Chapter 3 Design and simulation of MPF
3.5. Simulation results
3.5.2 Average luminance
Fig. 3-8 Lamp-Mura contrast of MPF and MPF-Plate case.
3.5.2 Average luminance
The average luminance between two lamps was also obtained, as shown in Fig. 3-9. The luminance of MPF case decreased when the MPF curvature increased over 70 mm-1. This situation was similar to MPF-Plate case with the threshold MPF curvature of 67 mm-1. Additionally, the luminance of MPF-Plate case in curvature range, 40~75 mm-1, is lower than that of MPF case. At optimized curvature 60 mm-1, MPF-Plate case yielded 2% decrease in luminance (7243 nits Æ 7070 nits) compared with MPF case. However, most important is that both cases had higher luminance than the conventional backlight system.
3.6 Summary
The concave prism profile of MPF was designed as parabola, and the curvature is the designed parameter. For optimizing the MPF curvature, backlight component models were built to create a simulation environment by LightTools. Additionally, MPF and MPF-Plate case with MPFs were designed to compare with the conventional backlight system. In the simulation, lamp-Mura contrast results were obtained. Then, the optimized curvature of MPF and MPF-Plate case, which had less lamp-Mura contrast as the conventional backlight, was found about 60 mm-1. Moreover, a higher luminance was also obtained at curvature 60 mm-1. Thus, we conclude the MPF curvature of 60 mm-1 has best performance to suppress lamp-Mura. We then fabricated MPF by screen printing process.
Chapter 4
Fabrication technologies and measurement instruments
Optical devices with micro-structures, such as prisms, micro-lens and diffractive optical elements, are generally used in micro-electro-mechanical systems (MEMS), display devices and projectors, et al. For devices with feature size of 1μm ~ 1mm, photolithography and etching techniques (standard semiconductor processes), diamond machining and excimer laser micro-machining (laser ablation) are applicable[20]. These methods are complex and are not efficient in fabricating the proposed MPF concave-prism structures. Therefore, an easy and economical fabrication method, screen printing process, supported by E vendor was adopted to fabricate MPF. Moreover, several instruments for measuring MPF profiles and optical performances of designed backlight systems are also introduced in this chapter.
4.1 Screen printing process of MPFs
The concept to fabricate the MPF concave-prism profile is to fill a UV-curable material on a conventional BEF, where this material has the same refractive index to the prism structure. The MPF fabrication schematic by using screen printing process is shown in Fig.
4-1. The dimension of the used BEF is 25μm height and 50μm width as shown in Fig. 4-2.
For controlling the amount of material filled, first, a mask (so-called screen) with a designed aperture ratio distribution was placed on a conventional BEF. Second, the UV-curable
material was dropped on the mask. Third, a squeegee, which is a flexible rubber, pressed and horizontally moved over the mask. Then, the UV-curable material flowed through mask apertures to fill the prism grooves on the BEF. Finally, a filled concave profile was formed after curing the filled material in UV-exposure.
According to the designed aperture ratio of the mask, the amount of filled material could be controlled to make different concave-prism profiles of MPFs. In this thesis, the term
“filling ratio (%)” depended on the mask aperture ratio was applied to classify different filling profiles of MPFs.
Fig. 4-1 Schematic of screen printing process.
Fig. 4-2 Schematic of used BEF structure.
4.2 Measurement instruments
After the fabrication, fabricated MPF profiles were examined and measured by an optical microscope (OM) and the surface profiler, Alpha-Step IQ[21], respectively. Then, optical performances of backlight systems were measured using a conoscopic system and a
charge-coupled device (CCD). In order to set up the diffusive plate and diffuser models in our simulation, our designed bidirectional transmittance distribution function (BTDF) measurement device was used to extract scatter properties of diffusive films to import into the simulation software, LightToolsTM.
4.2.1 Conoscopic system
The conoscopic system applies Fourier transform lens to transfer light beams emitted from the test area to a CCD array. Therefore, the angular properties could be easily measured on the CCD sensor plane. The CCD array consists of various directional CCD sensors to detect brightness, color, and angular distribution of transmissive light. Besides, it has two modes for measuring transmissive or reflective light. In our experiment, the transmissive mode is employed because the backlight system can emit light itself.
The options for testing under illumination are based on the combination of Fourier Optics and a cooled CCD sensor head. As shown in Fig. 4-3, the measurement is used in transmissive mode, where the first lens provides a Fourier transform image of the display surface. Each light beam emitted from the test area at incident angle, θ, could be focused on the focal plane at the same azimuth and at a position x=F(θ). The angular characteristics of the sample are thus measured simply and quickly, without any mechanical movement.
