Surface & Volume scattering light pipes

It is well known that the polished light pipe propagates the light by TIR without any optical losses. However, any regional defects of the surface will cause the optical scattering. Actually, the light is partially coupled out from the pipe by the surface roughness which can be generated on purpose in specific location by manufacturing process as shown in Fig. 2-4(a). The main drawback of the outcoupling by surface

As shown in Fig 2-4(b), when the light guided in the light pipe meets the volume scattering section, part of light will not respect the TIR anymore and then they couple out from the light pipe. This volume scattering results from the refraction of the light generated by particles which are poured into the pipe by fabrication process. The illumination should be theoretically controlled by modulation of doping the density of the particles. However, the technical limitation of the fabrication will not allow an exact control of doping distribution. Therefore, the uniformity of the illumination will be difficult to obtain in the volume scattering light pipe.

(a) (b)

Fig. 2-4 (a) Surface scattering light pipes (b) Volume scattering light pipes 2.2.2 Variable cross-section light pipes

The variation of the pipe cross-section can be implemented and classified by continuous or discontinuous cross-section light pipe. The discontinuous cross-section type is illustrated as Fig. 2-5(a). The flux coupled out from the pipe is directly size of the respective cross-sections. The partition of extracted flux is determined as

i propagation angle α of the guided ray is changed at each intersection with the inclined angle. When the propagation angle α becomes smaller than critical angle, the ray will not respect the TIR condition anymore, then the ray is extracted from light pipe.

If the inclined angle of the pipe surface and the incident angle are defined as and

1

i , respectively. Then the new propagation angle_iis defined as





_i  _i1 2 (2-4)

(a) (b)

Fig. 2-5 (a) Discontinuous cross-section light pipes; (b) continuous cross-section light pipes

2.2.3 Refractive & Reflective Outcouplers

The principle of the refractive outcouples consists in placing discontinuities along the pipe surface in order to locally exceed the TIR condition. In such situation, the light escapes from pipe by refraction as shown in Fig. 2-6(a). Where α is the incident angle and ρ is the inclination of refractive surface. The indices of the refraction of the PMMA and air are defined by n_pmma and n_air. Accordingly, the exit angle β can be

The relationship between exit angle, incident angle, and incline angle is shown in Fig.2-6 (b). The gray region covers the incident angle which relates to the light guided by TIR in the pipe. Furthermore, it is obvious that the light extracted from the pipe appearing large incline angle distribution. It is not practical for general light pipe which requires to illumination normal to the surface. Therefore, this kind of approach usually demands a correction plate to collimate the direction of light.

(a)

(b)

Fig. 2-6 (a) Refractive type outcouplers; (b) exit angles achieved by refraction for different orientations of the refractive surface.

The same to the refractive approach, the reflective outcouplers use discontinuities

along the surface of light pipe which deflects the guided ray out from the pipe. Two kinds of the reflective micro prism can be selected: coated micro-prisms and the TIR micro-prism. As illustrated in Fig. 2-7 (a), the reflective coated micro-prism is placed along the pipe surface to deflect part of guided light. The incident angle and the prism orientation are defined by α and ρ, respectively. Therefore, the exit angle can be

Fig. 2-7 (a) Outcoupling produced by a reflective micro-prism (b) Exit angles achieved by reflection for different orientations of the reflective micro-prism.

orientations of the micro prism. Different with the refractive case which has grazing angle of the extracted light, the reflective coated micro-prisms approach is more flexible. Therefore, for given any incident angle, there exists a better corresponding orientation to get the desired outcoupling distribution. However, the metallic coating on working face of prism is difficult to be done and the metallic absorption affects the optical efficiency of outcoupling rays. The cumulative optical losses will significantly deteriorate the optical efficiency of the light pipe.

The TIR micro prisms have an advantage that will not require any coating.

Moreover, TIR will not cause any optical losses. However, the TIR only occurs when rays reaching the prism which incident angle is greater than critical angle as

c obviously that the shadowing effects increase for grazing incident angle (α > 90). A way to repress the shadowing is to design variable cross-section light pipe.

Fig. 2-8 The influence of shadowing effect for a given angle of incident

As shown in Fig.2-9, it is a simulation model of the light propagating in the light

pipe by TIR. The cross-section of the dimension is defined by Y. If each point onz plane can be considered as a point source which emits in ₀ + direction, the z angular spectrum of the light is defined by a cone whose half angle is φ. Therefore, the half angular spectrum after propagation toz plane can be demonstrated equal to φ. ₁ Furthermore, if the reflective micro prisms are located between z and₀ z , the ₁ influence of shadowing can be analyzed. As shown in Fig.2-10, the depth of the three micro prisms P is placed in three random positions. As expect, the angular _y spectrum and illuminance in planez depends on the depth of prisms. Further, there is ₁ a region which does not receive any light (complete shadow).

Fig. 2-9 Propagation between two sections of light pipe.

Fig. 2-10 The influence of the shadowing on propagation of the flux and angular spectrum

In intuitive solution to reduce the shadowing is to decrease the depth of the prisms as shown in Fig.2-11. The prisms are located in the same position, but the depth has been reduced to 60%. As expected, the zone of the complete shadow has been considerably reduced. Nevertheless, some thing should be noticed that the efficiency of the light pipe h is proportional to the prism density G along the light pipe surface which is

Fig. 2-11 Reduction of the shadowing effect by reduced prism size

Where v_p(z) is the local spatial frequency of the prisms, andP is the prism size. _z Thus, the reduction of the prism size does not only reduce the shadowing effect, but also the efficiency of the light pipe. The way to diminish the shadowing, while keeping the efficiency, is to curtail the prism size and proportionally increase the prism spatial frequency. Actually, the general solution between the depth of prism P _z and the size of the prism cross-section Y is according to

20 Py

Y (2-9)

2.3 Summary

Consideration of the particular necessity of FSC scanning BLM, the profile of the light guide chooses the continuous variation of the cross-section type. Further, comparing with different functional light pipes, the reflective outcoupler provides more flexible parameters and better choice for designers. By properly design of the inclined angle of the micro-prism, the craved outcoupled angle can be obtained. In the following chapter, we will use the continuous cross-section light guide combining with reflective micro-prism to accomplish the backlight system, and the simulation software will be utilized to further design and optimize the micro structures.

