392, and 393 also may be made with twisted nematic liquid crystal cells, as is well known in the art. Then, the red, green and blue images from red-, green-, and blue-component image forming elements 391, 392 and 393 are combined to provide a colored image. First, a bottom center dichroic filter mirror 389 combines red component r and green component g by allowing red component r to pass through and reflecting green component g. Then red component r and green component g are combined with blue component b by a bottom right dichroic filter mirror 390 which allows red component r and green component g to pass through, and reflects blue component b, thereby providing a full-color image. The full-color image is incident upon lens 94 which focuses the image on a screen 395.

A reflective projector system 400 using lamp 120 is shown schematically in FIG. 6. Beam 140" from lamp 120 is focused by field lens 410 onto mirror 415 which directs focused light 440 to dichroic cube 420. The operation of dichroic cubes is described in detail in Sonehara, U.S. Pat. No. 5,098,183 (1992), Kurematsu et al., U.S. Pat. No. 5,170,194 (1992), Planner et al., U.S. Pat. No. 5,172,222 (1992), and copending, commonly assigned allowed application of Jones, Serial No. 08/074570, filed Jun. 7, 1993; and these disclosures are incorporated herein by reference. Briefly, dichroic cube 420 has three liquid crystal display cells 422r, 422b, and 422g (referred to collectively by reference numeral 422) mounted on faces of the cubes. Display cells 422 have pixels (not shown) which may be separately controlled by a computerized control means (not shown) so as to reflect an image. Dichroic cube 420 uses dichroic mirrors to separate incident white light 441 into red, blue and green components which are directed to red, blue and green display cells 422r, 422b, and 422g, respectively. The red, blue and green images from display cells 422r, 422b and 422g are then recombined by the dichroic mirrors and a beam 443 having the full-color image exits dichroic cube 420. Beam 443 is focused by projection lens system 430 to provide imaged beam 445 which is projected onto screen 450 for viewing. Projectors 380 and 400 may be either of the front projection or rear projection type. It should be noted that although conical prism 150 of FIGS. 2a and 2c has the desired effect of eliminating the hole in the beam 140 produced by light shield 126, the intensity of light diminishes with radial displacement. This is illustrated by FIG. 7a which shows path traces of a first pair of rays 710a and 710b (referred to collectively by reference numeral 710) and a second pair of rays 720a and 720b (referred to collectively by reference numeral 720). The rays in each pair 710 and 720 are separated by the same angle δφ as they emanate from gap 130. The pair of rays 710 which strike parabolic reflector 122 closer to apex 123 and are reflected closer to axis of symmetry 142 have a smaller radial separation δr₂ than the radial separation δri of the pair of rays 720 which strike parabolic reflector 122 farther from apex 123 and are reflected farther from axis of symmetry 142. A similar effect occurs azimuthally, as is diagramatically illustrated in the front view of parabolic reflector 122 of FIG. 7b. As shown there, a first pair of light rays 730 emanating from gap 130 with an azimuthal separation angle δμ are separated by a smaller distance δr₃ when they reflect from parabolic reflector 122 at a point near apex 123 than a second pair of light rays 740 which also emanate at an azimuthal separation angle δμ but reflect from parabolic reflector 122 at a point farther from apex 123 and therefore are separated by a larger distance δr₄. Therefore, the light emanating from gap 130 in a unit of solid angle which reflects from parabolic reflector 122 near apex 123 covers a smaller area than light emanating from gap 130 in a unit of solid angle which reflects from parabolic reflector 122 farther from apex 123. Therefore, as shown in the plot of flux density L(r) versus radius r of FIG. 7c, there is a decrease in the flux with increasing radial distance. However, this effect can be compensated for by using a prism 750 with a corrective curvature as shown in FIG. 7a. In this case the angle between the normal to outer surface 752 (and inner surface 751) of conical prism lens 750 and axis of cylindrical symmetry 142 is a monotonically increasing function of the radial distance. (Although conical lens 750 has a corrective curvature, it will be referred to herein as a "conical prism" because, as described below, conical prism lens 750 is designed such that each ray of light which strikes it traveling parallel to axis of symmetry 142 also exits it parallel to axis of symmetry 142. Furthermore, each ray of light which strikes conical prism lens 750 slightly displaced from parallel to axis of symmetry 142, also exits it slightly displaced from parallel to axis of symmetry 142.) Therefore, rays 720 which strike conical prism lens 750 farther from axis of symmetry 142 refract with a larger angle and are displaced by a larger distance. As shown in magnified view of the path of ray 720a of FIG. 7e, conical prism lens 750 is constructed such that, for a ray 720a traveling parallel to axis of symmetry 142, inner surface 751 where ray 720a' exits is parallel to outer surface 752 where ray 720a entered conical prism lens 750. As shown in FIG. 7a, the first pair of rays 710 are separated by a distance δr₂ prior to incidence on the prism. Near axis of symmetry 142 the curvature of the surfaces of conical prism lens 750 are relatively small and so rays 710" exit conical prism lens 750 separated by a distance δr₂' which is not substantially smaller than δr₂. However, since the second pair of rays 720 strike conical prism lens 750 farther from axis of symmetry 142 where the curvature is more pronounced, there is a greater difference between separation distance δr, of rays 720 incident on conical prism lens 750 and separation distance δr₃ of rays 720" exiting conical prism lens 750. Therefore, the curvature of conical prism lens 750 acts to concentrate the light farther from axis of symmetry 142 to provide a more uniform output flux density L"(r), as shown in FIG. 7d.

