WO2000017700A1

WO2000017700A1 - Optical arrangement for flat-panel displays

Info

Publication number: WO2000017700A1
Application number: PCT/GB1999/003131
Authority: WO
Inventors: William Alden Crossland; Timothy Martin Coker
Original assignee: Screen Technology Limited
Priority date: 1998-09-21
Filing date: 1999-09-21
Publication date: 2000-03-30
Also published as: JP2002525677A; KR20010075221A; CN1319196A; EP1116066A1

Abstract

A flat-panel display comprises a backlight for producing narrowband activation light; a means such as a liquid-crystal modulator (21) for modulating the activation light; and a photo-luminous output screen (23) which emits visible light in response to the activation light. Instead of the activation light simply passing through the modulator onto the screen, an optical arrangement (22) is interposed between the modulator (21) and the output panel so as to project an image of the modulator onto the output screen. This reduces the requirement for collimation of the activation light.

Description

OPTICAL ARRANGEMENT FOR FLAT-PANEL DISPLAYS

This invention relates to flat-panel, in particular liquid-crystal, displays using short wavelength activating light and photo-luminescent output elements known as PL-LCDs, as described in O95/27920 (Crossland et al) .

PL-LCD flat-panel displays have particular problems associated primarily with the requirement to collimate the excitation light. Other examples of FPD technology, such as TFT or STN screens, also have a requirement to collimate the light; however, this is generally because contrast, image brightness and, in the case of STN screens, the degree of multiplexability are all enhanced by collimation. Conventional liquid crystal FPDs suffer from the disadvantage that, as the degree of collimation is increased, the viewing angle suffers accordingly. Proposed solutions to this have' included diffusing screens at the front of the display (i.e. immediately between the colour filters and the observer) .

The PL-LCD architecture gets round all the viewing angle problems of conventional FPDs by placing visible phosphors at the front of the screen and exciting them with activation light from a backlight. However, this solution to the viewing angle problem tends to make the collimation requirement even greater for PL-LCDs, in that there is the additional problem of crosstalk between adjacent pixels. A conceptually simple solution to this is to provide a very high degree of collimation, such that light passing through one of the liquid-crystal pixels only impinges on the correct phosphor pixel. However it is difficult to collimate light from an extended source to this degree and it is even more difficult to do so efficiently. The amount of collimation required (or the allowable divergence) depends on the pitch of the pixels and the distance between liquid crystal modulator layer and phosphor screen.

In the case of conventional projection displays, such as cinema projectors, large-screen rear-projection TVs etc, very highly collimated light from a near point source is used. However, this light is not simply passed through the modulator and allowed to fall, still substantially parallel, upon the screen. It is instead gathered, focused and projected onto the screen by a projection lens. If this were not the case the collimation requirement (to avoid pixel crosstalk) would be quite beyond current optics because of the great distance between the modulator and the screen - this is true for all types of projection display, not just the PL-LCD architecture. In the invention the projection approach is used to aid the operation of PL-LCD flat-panel displays.

According to the invention there is provided a flat-panel display comprising an array of modulating elements designed to modulate activating light input from the rear, and a photo-luminescent output screen adapted to receive the modulated and to give a corresponding visible output to produce the display image; characterised by the provision of an optical arrangement projecting the plane of the modulating elements onto that of the output elements.

The advantage of the concept of using an optical arrangement to 'project' an image of the LCD modulator onto the phosphor output elements is that the two can now be separated by a distance that is no longer determined by the geometry of the pixel pattern and the degree of divergence (or collimation) of the activation light; the new separation is determined by the optics. However this is not to say that collimation is no longer relevant, but under this concept it affects image quality in differing ways. Indeed some schemes can operate with un-collimated light, though this is somewhat wasteful as it requires some light to be vignetted or blocked from reaching the phosphor output screen. Overall, whilst collimation is still relevant, its effect no longer bears a simple relation to pixel geometry.

