GB2385191A

GB2385191A - Backlight

Info

Publication number: GB2385191A
Application number: GB0203023A
Authority: GB
Inventors: Jason Slack
Original assignee: Screen Technology Ltd
Current assignee: Screen Technology Ltd
Priority date: 2002-02-08
Filing date: 2002-02-08
Publication date: 2003-08-13
Also published as: GB0203023D0

Abstract

A backlight for a liquid crystal display includes a plurality of substantially telecentrically collimated radiation sources 10. Light from sources 10 is homogenised by a biconvex lens array 30C, which is made up of a first array of positive lenses arranged to receive radiation from the radiation sources, and a second array of positive lenses located at the focal plane of the first lens array, such that a substantial portion of the radiation from the radiation sources is focussed by each lens of the first array onto a corresponding lens of the second array. Once the light leaving lens array 30C has propagated beyond the focal length of the second array it has a spatially uniform intensity, and provides uniform backlighting to a liquid crystal display situated beyond this point.

Description

BACKLIGHT This invention relates to backlights for displays and to displays using such backlights.

It is well known that the brightness uniformity across the surface of a displayed image is very important to the perceived quality of the image. It is therefore important in a display system requiring a backlight (such as a liquid crystal display or"LCD") that the backlight illuminates the display, and therefore the image to be displayed, evenly.

LCDs operate most effectively with well-collimated backlights. Conventional LCD backlights typically use diffusion screens to homogenise radiation generated by a radiation source or sources prior to it being imaged onto a target liquid crystal screen. the liquid crystal screen then modulates the radiation for viewing (in the case of visible radiation) or modulates it so as to excite an array of phosphors (in the case of ultraviolet light in a photoluminescent LCD system).

The term"homogenisation"in this context means to remove at least some artefacts and to flatten the intensity distribution of the radiation. Diffusion homogenisers suffer from the disadvantage that they tend to scatter the radiation over a wide range of angles.

In particular, the scattering characteristic of a diffusion screen substantially reduces the collimation of the light reaching a target screen. Well collimated (i. e. low divergence) radiation is an important requirement for backlighting because most modulation devices such as LCD panels operate better with collimated backlight radiation. In the case of an image projector, the directivity provided by well collimated radiation also allows the use of projection lenses with larger f-numbers where required, which are smaller and less expensive than corresponding lenses having lower f-numbers.

It is known to use so-called"fly's eye"lens arrays (two-dimensional arrays of lenses, which may be substantially spherical or aspheric) in combination with a lamp to provide homogeneous backlighting within a projection apparatus. A typical arrangement for this type of apparatus is shown in Figure 1 of the accompanying drawings. The apparatus illustrated in Figure 1 comprises a lamp source 1 coupled to

a parabolic reflector 2, two fly's eye arrays 3,4 providing a degree of homogenisation, an LCD screen 6, and a collector lens 5 to relay the homogenised light onto the LCD screen.

This invention provides a backlight comprising: a plurality of substantially telecentrically collimated radiation sources; a first array of positive lenses, arranged to receive radiation from the radiation sources; and a second array of positive lenses, located at the focal plane of the first lens array such that a substantial portion of the radiation from the radiation sources is focused by each lens of the first array onto a corresponding lens of the second array.

The invention overcomes or at least alleviates many of the disadvantages of the prior art by providing a mechanism for redistributing (rather than attenuating) radiation from a plurality of radiation sources by using a system of lenses rather than (for instance) a diffusion screen. The lenses of the present invention can exhibit lower absorption losses than components such as the diffusion screen of the prior art.

Embodiments of the invention can also advantageously produce better (and more controllable) collimation than that provided for by conventional backlights using lamps and diffusing screens.

The operation of a fly's eye lens array is well documented-for example in EP-A-0753780. However, while it is known that fly's eye arrays have been used in LCD data projectors, in these previous applications the homogeniser has been used with only a single light source. A secondary'collector lens'has also been required to relay the homogenised light onto the display device. In contrast, in the invention, a plurality of light sources is used (e. g. an array of light sources) with a fly's eye array, avoiding the need for a collector lens previously required to image onto the liquid crystal panel effectively.