For conical prism lens 750 to eliminate the hole in the beam produced by a light shield, it is necessary for light which is reflected by parabolic reflector 122 and passes just outside the radius r of the shield to be refracted so that it exits conical prism lens 750 along the axis of symmetry. Therefore, the relation r = d = h * sin φ * [1 - cos φ * (n² - sin² φ)^"1/2] , where h is the thickness of conical prism lens 750, φ is the angle between the normal to surface 752 of conical prism lens 750 and the radial direction, n is the index of refraction of conical prism lens 750, d is the radial displacement of light rays on passing through conical O 96/13943 PC17US95/14109

prism lens 750 and r is the radius of the light shield, must at least hold for the light rays which exit conical prism lens 750 along the axis of symmetry. Because conical prism lens 750 does not necessarily have a uniform thickness h, the above relation must hold for an effective thickness. As shown in FIG. 7e, the effective thickness h' is the projection of the distance ray 720a' travels through conical prism lens 750 on normal vector 762 to the surface of conical prism lens 750 at point 764 where ray 720a enters (which by construction is the same as the projection of the distance ray 720a* travels through conical prism lens 750 on normal 768 to the surface of conical prism lens 750 at point 766 where the ray 720a' exits). Algebraically, this condition is expressed as d f(p) d g(p-d) dp dp where f(p) is the curve describing the outer surface 752 and g(p) is the curve describing the inner surface 751 as a function of radial distance p.

The output flux density L"(p) is determined by equating the total flux incident on conical prism lens 750 on a ring of radius p and width δp with the total flux produced by conical prism lens 750 on a ring of radius (p-d) and width δp * (1-dd/dp). This provides the output flux density L"(p) given by

T Υ_rΛ = I , _Λ^ 2π p δρ - ΛPJ _2π (p-d) δp (1-dd/dp)

^{= L(p)} * (p-d) (1-dd/dp) ^•

The present invention is not limited to the use of a parabolic reflector. For instance, FIG. 8a shows a cross-sectional view of a lamp 500 where a bi-concave collimating lens 545 is used to collimate light 540 from an elliptical reflector 522 prior to incidence on a conical prism lens 550. A light source 524, such as an arc lamp, located at a focus of the elliptical cross-sections of elliptical reflector 522 produces light which is reflected towards the other focus of the elliptical cross-sections. A shield (not shown) mounted on light source 524 prevents light from exiting lamp 500 without striking reflector 522 and produces a hole in converging light 540 which makes the beam of converging light 540 unsuitable for use as a source of illumination for a projector. Converging light 540 is dif¬ fracted by bi-concave collimating lens 545 to produce collimated light rays 540' which still has a central hole. Conical prism lens 550 displaces collimated light rays 540' towards axis of symmetry 542 by an amount equal to the radius of the hole incident on conical prism lens 550, thereby producing a collimated beam 540" without a hole. Therefore, bi-concave collimating lens 545 and conical prism lens 550 form a lens system 560 which acts to collimate the beam of converging light 540 and remove the hole therefrom.

