Module 31: Projection Displays
Module 31: Projection Displays
31.1. Introduction
Flat panel displays become larger every year, but until large flat panel displays decrease significantly in cost, projection systems will be marketable. In a sense CRTs are projections systems, but for large areas, the evacuated glass envelope becomes too large and heavy. Projection systems using some sort of spatial light modulation (SLM) have an advantage since the optical path can be tightly folded in a compact form to save space.
Projection displays generally use a lens to focus the magnified image of a small internal display, thereby projecting the image onto a screen. Front projection systems typically use a long throw distance (the room) to project onto a wall whereas rear projection systems use screens with gain to redirect the light more efficiently.
Another distinction in projection systems revolves around how the image is created. Light valve projection systems use a fixed light source and an independent modulation element to form the image. They can have high brightness and contrast because the light source and modulation can be independently optimized. CRT projection can have reasonable brightness and efficiency using conventional CRT technology and optically merging the three monochrome CRT images that generate red, green and blue.
31.2. Types of Projection Configuration
The primary distinction between projection systems is the configuration: front or rear projection. The viewer and the projection unit are on the same side of the front projection screen, while the rear projection system is hidden behind the screen. In most applications, front projection is not used in homes since it requires a long throw length, and diffuse reflections from ambient light degrade the contrast significantly.
Rear projection systems consume valuable space behind the screen, even though the optics can be packaged up compactly, it does use space. A simple configuration of a rear projection screen is shown below:
31.3. Rear Projection - Contrast and Gain
Independent of projection technology, the light source, image generator, and screen are all critical components in optimizing performance. Comparatively speaking, screen performance is the main advantage of the rear projection over the front projection screen technology.
Consider the following screen example. The diverging beam of projected light is directed into the viewing angle of a Fresnel lens as shown below. The image is formed on a second sheet which has particles dispersed in it to scatter light over a certain angular range. Focusing and defocusing the light results in passive gain, which means that the image will be brighter within the intended viewing angle than it would appear on a Lambertian screen.
The second element also has lenticular elements molded into the lens to expand the horizontal range. The screen would not be very satisfactory if the horizontal viewing range were as narrow as the vertical.
These elements also serve to increase light entering the back of the lenticular sheet becomes concentrated and emerges at the front between black stripes, located here. These black stripes absorb the light that falls on the screen between lenticular elements, but the projected light passes through the screen with less than ~10% loss. A larger fraction of the ambient light falling on the screen from the room is interpreted and absorbed by the black stripes - this improves contrast.
31.4. Wavelength and Polarization Separation
Many projection displays effectively manage light more efficiently than direct view displays. As an example, shadow mask CRTs and LCDs that utilize spatial color synthesis techniques to create full color image are less efficient than monochrome displays. In a color CRT only about one-fifth of the electrons pass through the shadow mask and become available for visible light creation. In LCDs the color filter arrays can reduce transmission by 3-5 times. In projection devices light can be managed much more effectively. Dichroic mirrors are often used to separate light into its primary components, as well as to recombine the light after it has been processed. Polarizers that are used on direct view displays throw away the light through absorption, but clever optical design can recycle unwanted polarized light and convert it into a polarization that can be reused. Before we go into the various projection technology, it is worthwhile to cover some standard tricks of the trade.
31.5. Color Separation and Projection Lenses
In full color, light valve systems, reasonable color separation efficiency is usuallyachieved with dichroic mirrors, which reflect a well-defined bandwidth from a portion of the spectrum and transmit the rest. Dichroic mirrors can perform two functions in projection display technology: such as splitting out the colors needed from a broadband light source, or recombining colors from separate spatial light modulators or individual light sources.
Dichroic mirrors are known to the optics community as interference filters, which consist of alternating layers of two materials with different indices of refraction. The reflection at any given boundary is small but when added up in phase, they are very efficient. The angle of incidence with the dichroic mirror is critical, because of the change in pathlength associated with higher incidence - therefore the angle at which the mirror is oriented in the optical system is a design parameter.
The size of the projection is often critical especially in today's market where many projectors can be carried during travel. One way to create a folded optical design is the Scholl method, as shown below:
The mirror configuration is the same as of a pentaprism, but the prism has been eliminated. The nominal angle of incidence for all beams in the system is 22.5°. Dichroic 1 reflects red and passes the rest of the spectrum, while dichroic 2 reflects blue. In this particular system dichroic 2 is only required to transmit green light, but could also be used to transmit red.
Some CRT projections combine three color images in a cube in which two dichroics intersect one another diagonally, resulting in an angle of incidence of 45°. In this configuration, the image can be transferred to the screen with only one projection lens instead of three.
This is advantageous for front projection systems, where the distance from the projector to the screen and screen size may vary. For rear projection systems, the distance to the projection screen is fixed, and it is simple to project images from a CRT to the screen through three separate lenses. In this configuration the three lenses tend to be large because shrinking the CRT would decrease light output and resolution.
Projection lenses collect surprisingly little of the light emitted from a CRT. For example, a typical lens with a 1/f aperture collects light from a cone with a half-angle of only 27°. Only 20% of the Lambertian distribution passes through this cone. This inferior collection factor decreases the efficiency advantage of using three independent monochrome tubes. This is one reason why projection systems are only typically used only for generating images that are too large for direct view displays.