For a given amount of collimation, the pixel geometry without an optical arrangement will determine the maximum separation of the phosphor and the liquid crystal modulator for zero crosstalk. In many cases this maximum separation is less than the thickness of the glass panel, a solution to this is to place the phosphor inside the modulator rather than outside. Whilst this is possible in theory at least, in practice it is difficult and expensive. However the use of a suitable optical element decouples the separation distance from the pixel geometry and can therefore allow the separation to be increased more or less arbitrarily, so that the phosphors can remain outside the modulator. Thus one application of the invention is as an alternative to 'phosphor inside' solutions.

Another advantage of using this 'projection' approach, especially in relation to other conventional architectures that may employ similar principles, is that the PL-LCD architecture is essentially monochromatic in terms of the light that has to be modulated and projected. Thus the optical arrangement need not include the additional complexity that is normally required in optics to counteract wavelength dispersion - for example achromatic doublets with their attendant expense are not required, singlet lenses will suffice. Additionally, a more subtle advantage is that the resolution of the final image is solely determined by the phosphor arrangement on the output screen. For instance if the 'dot pitch' on the output screen is 3/mm then the resolution of the image will be 3 dots/mm. This is so because the pixellated phosphor screen effectively resamples the image produced by the optics in a way that is analogous to digital sampling of audio material . This aspect is most obvious when one considers that the phosphors are most often arranged on the output screen within a black matrix. If the resolution is such that each pixel is imaged onto the output screen in a less than perfect manner, then a 'fuzzy' out-of-focus image of each pixel results. However, and within certain limits, the light within this fuzzy image falls either on the intended pixel or within the black matrix surrounding it, so the correct phosphor pixel is still activated and an accurate image is still obtained. In this way simple and cheap optics can be utilised without jeopardising image quality.

The principle of using an optical arrangement to project an image is identical to that of a projection display except that in this architecture all the components are preferably contained in a single housing or rigid assembly and the distance over which the liquid-crystal pixels are imaged (i.e. projected) is typically of the order of millimetres or centimetres at most, certainly less than one metre. In general in embodiments of the invention the imaging distance is small in comparison to the linear dimensions of the display. However, the small imaging distance or optical 'throw' makes the use of normal lenses with an aperture similar to that of the panel totally impractical. One solution in this case is to use an array of micro- lenses . These have suitable focal lengths, i.e. of the order of a few millimetres, but can be made in array sizes sufficient for any flat-panel application.

Micro- lenses can be used in various ways; one, referred to as integrated imaging, occurs when the pitch of the micro-lenses is much less than the size of the image/object. The basic theory of integrated imaging with micro-lens arrays is described in N F Borelli et al; Imaging and radiometric properties of micro-lens arrays, Applied Optics Vol. 30(25), 1 Sep 91, pp. 3633-3642. In the case of the invention, the object is typically tens of centimetres in size, being an LCD panel, and therefore integrated imaging will take place.

Even with the use of field lenses, as described in this paper, with 'ordinary' amounts of collimation secondary images are created which degrade the image obtained. The most obvious way to avoid this problem is to increase the degree of collimation of the activation light. This aspect is particularly applicable to PL-LCD displays, because the narrow-band excitation light allows the use of dielectric collimation (see the applicant's earlier application No. WO 98/49585) . This is a particularly effective way of collimating extended light sources because, generally, high-angle side lobes, which are often an artifact of refractive collimating films for white light, can be eliminated and regeneration of the non-collimated light is better facilitated. Another solution, as described in GB2329786A (CRL Ltd) , is to use arrays of apertures as well as micro- lenses but this approach is very wasteful of the available illumination from the backlight. Another point to note is that normal projection optics project magnified images (in many cases of course a considerably magnified image) . In many applications of the invention there is no need to magnify the image of the modulator, for example in a desktop computer monitor. In these cases the image of the modulator is projected with no magnification, but this can be alternatively expressed as unity or 1:1 magnification. In turn this is sometimes referred to as image relay or image transfer. Another common method of carrying out relay imaging is with GRIN-lens arrays and these can be simply adapted for use with this invention.