4.2.2 BTDF measurement device
In order to measure BTDFs of diffusive films, a light source device which can provide collimating and different-incident-angle light is required. Accordingly, we designed a device for this purpose, as shown in Fig. 4-4. Nine light-emitting diodes (LEDs) sited on the arc are chosen as the light sources. The LED is green light with a peak wavelength about 520 nm.
Each LED represents each incident direction, from 0 to 80, step by every 10 degrees. Besides, the small aperture on the exit plane of LEDs and the small hole on the top plate ensure the exit light is collimated.
Fig. 4-4 Schematic of our designed device for BTDF measurement.
When measuring BTDFs of diffusers, the BTDF measurement device is placed under the conoscopic system to serve as the light source. The luminance information in each direction can be measured by the conoscopic system. A photograph of the experimental setting is shown in Fig. 4-5.
Fig. 4-5 Diagram of BTDF measurement.
4.3 Summary
Before the MPF fabrication, simulation software LightTools built models of backlight components to do simulations to find optimized design results. The BTDF measurement device measured the BTDF value of optical films to import into LightTools to build corresponding models. The optical property of simulation models is similar to real objects.
The conoscopic system not only can measure the BRDF of the reflector but also the optical performances of backlight systems, such as angular luminance distribution, color information, et al.
To fabricate the concave-parabolic prism structure of MPFs, the idea in this thesis was to fill some material in the prism groove of conventional BEFs to be the concave-parabolic profile. Screen printing process was adopted to accomplish this idea. By adjusting the aperture ratio of the mask (screen), the amount of filled UV-curable material in the prism groove was controlled; on the other hand, different prism profiles of MPFs could be realized. Compared with the imprint method, the cost of a mask in the screen printing process is lower than that of a mold. Thus, we conclude that screen printing process has an advantage of high design flexibility to find the optimized MPF in simple process and low cost. Moreover, fabricated optimum MPF is possible to transfer to a mold for roll-to-roll mass production.
Chapter 5
Experimental results and discussions
After fabricating MPFs by screen printing process, surface profiles of fabricated MPFs were measured by a surface profiler. Then, the relationship between filling ratios and fitting curvatures of MPFs was also obtained. The experimental results, which included brightness, viewing angle, output light efficiency and real luminance images, of three experiment backlight structures (conventional, MPF and MPF-Plate case) will be presented and discussed.
5.1 Fabrication results
The optimized MPF curvature among fabricated MPFs can be obtained by measuring the surface profiles. The surface profiles between two prisms width were measured by Alpha-Step IQ[20], as shown in Fig. 5-1. The result showed that the higher filling ratio, the flatter surface profile. In order to verify whether the fabrication result meets the designed MPF curvature, measured MPF profiles were fitted with a parabolic curve to find an approximate curvature.
For example, fitted results of filling ratio 10% and 25% MPFs are shown in Fig. 5-2. Then, the relationship between filling ratio and fitting curvature of MPFs were obtained, as shown in Fig. 5-3. Among the results, the approximate curvature of filling ratio 10% MPF of 58 mm-1 was the closest to the optimized curvature 60 mm-1 of MPF and MPF-Plate case. Thus, we conclude that filling ratio 10% MPF is the optimized MPF which can suppress lamp-Mura as the conventional backlight system. In following experiments, filling ratio 10% MPF will be
adopted to do experiments for verifying our conclusion.
Fig. 5-1 Measured MPF surface profiles.
(a) (b) Fig. 5-2 Fitting profiles of filling ratio (a) 10% and (b) 25% MPFs.
Fig. 5-3 Relationship between filling ratio and curvature.
5.2 Experimental results
Filling ratio 10% MPF was applied to MPF and MPF-Plate case to do experiments. Then, optical performances, such as lamp-Mura contrast, normal brightness, viewing angle and output light efficiency, of three experiment backlight structures (conventional, MPF and MPF-Plate case) were measured and compared.
5.2.1 Lamp-Mura contrast and average luminance of MPF case
To verify that the optimized MPF curvature had best performance of suppressing lamp-Mura, lamp-Mura contrasts of MPF case with fabricated MPFs were obtained to compare with simulation results. For the conventional backlight and MPF case, 9 points luminance between two lamps at the center of the 32” backlight system was measured by the conoscopic system, as shown in Fig. 5-4. Then, lamp-Mura contrasts were calculated according to the measured luminance distribution. The simulation and experimental results of lamp-Mura contrast and average luminance are shown in Fig. 5-5.
The lamp-Mura contrasts in the experimental and simulation results had similar trend. In
the experimental result, the filling ratio 10% MPF yielded minimum lamp-Mura contrast, 0.010, lower than the conventional backlight (0.012), which agreed with the simulation result.