The skilled man will understand that within the meaning of the present application, telecentricity is defined to mean that light illuminating an object such as a lens or display has its principal axis substantially perpendicular to the plane of the object and the intensity. distribution of rays around the principal axis is substantially the same at all points on the surface of the object. More generally, a telecentric system is one in which one or both of the pupils are located at infinity and the principal ray

enters or leaves the system parallel to the axis. Another way to think of this is to remember the definition of a lens focal point. Rays of light that are parallel to the optical axis of a lens will enter the lens and converge to the focal point. At least this is true for the paraxial or"thin lens"case. Also, rays of light that enter a lens after passing through the focal point will emerge parallel to the optical axis. So, by definition, if a small aperture or stop is placed at a lens focal point, only light that is nearly parallel to the optical axis can either enter or emerge from the system, depending at which focal point the stop is placed. Such an arrangement can be called telecentric, and the aperture is the telecentric stop. Systems can have telecentric light entering the system (object space), exiting the system (image space) or both.

Various other respective aspects and features of the invention are defined in the appended claims. Features from the dependent claims may be combined with features of the independent claims as appropriate and not merely as explicitly set out in the claims.

Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings in which: Figure 1 illustrates a known backlighting arrangement for a projection apparatus using fly's eye lens arrays and a collecting lens ; Figure 2 is a schematic diagram illustrating a backlighting arrangement according to an embodiment of the invention; Figure 3 illustrates the geometric relationship between the telecentric input and the output of a fly's eye homogeniser; Figure 4 illustrates an embodiment of the invention comprising two lens arrays; Figure 5 illustrates another embodiment of the invention, in which the two lens arrays are joined to make a single biconvex lens array; and Figure 6 illustrates an embodiment of the invention in which the radiation sources are an array of substantially point sources collimated using a parabolic reflector.

In the following discussion reference will be made to both ideal/theoretical situations and the situation in reality. It should be appreciated that reference is made to ideal/theoretical situations for the purposes of clarity but that the broad principles involved are also applicable to the situation in reality. To this end, the skilled man will

of course realise that terms representing the performance of an ideal theoretical system such as"perfectly collimated"are not to suggest that a physical embodiment of the invention is capable of such idealised performance. Instead, the skilled man will understand the manufacturing and alignment constraints which place an upper limit on the performance of a real embodiment of the invention.

Referring now to Figure 2, a back-lighting system comprises a plurality of radiation sources (10) which are optically coupled to a collimator (20). The radiation sources and the collimator are shown in highly schematic form. In the embodiment, the radiation sources could, for example, be Light Emitting Diodes (LEDs) incandescent lamps or fluorescent lamps and the emitted radiation could be, for example, visible or ultraviolet. The collimator could, for example comprise parabolic reflectors optically coupled to one or more respective radiation sources. In the illustrated arrangement a single schematic collimator receives radiation from the whole array of radiation sources. However, other arrangements may comprise a plurality of collimators, with one collimator being arranged to receive radiation from a subset of the radiation sources. In yet other arrangements a plurality of collimators may each be arranged to receive radiation from a single radiation source.

The combination of radiation sources and collimators serves to provide a substantially collimated radiation output. Due to the finite extent of each radiation source (i. e. it is assumed not to be a perfect point source) the collimated radiation from each source will exhibit some degree of divergence, having an angular intensity profile that may be represented for instance by a Gaussian distribution. An ideal backlight radiation source would have a spatial intensity distribution at the target that is uniform (i. e. the same intensity of radiation at all points on the surface of the target).

In a practical system, of course, the radiation from the radiation sources, both before and after collimation, may be subject to undesirable inhomogeneities in brightness, including non-uniform intensity profiles and geometric artefacts arising from manufacturing or alignment defects in the radiation sources and the collimator. The substantially collimated radiation therefore needs to have such inhomogeneities removed (as far as practically possible) if uniform illumination is to be provided by the apparatus. This is carried out by optically coupling the array of radiation sources and collimators to a fly's eye integrator (homogeniser) 30.

Again, the fly's eye integrator is shown highly schematically in Figure 2. Its operation will be discussed with reference to Figure 3 below, and then two ways of implementing the fly's eye integrator will be described with reference to Figures 4 and 5 below.