Despite the number of applications for unity magnification, it can be advantageous to project with greater than unity magnification. One such application is where displays are to be tiled seamlessly, in spite of the space required at the side of the modulator itself for the addressing circuitry. This idea was disclosed in GB2236447 (KC Tung) and was also explored in the previously mentioned CRL patent. The concept of tiling sub-displays without visible joins is referred to here as seamless tiling, and an image so produced is referred to as a seamless image.

Prior art seamlessly tiled displays have all applied this magnification principle to conventional, rather than PL-LCD, display architectures. When the principle is applied to a PL-LCD architecture a number of advantages arise. In prior art examples given here, although the image may appear to be seamless, in practice there are viewing angle problems (a large seamless display is somewhat self-defeating if it only appears seamless from certain viewing angles) . The solution described in the CRL patent is either to overlap the images from the individual sub-panels or to form the image on a diffusing screen (with the result that the viewing angle characteristics are determined by the screen, not the optics that create the image) . However, none of these methods are fully practical - the solutions described in GB2329786A in particular are very inefficient - nor do they really produce an image with a full Lambertian viewing angle characteristic over its entire area. Embodiments of the invention may, therefore, provide a display assembly comprising a tiled matrix of display units, each in turn comprising an array of modulating elements designed to modulate light input from the rear, and circuitry for addressing the modulating elements; characterised by the provision of an optical arrangement projecting a magnified image of the plane of each of the modulating elements onto a photo- luminous output screen, in such a way that a seamless image is produced.

The geometric requirement for projection with magnification in order to achieve seamless tiling is the same as for unity magnification applications, that is, that the optical throw is very short. To this end integrated imaging micro-lenses can also be used in order to magnify the image; in this case the pitches of the arrays are not constant, but rather they increase as one moves away from the modulators. The basic principle applied here is described in "Close-up imaging of documents and displays with lens arrays", R H Anderson, Applied Optics Vol. 18(4) , 15 Feb 79, pp.477-484. Micro-lens arrays designed according to this principle are sometimes referred to as Super-Gabor lenses. GRIN lenses can also be used for this purpose, in which case the pitch of the GRIN lenses within the array has similarly to be increased from one end to the other.

For a better understanding of the invention embodiments of it will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 shows how the collimation angle is set by the pixel geometry in the absence of an optical arrangement according to the invention; Figure 2 shows diagrammatically how an image is projected onto an output screen carrying phosphors; Figure 3 shows one possible optical arrangement;

Figure 4 shows how the invention can be used to tile displays together seamlessly;

Figure 5 is a ray trace of a micro- lens component as described in Figure 3 ;

Figure 6 shows how secondary images arise where the activation light is not 'tightly' collimated; and

Figure 7 and Figure 8 show the modelled values of pixel crosstalk for a number of cases, demonstrating how the invention is used to reduce crosstalk.

Figure 1 shows how the collimation angle is determined by the geometry of the panel. In order for there to be no inter-pixel crosstalk the collimation of the backlight needs to be less than or equal to the collimation angle φ. This angle is given by the equation:

φ = atan(d/D)

Conversely of course for a given degree of collimation of the activation light and inter-pixel gap, the maximum separation of LCD and phosphor is determined. Figure 2 shows an LCD panel 21, an optical arrangement 22 and an output screen 23 carrying phosphors. Activation light from a backlight (not shown) is modulated by the panel 21, and projected by the optical arrangement 22 onto the phosphor output panel 23. In this figure the separation from LCD to phosphor is determined not simply by the degree of collimation of the backlight, but by the function of the optical arrangement .