On the other hand, MPF case with filling ratio 10% MPF resulted in invisible lamp-Mura as the conventional backlight. Moreover, for the MPF-Plate case with the filling ratio 10% MPF, a lamp-Mura contrast, 0.018, was also obtained by experiments. However, this lamp-Mura contrast higher than the conventional does not mean that lamp-Mura is appeared. The following experimental will verify whether MPF-Plate can suppress lamp-Mura.
Moreover, the average luminance of MPF case was lower than the conventional backlight. Compared with simulation results, the average luminance of MPF case was higher than that of the conventional backlight when the curvature was below 70. The mismatch result of the average luminance will be discussed at the end of this chapter.
Fig. 5-4 Measured luminance distributions of conventional backlight and MPF case.
(a) (b)
Fig. 5-5 Comparisons of simulation and experimental results in the MPF case, (a) lamp-Mura contrast and (b) relative average luminance.
5.2.2 Normal brightness and viewing angle
The experimental result of the angular luminance cross-section was measured, as shown in Fig. 5-6. Calculated values of luminance angle and of relative normal brightness are shown in Table 5-1. For MPF case, the FWHM was improved to 90°, and the normal luminance can be maintained as high as the conventional backlight. For MPF-plate case, although the normal luminance was reduced to 90% of that of the conventional backlight, the FWHM was improved further to 116°. Compared the MPF case with the MPF-Plate case, MPF case with high normal brightness is adaptable to a high brightness backlight module; MPF-Plate case with wider FWHM is suitable for a wide viewing angle LCD.
(a) (b)
Fig. 5-6 Diagrams of luminance cross-section in (a) horizontal and (b) vertical.
Table 5-1 Relative normal brightness and luminance viewing angle.
5.2.3 Relative output light efficiency
From the result of the angular luminance contour as shown in Fig. 5-7, the luminance viewing angles of MPF and MPF-Plate case were both wider than the conventional backlight.
Consequently, for MPF and MPF-Plate case, an 8% and a 19% improvement in output light efficiency were obtained, respectively, by integrating the flux of whole viewing angles, as shown in Table 5-2. Compared with the conventional backlight and MPF case, MPF-Plate case had best performance of increasing output light efficiency. This result also verified the design motivation of MPF-Plate case.
(a) (b) (c)
Fig. 5-7 Diagrams of angular luminance contour, (a) the conventional backlight, (b) MPF case, and (c) MPF-Plate case.
Table 5-2 Relative output light efficiency.
5.2.4 Real luminance images
A charge-coupled device (CCD) camera (PM-1600F Color Series Imaging Photometers and Colorimeters, Radiant Imaging, Inc.[22]) was used to record real images and contours of the output luminance distribution from backlight systems. The recorded real images are close to what the human eye sees. The captured results of three experiment cases are shown in Fig.
5-8. In the experiment, we also recorded the real image of MPF case with a conventional BEF to be a contrast with three original cases, as shown in Fig. 5-8 (b). This image had obvious lamp-Mura, and a periodic pattern was also appeared in the contour. In contrast, both MPF and MPF-Plate case had no periodic luminance pattern, therefore, they resulted in an invisible lamp-Mura image as the conventional backlight, as shown in Figs. 5-8 (a),(c), and (d). This result showed that the lamp-Mura contrast of 0.018 in the MPF-Plate case also can yield
invisible lamp-Mura. In summary, the real luminance images verified that MPF and MPF-Plate case can suppress the lamp-Mura effectively.
Fig. 5-8 Captured results by a CCD camera, the captured size on the backlight is 20cm by 20cm.
5.3 Summary and discussion
The optimized curvature of MPF was successfully fabricated and verified that it can suppress lamp-Mura effectively. Among fabricated MPFs of different filling ratios, the MPF filling ratio 10% of curvature 58 mm-1 was the closest to the optimized curvature 60 mm-1 of MPF and MPF-Plate case. Thus, MPF filling ratio 10% was chosen to be the optimized MPF which can suppress lamp-Mura. In MPF case, that the lamp-Mura contrast (0.010) was lower than that of the conventional backlight (0.012), agrees with the simulation result. In MPF-Plate case, the lamp-Mura contrast was 0.018, higher than 0.012. However, both MPF and MPF-Plate case resulted in an invisible lamp-Mura luminance image as the conventional backlight. Thus, we conclude the lamp-Mura contrast of 0.018 can yield invisible lamp-Mura.
Moreover, the luminance difference between simulation and experimental results of MPF case increased when MPF curvature decreased, as shown in Fig. 5-5 (b). From the optical microscope (OM) image of filling ratio 100% MPF as shown in Fig. 5-9, many particles were in the filled material. Therefore, we assume the filled UV-curable material causes the light scattering.