In the present arrangement, when the substantially collimated radiation is transmitted through the homogeniser, it is homogenised to have a more spatially uniform intensity. The homogenised radiation is then projected onto a target display such as an LCD screen 40. Radiation reaching the target display 40 will originate from one or more of the radiation sources depending on the configuration of the apparatus. Specifically, the greater the separation between the display and the backlight the greater the overlap (at the target) between radiation from adjacent radiation sources, due to the divergence of the radiation output from the second lens array. Therefore, it may be that certain arrangements provide for regions of the surface of the target screen to receive radiation from any reasonable number of radiation sources (limited of course by the number of radiation sources present in the arrangement).

An important feature of the fly's eye array 30 in embodiments of the invention is that two lens arrays are provided, being physically separated by a distance equal to the focal length of the lenses receiving radiation from the radiation sources. In the schematic drawings of Figures 2 to 6, this implies the focal length of the lenses drawn on the left-hand side of the fly's eye array 30.

The optical operation of a single pair of elements from a fly's eye homogeniser comprising two lens arrays will now be described with reference to Figure 3. Two individual lenses are shown, with a left-most lens LI representing an element of the first lens array to receive radiation from the radiation sources and a right-most lens L2 representing a corresponding element of the other lens array. The two lenses are here shown a distance f apart, corresponding to the focal length of the lens LI (which should preferably be substantially the focal length of the lens L2). A parameter D is the diameter of each lens element, which for the present arrangement is the same for the corresponding lenses of both lens arrays.

In terms of physical size, a typical diameter D would be 250pm with a focal length of 740um Because of this relatively small size, these arrays are often referred to as"microlens"arrays.

Schematic rays representing telecentric input radiation 100 of collimation 0 are shown as incident upon the centre and outermost points of the lens LI. The rays shown in Figure 3 serve to illustrate the principal ray (i. e. parallel to the optical axes of lenses L1 and L2) and the marginal rays (i. e. having a divergence from the principal ray of o) which make up a cone representing the telecentrically collimated radiation impinging on one point of the lens LI. It should be noted that this representation is purely illustrative and radiation will in reality be incident on lens LI all the way across. The principal rays of these three cones are imaged to the principal point 300 where the second lens L2 and the aperture stop 200 are located. The marginal rays are similarly imaged to points PI and P2 of lens L2. Rays having a divergence from the optical axis of the lenses of less than 0 but greater than zero will be imaged onto lens L2 between the principal point and PI or P2 depending on the magnitude and direction of the divergence.

Note that the f-number of the lens LI is equal to f/D (where f=focal length of lens and D=diameter of lens). It can also be seen from the diagram that the angle made by the cone of radiation is determined by this f-number of the lens, so the fnumber of the illumination and lens are matched. The mathematical relationship between the two is defined by f-number= 1/ (2tan (e)) The entrance pupil of the system is the image of the aperture stop 200 (i. e. PI, P2 and principal point) through the lenses preceding the stop (i. e. LI only in this example) and it can be seen that it is imaged at infinity to the left of LI. From this, it follows that the system has a telecentric input. As regards the output from lens L2, it can be seen that there are three schematic rays parallel to the optical axis and these ray bundles form an image at infinity. The output of the system is therefore also telecentric.

The lens L2 also forms an image of lens LI at infinity. It is common with the fly's-eye configuration to use a lens (a collector lens) to bring the image closer.

However, to do so requires a second lens at the image plane to make the light telecentric again.

So, two parallel rays impinging on lens LI would be focused to a point on lens L2. If these parallel rays are perpendicular to the plane of the lens LI, the parallel rays will be focused upon the principal point 300 of the lens L2. If these parallel rays are not perpendicular to the plane of the lens LI, the parallel rays will strike the lens L2 at a point other than the principal point 300. In either case, the rays that were parallel upon striking the lens LI will now be non-parallel, with an angular separation in dependence on their relative spatial separation upon striking the lens LI.

Two non-parallel rays impinging on the lens LI will be focused on the lens L2 at different points. The combination of the lenses LI and L2 will tend to cause the angle between the non-parallel rays to decrease. This applies in many cases, but not the case for instance in the boundary condition of two rays impinging on diametrically opposite edges of the lens LI.