The positions of conical prism lens 550 and bi-concave collimating lens 545 in the lamp 500 may be reversed. FIG. 8b is a cross-sectional view of a lamp 600 where conical prism 650 is used to remove a hole in converging light 640 from elliptical reflector 622, and then a bi-concave collimating lens 645 is used to collimate light 640' exiting conical prism lens 650. As before, a light source 624, such as an arc lamp, located at a focus of the elliptical cross-sections of the elliptical reflector 622 produces light which is reflected by elliptical reflector 622 towards the other focus of the elliptical cross-sections. A shield (not shown) mounted on the light source 624 prevents light from escaping from lamp 600 without striking elliptical reflector 622, and produces a hole in converging light 640 which makes the beam of converging light 640 unsuitable for use as a source of illumination for a projector. Converging light 640 is displaced towards axis of symmetry 642 by conical prism lens 650, thereby removing the hole in the beam of converging light 640. Because the angle between the incident rays of converging light 640 and the normal to the surface of conical prism lens 650 of FIG. 8b is smaller than the angle between the incident rays of converging light 540 and the normal to the surface of conical prism lens 550 of FIG. 8a, the reflective losses are smaller in lamp 600 of FIG. 8b than in lamp 500 of FIG. 8a. It should be noted that a correction to the conical surface of conical prism lens 650 will be required because of the differing angles of impact of the rays of converging light 640 with radial displacement. Because outer surface 651 and inner surface 652 of conical prism lens 650 are parallel, each ray 640' exits conical prism lens 650 parallel to the direction ray 640 entered it, so in this case the rays 640' exiting the conical prism lens 650 are converging. Bi-concave collimating lens 645 then acts on the converging beam 640' to produce collimated beam 640" where the rays of beam 640" travel parallel to each other. Bi¬ concave collimating lens 645 and conical prism lens 650 therefore form a lens system 660 which acts to collimate the beam of converging light 640 and remove the hole therefrom.

The hole elimination system can be a mirror system, instead of a lens system. FIG. 9a shows a schematic cross-section of such a lens system. Lamp 820 comprises an arc lamp 824 as the source of light, a parabolic reflector 822 for collimating light, and a light shield 826 for re-directing forwardly emitted light back towards parabolic reflector 822. (For the sake of avoiding clutter, other elements of lamp 820, corresponding to those depicted for lamp 120 in FIG. la, have been omitted.) For the reasons previously explained, the collimated light beam (represented by light rays 840a through 840f) exiting parabolic reflector 822 has a central hole. This hole can be eliminated by a mirror system comprising a frusto-conical mirror 850 and a conical mirror 860, each disposed symmetrically around cylindrical axis of symmetry 842. Frusto-conical mirror 850 reflects outer light rays 840a and 840f inwards, in a direction generally perpendicular to axis of symmetry 842. Upon impinging on conical mirror 860, reflected rays 840a' and 840f are reflected thereby in a direction parallel to axis of symmetry 842 as light rays 840a" and 840f '. Together with undisplaced rays 840b through 840e, rays 840a" and 840f ' form a collimated beam of light without a central hole therein.

FIG. 9b shows a possible construction of the mirror system of the previous figure. A frusto-conical piece 870, made of a transparent material such as glass or plastic (e.g„ polycarbonate or acrylic) has a central conical recess 872 formed therein. Frusto-conical reflector 850 can be formed on exterior surface 874 by deposition thereon of a suitably reflective material. Similarly, conical reflector 860 can be formed on conical interior surface 876. Those skilled in the art will appreciate that FIG. 9b illustrates only one possible construction, and that many other equivalent constructions are possible, for example by using metal or dielectric reflectors or by using total internal reflection at glass/air interfaces.