The problem of collecting light in a more effective manner has been addressed by incorporating an interference filter in the CRT face plate. In this application the angular dependence of the mirror can increase the light output from a CRT. The figure below shows the light emerging from the phosphor layer and encountering an interference filter before entering the face plate.
The filter cuts over from transmitting to reflecting in the 30°-40° range thus narrowing the emission angle light that would accur at layer angles and not be collected by the projection lens.
Light that is emitted at larger angles will be ejected by the filter and not collected by the projection lens. Since the phosphor layer is diffusely reflecting, a portion of the reflected light can be scattered back into the collection optics where the filter transmits. Therefore, more light is collected by the projection lens, and unwanted wavelengths emitted by the phosphor layer can be rejected by the filter. A gain in luminance can be achieved, however, interference filters are costly.
31.6. Polarization Separation
Most liquid crystal technology requires polarized light input. Practically speaking, more than 50% of the light is absorbed by the first polarizer, and the second polarizer can also absorb some of the selected components. Because of this limitation, various projection designs that utilize LCD technology try to manage polarized light.
The figure below shows a compact projection design in which the three color images are combined by crossed dichroic mirrors.
Crossed dichroics have been used transmissively to combine color channels in both CRT and LCD based systems, but in this configuration also work in reflection. As a result the same dichroic beam splitter can be used to both separate and combine the colors. Half of the system aperture is filled with the polarization that passes directly through the polarizing beam splitter. The other polarization is reflected internally and becomes rotated by 90° after twice passing through a quarter wave plate. The reflected light emerges below the original beam and fills the remainder of the system aperture.
The electrooptic spatial light modulators in the system act as switchable quarter wave plates. Light that has its polarization rotated by one of the modulators subsequently reflects from the internal diagonal in the polarizing beam splitter and passes out through the projection lens. The rest of the light passes through the beam splitter and returns to the source. Light that is returned to the source may be reflected and added to the output. A major challenge is to conceal the bonding between the two values of the aperture.
A modifiedsystem shown below does not double the aspect ratio of the source but alters the capability to recycle the unwanted polarization.
In this configuration the mirror reflects the S-wavelength, which is polarized perpendicular to the plane containing the incident and reflected propagation directions, back into a polarizing beam splitter. P wave is defined as polarized in the plane of incidence, which passes through the beam splitter and can be used by the system. The S- wave is then reflected by the beam splitter back to the source. A quarter wave plate is placed between the lamp and the polarizing beam splitter. The effect on the polarization of the reflected light is the same as in a circular polarizer used as a contrast enhancement filter.
The principle axes of the quarter wave plate are at 45° with the respect to the polarization axes of the S and P waves. Assuming that the light undergoes specula reflection inside the lamp, two passes through the quarter-wave plate rotate the plane of polarization by 90°. The recycled light in this case becomes P-wave. A problem with this system is that the intensity and thespectrum can be influenced by multiple reflections.
Another possible configuration is shown below.
In this case the unpolarized light will traverse through a polarizing beam splitter. The S and P polarizations emerge from the beam splitters and are reflected by mirrors towards the liquid crystal spatial light modulation from opposite directions. After passing through the modulator, the two beams retrace one another's paths back to the beam splitter. At individual pixel positions the modulator rotates the polarization of the two beams by 90°. In this case the S changes into the P wave and the P changes into the S wave, and the two beams will retrace each other's paths into the light source. In the pixel position where the two beams pass through the modulator un unaffected, however both will emerge from the forth face of the beam splitter and enter the projection lens. Both ferroelectrics and twisted nematics can be used to simultaneous modulation of oppositely -propagating, orthogonal polarized light beams. For this particular configuration, the system treats the two polarizations symmetrically, but it is difficult to devise a color version that is as compact as the system described above.
31.7. CRT Based Projectors
Typically projectors based on CRT technology are comprised of three individual CRTs with dedicated projection. Most CRT projection systems are designed for rear projection applications.
The figure below shows an example of a CRT projection that is designed for high definition television (HDTV).
Typically, the faceplate of the CRT is concave. Negative curvature on the inside of the faceplate improves the image quality in two ways:
- The curvature tends to flatten the field of the projection image (which is more difficult to accomplish with extra lens elements as the projection distance increases).
- The curvature also makes the projected image brighter at the edges. The angular distribution of light emitted by the phosphor face plate at any position is centered about the direction normal to the face plate. The normal is tilted into the lens at the edges, so light emitted there is collected if the screen were flat.
The coolant between the first projection lens and the face plate improves the longevity of the phosphor and tube in three ways:
- It conducts heat away from the face plate.
- It reduces temperature gradients, minimizing tube cracking.
- It is index matched with the faceplate to suppress unwanted reflections, thereby improving contrast and brightness
To optimize brightness, large optical elements in CRT based projectors are utilized. The faceplates are usually 5-7 inches (diagonal) to maximize efficiency. The efficiency of phosphor depends on the excitation energy pulse phosphors saturate at high input power because all the activation centers become excited. The light output depends on the density and decay time of activation centers, and after saturation, increasing the input power does not increase the light output. Making smaller CRTs with higher resolution would require less phosphor at each pixel and less light output at saturation. The figure below shows the efficacy of some common phosphor materials.