It is important to note that incoming rays impinging on lens LI at such an angle that they will not be focused onto lens L2 are being ignored at this stage but will be discussed later. In Figure 3 such rays are stopped by the aperture stop (200).

The discussion of Figures 2 and 3 has related to schematic systems. Two practical embodiments of the invention, using the principles described with reference to Figures 2 and 3, will now be described with reference to Figures 4 and 5.

Referring now to Figure 4, a first arrangement comprises a first lens array 30a arranged to receive radiation from an array of spatially-separate, telecentrically collimated radiation sources (not shown). A second lens array 30b is situated at the focal plane of the first lens array 30a, there being a corresponding lens element in the second lens array 30b for each lens element in the first lens array 30a. Each pair of lens elements (i. e. each element in lens 1 and the corresponding element in lens array 2) can be considered as behaving as described above with reference to Figure 3, so that each such pair of lens elements in Figure 4 behaves as the lenses LI and L2 in Figure 3.

Each of the first lens elements in Figure 4 serves to project an image of the radiation source (not shown, but as with the other diagrams, positioned to the left of the first array 30a) onto the corresponding element in the second array 30b. The

second lens array 30b then focuses incoming light to infinity (the far field). Note that this equates to each element in the second lens array 30a projecting an image of the corresponding element of the first lens array 30a to infinity. In other words, radiation from the radiation sources is projected onto the entrance plane of each of the elements of the second lens array through the first lens array in such a way as to form a secondary radiation source image on each element of the second lens array. Further, an object such as a liquid crystal display (not shown, but as with the other diagrams, positioned to the right of the second array 30b) is illuminated with radiation emitted from the second lens array.

In the case of the incoming radiation being perfectly collimated and having a principal ray at a normal to the plane of the lens, the image formed by a lens element of the first lens array onto the corresponding lens element of the second lens array would be a single point at the centre of the lens (the principal point).

In the more realistic case of less imperfectly but well collimated radiation, the image formed may fill either a portion of, or the entire lens of the second array 30b.

Preferably the first and second lens arrays are configured such that the image of a radiation source generated by a lens element of the first lens array maps substantially over the whole area of the corresponding element of the second lens array. To achieve this, the second lens array is be situated at the focal plane of the first lens array and the degree of collimation of the collimated radiation is substantially matched by the fnumber of the first lens array as described above. For instance, for radiation having a divergence of 15 degrees (e. g. 90% of the incident radiation is within 15 degrees of the principal ray) an f-number of f/l. 9 is required in order for an element of the first lens array to project an image over substantially the whole face of a second lens element of the same size. Also, in the present arrangement the f-number of the lenses is matched to that of the radiation source such that radiation exiting from first lens array (2) is substantially incident only upon the corresponding elements of lens array (3).

Otherwise, radiation will leak into adjacent lens elements. This leakage causes collimation of radiation having a divergence greater than the f-number of the lenses, thus creating'secondary images'of the light source at the output of the second lens array. This increase in the collimation angle of the incident radiation source is due to

leaked radiation being diverted by the adjacent lens in the direction of the normal to the lens arrays.

It can be seen therefore that perfectly collimated radiation entering the homogeniser will exit with a degree of collimation corresponding to the f-number of the lens arrays. Similarly, it can be seen that radiation having a degree of collimation equal to the f-number of the lens arrays exits the homogeniser perfectly collimated.

This has the effect of homogenising the radiation locally within the beam (the fullwidth half maximum (FWHM) of the intensity distribution of the beam is increased).

Also, as the beam propagates away from the homogeniser, radiation originating from adjacent radiation sources also mixes so that the radiation is homogenised non-locally.

Each element of the second lens array serves to form an image of the corresponding element of the first lens array at infinity (i. e. in the far field). Due to the divergence of the radiation leaving the second lens array, the edges of these images (from the respective array elements) will overlap at a predetermined distance from the second array. This distance will depend (among other things) on the separation between the radiation sources. In other arrangements the edges of parabolic mirrors within which radiation sources operate are tessellated to provide an array of telecentrically collimated radiation sources with substantially no gap between them.

This reduces the distance from the second lens array at which the images of the second lens, and thus the intensity distributions of the radiation, overlap, in turn allowing for a more compact system.