According to our assumption, we simulated a MPF with scattering filled material. This material was the pure UV-curable material added scattering particles ( 105 particles/mm3). In the MPF case with different curvatures, the luminance distributions with and without scattering material are shown in Fig. 5-10 (a). Additionally, the calculated result of the luminance with scattering material relative to that without is shown in Fig. 5-10 (b). Where the luminance decreased when the curvature decreased or the filling ratio increased. This result verified that the luminance is inversely proportional to the amount of scattering material.
Fig. 5-9 OM pictures of filling ratio 100% MPF.
(a) (b)
Fig. 5-10 Simulation results of (a) luminance distribution of MPF case with and without scattering filled material, (b) relative luminance of MPF case with filled material of scattering compared to that without scattering filled material.
Moreover, MPF and MPF-Plate case enhanced FWHM to 90° and 116°, respectively, in luminance viewing angle and had an 8% and a 19% improvement in output light efficiency, respectively. The higher efficiency in MPF-Plat case was also verified.
The normal luminance of MPF-Plate case is lower than that of MPF case as shown in Fig.
5-6 and Table 5-1. One possible reason is that the number of interfaces is reduced in the MPF-Plate. Because the light guiding effect exists in a pair of interfaces, the reduced number of interfaces will cause less light guiding effect and less light transferring to normal direction.
Thus, MPF-Plate case had lower normal luminance than MPF case did.
Chapter 6
Patterned-MPF
A slim LCD-TV is a trend in current display market because it is suitable to mount on a wall to free up more space. However, the thickness of LCD-TV depends on the backlight structure. To achieve a slim LCD-TV, backlight thickness has to be reduced. When the backlight thickness decreases, the lamp-Mura will become more serious. Additionally, the MPF had a limitation to suppress lamp-Mura in a slim backlight system. Therefore, compared to the MPF with uniform prism structure, a Patterned-MPF was proposed to improve the uniformity in a slim backlight system.
The Patterned-MPF has various curvatures depending on the position from lamps, as shown in Fig. 6-1. This pattern concept has been applied to diffusive films for enhancing the uniformity of side-emitting backlight system by Y.-C. Lo[23]. Because lamp-Mura is caused by a non-uniform light distribution, thus, the pattern concept can also be applied to MPF to optimize the MPF curvature at different positions. In the following designs and simulations, MPF case will be chosen to design backlight system at a thickness of 20 mm.
Fig. 6-1 Schematic of Patterned-MPF.
6.1 Design and simulation results
To design the curvature distribution of Patterned-MPF, luminance distributions of different MPF curvatures have to be determined first. The luminance simulation results of different thicknesses are shown in Fig. 6-2. Considering the uniformity at each position, we found output luminance is limited at 0.5 pitch. This luminance value was called the target luminance in this thesis.
(a) (b)
(c) (d)
Fig. 6-2 Luminance simulation results of MPF case with different curvatures at the backlight thickness of (a) 10 mm, (b) 12 mm, (c) 14 mm, and (d) 16 mm.
The luminance at each position is designed as the same as the target, which becomes the initial curvature distribution for Patterned-MPFs, as shown in Fig. 6-3. As a result, when the thickness decreased, the variation of curvature distribution increased because of more serious lamp-Mura. In contrast, in a thick backlight system, a uniform MPF can result in a uniform light output.
Fig. 6-3 Initial curvature distribution design of Patterned-MPF at the backlight thickness of 10 mm, 12 mm, 14 mm, and 16 mm.
The Patterned-MPF with the initial curvature distribution was then simulated to evaluate the uniformity of light output. The simulation results of output luminance distribution are shown in Fig. 6-4. The initial designed Patterned-MPFs yielded a not uniform enough luminance distribution. Therefore, the curvature distribution needs to be future modified for a more uniform light output. For example, in Fig. 6-4 (a), the luminance at 0 and 1 pitch were higher than the target luminance (5500 nits). Thus, the corresponding curvatures have to be increased for decreasing luminance. This is because the luminance of 0 or 1 pitch (above the lamp) decreased with increasing curvature, as shown in Fig. 6-2. Accordingly, the uniform light outputs (blue curves) were then yielded, as shown in Fig. 6-4, and the corresponding optimized curvature distributions are shown in Fig. 6-5.
(a) (b)
(c) (d)
Fig. 6-4 Luminance simulation results of MPF case with designed Patterned-MPF at the backlight thickness of (a) 10 mm, (b) 12 mm, (c) 14 mm, and (d) 16 mm.
Fig. 6-5 Optimized curvature distribution design of Patterned-MPF at the backlight thickness
Moreover, lamp-Mura contrasts calculated from the luminance distributions in Fig. 6-4
Moreover, lamp-Mura contrasts calculated from the luminance distributions in Fig. 6-4