An alternative mirror system is depicted schematically in FIG. 10. Lamp 920 comprises arc lamp 924, parabolic reflector 922, and light shield 926, which interact in a manner previously described to produce a collimated beam of light (represented by light rays 940a through 940f) having a central hole. However, each of light rays 940a through 940f are inwardly reflected, in a direction perpendicular to cylindrical axis of symmetry 942, by frusto-conical mirrors 950a through 950c. However, each ray is again reflected by the next inner frusto-conical mirror, this time in a direction parallel to axis of symmetry 942. For example, ray 940a, after reflection by frusto-conical mirror 950a, is next reflected by frusto-conical mirror 950b. For light reflected by the innermost frusto-conical mirror 950c (i.e., light rays 940c and 940d), the second reflection is effected by conical mirror 960, but also in a direction parallel to axis of symmetry 942. In this manner, the net effect is that each ray of light has been displaced inwardly, by a distance equal to about equal to the radius of the central hole, thereby filling it. It will be appreciated that the interior frusto-conical mirrors (950b and 950c) must be reflective on both their exterior and interior surfaces, but that the outermost frusto-conical mirror (950a) need be reflective only on its interior surface, although as a matter of manufacturing convenience it may also be made reflective at its exterior surface. It will also be appreciated that in this particular instance a series of three frusto-conical mirrors 950a through 950c has been depicted, but that a different number of such mirrors may be used. So far the hole elimination system (whether lens or mirror) has been described as cylindrically symmetric. So the inward displacement of light is independent of rotation around the polar axis. It will be clear to a person skilled in the art that the degree of radial displacement of the light can be made a function of the polar angle of rotation around the optic axis. In particular, in projecting images of TV or computer screens of rectangular aspect ratio around 4:3, light efficiency can be improved by having the degree of inward deflection and compression increased for the radial direction that is normal to the longest side of the rectangular image. This becomes much more important for wide screen formats like HDTV at 16:9, and will increase the light efficiency more. A number of differing approaches can be used to accomplish this non- cylindrically symmetric result. One is to draw the desired element profiles for the top and side views; and then to interpolate the intermediate x-sections using eg. an elliptical variation. One is to draw the required views at more than 2 polar angles and to linearly interpolate the intermediate x-sections. Whichever system is used, it is important to avoid sharp discontinuities in form which could lead to singularities in the output light pattern e.g. cusps. In other words, it is not necessary the hole elimination system (and the beam its generates) be cylindrically symmetric about the central axis of symmetry (e.g., elements 142, 542, 642, 842, and 942); they need only be at least bilaterally symmetric about the axis, for example with the hole- eliminated beam having a rectangular or elliptical cross section.

In summary, an apparatus for elimination of a hole in a light source using a reflector has been described in terms of preferred embodiments. The invention, however, is not limited to the embodiments depicted and described, since many variations are within the scope of the present invention. For instance: the conical prism of FIG. 3 may have other shapes near the apex since no light is incident on this region; although an arc lamp is preferred because of its small size, other types of light sources, such as fluorescent and incandescent bulbs, may be used; any shape of reflector may be used as long as the lens system, which includes the conical prism lens, acts to collimate the beam, etc. Further, the foregoing detailed description of the invention includes passages which are chiefly or exclusively concerned with particular parts or aspects of the invention. It is to be understood that this is for clarity and convenience, that a particular feature may be relevant in more than just passage in which it is disclosed, and that the disclosure herein includes all the appropriate combinations of information found in the different passages. Similarly, al¬ though the various figures and descriptions thereof relate to specific embodiments of the invention, it is to be understood that where a specific feature is disclosed in the context of a particular figure, such feature can also be used, to the extent appropriate, in the context of another figure, in combination with another feature, or in the invention in general. Therefore, it should be understood that the scope of the invention is defined by the appended claims.

Claims

Claims What is claimed is:

1. A lamp for providing a collimated beam of light comprising: a light source having an emission point from which source light is emitted; a reflector for reflecting said source light to produce reflected light; a shield mounted on said light source to intercept said source light emitted in directions away from said reflector, said shield also intercepting said reflected light to produce a pre-lens hole in said reflected light having a pre- lens hole center and a pre-lens hole width; and a hole elimination system for displacing light rays in said reflected light inwardly towards said pre-lens hole center to produce said collimated beam of light, said collimated beam of light being without a hole.