So, the combination of the two lens arrays acts as a homogeniser. After the homogenised output has propagated a certain distance (which is greater than the focal length of the second lens array) it will exhibit a spatially uniform intensity. Therefore, when placed at such a distance (or greater) behind a transmissive display device such as a liquid crystal display, the apparatus according to this arrangement will provide substantially uniform backlighting of the display device. The distance at which spatial uniformity occurs is determined also by the packing density of the individual sources of radiation and decreases with increasing packing density.

In this and other arrangements each lens in the array 30a has its optical axis aligned with the corresponding element in the array 30b. Also the elements in the lens array 30a have substantially the same focal length as the elements in lens array 30b.

The lenses may be spherical, substantially spherical or aspheric.

Referring now to Figure 5, a second arrangement is illustrated in which two lens arrays consistent with those described with reference to Figure 4 are joined (or indeed fabricated as a unitary structure) to provide a single biconvex lens array 30c.

This arrangement is useful when the pitches of the lens arrays are sufficiently small to make the glass, plastics or other lens material of a practicable thickness. For instance, an f/3. 6 lens with a pitch (distance from lens centre to lens centre) of 200 urn would

have a thickness of 200 gm x 3. 6 x n, where n is the refractive index of the material.

In general the paraxial relationship defining the thickness of the biconvex array is n. f-number = t/D where t is the thickness of the biconvex array.

This arrangement has several additional advantages, including compacting the system, reducing system alignment requirements (i. e. the two lens arrays do not need to be aligned within a supporting structure because they are a single entity) and reducing weight.

The two operating surfaces of the fly's eye lens 30c shown in Figure 5 are separated by their focal length in the same way that the two lens arrays in Figure 4 are separated by the focal length of the first lens array. Of course, the fact that the lenses are separated by material other than air (i. e. having a refractive index other than unity) needs to be taken into account in designing their physical separation.

Referring to Figure 6, the radiation sources 10 may be an array of substantially point sources collimated using parabolic reflectors 20. The radiation sources may be for example, LED's enclosed within small parabolic reflectors. Such devices are available commercially and are suitable for coupling to the homogenisers described above. As in Figure 6, a single reflector may be arranged to project light upon a plurality of lens elements. Alternatively, individual reflectors could be arranged to project light upon respective individual lens elements in the array.

Claims

CLAIMS 1. A backlight comprising: a plurality of substantially telecentrically collimated radiation sources; a first array of positive lenses, arranged to receive radiation from the radiation sources; and a second array of positive lenses, located at the focal plane of the first lens array such that a substantial portion of the radiation from the radiation sources is focused by each lens of the first array onto a corresponding lens of the second array.
2. A backlight according to claim 1 in which the f-number of each lens within the first array of lenses is substantially matched to the divergence of the radiation from the radiation sources so that each lens in the first array images radiation over substantially the whole surface of the corresponding lens of the second array.
3. A backlight according to claim 1 or claim 2, in which each lens in the second array of lenses has its optical axis aligned with the corresponding lens in the first array of lenses.

1
4. A backlight according to any one of claims 1 to 3, in which each lens in the second array of lenses has substantially the same focal length as the corresponding lens in the first array of lenses.
5. A backlight according to any one of the preceding claims, in which the first and second lens arrays are arranged as a unitary structure so as to provide a single biconvex lens array.
6. A backlight according to any one of the preceding claims, in which the collimated radiation sources each comprise one or more radiation sources and one or more collimators.

<Desc/Clms Page number 12>
7. A backlight according to claim 6 in which the or each collimator is a parabolic mirror.
8. A backlight according to any one of the preceding claims, in which the radiation is visible light.
9. A backlight according to any one of claims 1 to 7, in which the radiation is ultraviolet radiation.
10. A backlight substantially as hereinbefore described with reference to the accompanying drawings.
11. A backlit liquid crystal display comprising a backlight as defined in preceding claim and a liquid crystal screen arranged to receive radiation from the backlight.
12. The backlit liquid crystal display of claim 11 in which the screen is displaced from the backlight by a distance whereby radiation originating from adjacent radiation sources is mixed at the liquid crystal screen.