Figure 3 shows a possible embodiment of an optical arrangement according to the invention. What is shown is two substrates each with a micro-lens array on both sides. Activation light from the backlight 31 is modulated by the LCD panel (not shown) and is then refracted by the micro-lens arrays in such a way that an image of the LCD panel is formed on the phosphor output screen 32. The lenses in the first array 33 create an inverted image of the panel which is then re-inverted (i.e. erected) by the last array 34. The two central arrays 35 act as field lenses for the first and last arrays, increasing their apparent field of view. The second aspect of the invention, namely that of seamless tiling, is shown diagrammatically in Figure 4. Two sub-displays 41a and 41b are shown together with two optical arrangements 42a and 42b. These components 42a and 42b act to produce a magnified image of each sub-display on the seamless output panel 43. These two images are aligned to create a seamless image, whilst still allowing room around the modulating areas of the sub-displays for addressing circuitry and a mechanical arrangement for holding the sub-displays together in a regular array or matrix.

Figure 5 is a ray trace diagram of a micro-lens component such as was shown in Figure 3. Two field points 51 are shown being imaged by the micro-lens component 52 onto the image plane 53. As can be seen, the two image points 54 are made up of rays that pass through a number of separate lenslet pairs; hence the term 'integrated imaging' . The throw is about 12 mm.

Figure 6 shows a similar ray trace diagram to Figure 5. In this diagram a central or main image point 61 is shown and two secondary image points 62a and 62b are also present. These are caused by the higher ray angles emanating from the object point 63. It can thus be seen that 'tighter' collimation would reduce or eliminate the secondary images. Figure 7 shows the modelled pixel crosstalk for two situations. The first is that where there is no optical arrangement, the activation light passing straight through the modulator to fall on the phosphor output screen. In this case the LCD-to-phosphor separation is 0.8mm, which is typical for a PL-LCD display without the optical arrangement, being essentially the thickness of the top glass plate of the LCD. The second case is where a suitable optical arrangement (microlens array) is introduced. In this case it is seen that the pixel crosstalk with the arrangement 71 is always less than the crosstalk without the arrangement 72, for collimation angles of less than 25 degrees at least.

Figure 8 is a similar diagram to Figure 7, except that a different optical arrangement has been used. In this case the crosstalk with the optical arrangement 81 is only less than the crosstalk without the optical arrangement 82 for collimation angles of less 13 degrees .

The embodiments so far described only mention magnifying optics in relation to tiled displays, but it should be understood that magnification can be used for non-tiled applications. Additionally the activation light that has been referred to is preferably short- wavelength light, that is blue or UV light, most preferably with a central wavelength of 388nm and a bandwidth of 15nm.

The photo- luminescent output screen may include a single, large output element in the form of a continuous layer; this would be suitable for a monochromatic display. However, preferred embodiments would include a plurality of output elements or pixels arranged in colour triads in the known manner for PL-LCD displays .

Claims

1. A flat-panel display comprising:

a means, such as a backlight, for producing narrowband activation light; a means (21) for modulating the activation light; and a photo-luminous output screen (23) which emits visible light in response to the activation light;

characterised in that an optical arrangement (22) is interposed between the modulator (21) and the output panel, the optical arrangement being adapted to project an image of the modulator onto the output screen.

2. A display according to claim 1, in which the optical arrangement is adapted to project an image of the modulating means with unity magnification.

3. A display according to claim 1, in which the optical arrangement is adapted to project a magnified image of the modulating means .

4. A display according to claim 1, in which the modulating means comprises a plurality of separate modulators (41) arranged or tiled in a regular array or matrix and in which the optical arrangement is adapted to optically enlarge the sub-displays so as to form a seamless image on the output screen, thereby forming a seamless visible image.

5. A display according to any preceding claim, in which the optical arrangement is one or more micro- lens arrays or a set of Super-Gabor lenses.

6. A display according to any of claims 1 to 4 in which the optical arrangement comprises one or more GRIN- lens arrays.

7. A display according to any preceding claim, in which the activation light consists of narrowband UV light.

8. A display according to claim 7, in which the activation light has a central wavelength of 388nm and a bandwidth of approximately 15nm.

9. A display according to any of claims 1 to 7, in which the activation light is visible blue light.

10. A display according to any preceding claim, in which the output screen includes a plurality of photo-luminous elements arranged so as to form pixels.