2. The lamp of claim 1 wherein said reflector is a parabolic reflector having a reflector axis of cylindrical symmetry, said light source being mounted in said reflector along said reflector axis of symmetry, and said emission point being positioned at a focus of said parabolic reflector so that said reflected light travels substantially parallel to said reflector axis of cylindrical symmetry.

3. The lamp of claim 2 wherein said hole elimination lens system includes a conical prism lens with a prism axis of cylindrical symmetry coincident with said reflector axis of cylindrical symmetry.

4. The lamp of claim 3 wherein said pre-lens hole is circular with a radius r, said conical prism lens at radial distance r has a thickness h, an index of refraction n, and an angle φ between said prism axis of symmetry and an outside surface of said conical prism, such that r « h * sin φ * [1 - cos φ * (n² - sin² φ)^{*1 2}] .

5. The lamp of claim 4 wherein said index of refraction n is approximately 1.5, said angle φ is approximately 30°, and said thickness h is approximately 25 mm.

6. The lamp of claim 3 wherein said conical prism lens has an exterior surface described by a function f(p) and an interior surface described by a function g(p) where p is a radial distance and O 96/13943 PC17US95/14109

d f(p) _ d g(p-d) dp dp ' where d is a radial displacement of a light ray due to said conical prism lens.

7. The lamp of claim 6 wherein df(p)/dp increases with increasing p, whereby a flux density of an outer region of said collimated bean is increased relative to said collimated beam produced when df(p)/dp is a constant.

8. The lamp of claim 1 wherein said hole elimination lens system includes a Fresnel lens.

9. The lamp of claim 8 wherein said Fresnel lens produces a banded beam having dark bands of thickness q immediately adjacent said Fresnel lens, light rays in said banded beam having a divergence angle ψ, wherein said banded beam, when incident on an illumination plane p positioned at a distance p from said Fresnel lens, where the inequality p » q/sin ψ is satisfied, provides substantially uniform illumination on said illumination plane.

10. The lamp of claim 1 wherein said reflector is an non-parabolic reflector having a reflector axis of cylindrical symmetry, said light source being mounted in said non- parabolic reflector along said reflector axis of cylindrical symmetry, and said hole elimination lens system including a conical prism lens with a prism axis of cylindrical symmetry coincident with said reflector axis of cylindrical symmetry and a collimating lens.

11. The lamp of claim 10 wherein said collimating lens is located between said non- parabolic reflector and said conical prism lens, said collimating lens collimating said reflected light.

12. The lamp of claim 10 wherein said conical prism lens produces a holeless beam from said reflected light incident on said conical prism lens, and said collimating lens produces said collimated beam of light from said holeless beam incident on said collimating lens.

13. The lamp of claim 10 wherein said non-parabolic reflector is an elliptical reflector.

14. The lamp of claim 1 wherein said collimated beam produced by said hole elimination lens system has substantially uniform brightness.

15. The lamp of claim 1 wherein a projection from said emission point through an inner edge of said shield falls upon said reflector.

16. The lamp of claim 1 , wherein said hole elimination system is a mirror system.

17. The lamp of claim 16, wherein said mirror system comprises a frusto-conical mirror and a conical mirror, said frusto-conical mirror being disposed symmetrically around said conical mirror.

18. The lamp of claim 16, wherein said mirror system comprises a conical mirror and a plurality of frusto-conical mirrors successively and symmetrically disposed around said conical mirror.

19. An apparatus comprising : a lamp for providing a collimated beam of light, said lamp including a light source having an emission point from which source light is emitted; a reflector for reflecting said source light to produce reflected light; a shield mounted on said light source to intercept said source light emitted in directions away from said reflector, said shield also intercepting said reflected light to produce a pre-lens hole in said reflected light having a pre-lens hole center and a pre-lens hole width; and a hole elimination lens system for displacing light rays in said reflected light towards said pre-lens hole center by a distance equal to half said pre- lens hole width to produce said collimated beam of light, said collimated beam of light being without a hole; and a means for image forming, said collimated beam of light being incident on said means for image forming, said means for image forming including an electrically-modulated light scattering material.

20. The apparatus of claim 19 wherein said means for image forming includes an encapsulated liquid crystal material.

* * * * * * * * * *