Chapter 3

Design and Simulation

3.1 Introduction

In this chapter, the main stress will fall on the tandem wedge shaped type light guide components. These components are assembled into desired penal size, and utilized in the backlight module. The BLM is proposed for a scanning field-sequential-color liquid-crystal display (FSC LCD). Therefore, the requirement and specification of the backlight system will be briefly introduced. Moreover, the design of the wedge shaped light guide combining with prismatic micro-bump structures will be described in detail. And the simulation model which is established to characterize the feature of the tandem light guide system will also be represented as well as the simulation results.

3.2 Simulation Software

The optical simulator TracePro^® is developed by Lambda Research Corporation, and it is a general ray tracing software for illumination analysis, optical analysis, radiometry analysis, and photometry analysis. It is used to simulate and optimize the uniformity and efficiency of the wedged light guide combining with prismatic micro-bump structures.

3.3 Proposed Backlight Module for FSC LCD

Coupling with the optically-compensated-bend (OCB) mode LC and full color LEDs, a FSC LCD can be realized without any color filters. However, due to the insufficient LC saturated time, the duty cycle of the LEDs is restricted, and that will

cause the backlight not bright enough. In order to improve the issue of the LC response time in FSC LCD application, a specific configuration of the backlight system is required: spatial-temporal partitioned scanning backlight driven by FSC method.

In the previous research, a small size FSC scanning backlight module has been designed and fabricated by using side lit approach [22]. However, this side emitting type BLM is inadequate for developing the large scale FSC scanning backlight system.

Therefore, the objective of proposed BLM should be able to accomplish the large size FSC spatial-temporal partitioned scanning backlight system.

The structure of the spatial-temporal scanning backlight for an OCB-mode FSC-LCD is shown in Fig. 3-1(a). The backlight consists of

(1) Tandem wedge shaped light guides have prismatic micro-bump structures over the bottom that control the direction of the light extraction as well as the uniformity and efficiency as shown in Fig. 3-1(b),

(2) 4-in-1 full color LEDs light bar is set in front of the incident surface of the corresponding LG and under the end of the previous one,

(3) A diffuser on the top of the wedge shaped light guides that diffuses light and illuminates local dim regions, and

(4) A brightness enhanced film (BEF) with a saw-tooth cross-section guide light toward the front direction in order to increase the normal component of the light toward the LCD.

Each tandem wedge shaped light guide for FSC LCD application should be made block-wise, so that the illumination of the single block will be isolated from other blocks. If there is no partition between the LG plates, the light from one single block

LED light-bar corresponding single block which consists of 15 packages of 4-in-1 R, G, G, B LEDs are adopted as the light source.

(a)

(b)

Fig. 3-1 (a) Structure of FSC scanning BLM; (b) wedge shaped light guide unit.

3.3.1 Specifications and Criterions

Table 3-1 Specifications and criterions of proposed BLM.

The specification of the developed prototype OCB mode FSC LCD is shown in Table. 3-1. The tandem wedge shaped light guides are assembled into 32-inch diagonal panel size with aspect ratio 16:9. The entire BLM is divided into 12 horizontal blocks in the consideration of LC response time, the optical efficiency of the light guide unit, and the panel resolution. The vertical pixel numbers should be adequate choice for our model, and the following condition should be satisfied [23].

[1 sub-frame time] > = [scanning time of the whole gate lines] / [scanning blocks]

+ [response time of liquid crystal] + [LED flashing time] (3-1) Each color sub-frame takes about 1/180 sec. The duty cycle of the illumination pulse is limited to about 50%. From the specification, the total number of gate lines is 768. If a BLM is divided into 12 scanning partitions, each single block will be in charge of 64 lines of the TFT pixel array correspondingly, and the scanning duration over those lines is about 0.45 ms, which is the acceptable tolerance for the LC response time. Fig. 3-2 shows the simple time relation on scanning of LC, TFT-array cell and LED backlight.

The criterions of the uniformity for full panel and efficiency per light guide unit are 80% and 70%, respectively. The thickness of the overall BLM is expected to less than 30mm and without any shields.

Fig. 3-2 Timing chart for multi-flashing method

3.4 Design of Wedge Shaped Light Guide

Considering the optical efficiency of the wedge shaped light guide unit and the required space for setting LED light bar, the tilt angle of the wedge shaped light guide δ is defined by 13.6 degree as shown in Fig. 3-3.

Fig. 3-3 Schematic diagram of tandem wedge shaped light guide

In order to obtain the uniformity and directional light distribution, the prismatic micro-bump structures are built along the bottom of the light guide. For convenience, every LED is regarded as a point source; therefore, the relationship between incident ray, exit light and micro-bump structures can be illustrated with Fig. 3-4. The incident angle is defined as α, and the tilt angle of wedge type LG and incline angle of micro-bump structure are determined by γ and δ, respectively. Accordingly, the exit angle β can be obtained by Snell‟s law of refraction as this formula.

β =sin ¹[n_PMMAsin(α-(2γ+2δ)] (3-2)

Fig. 3-4 Light extraction is introduced by a reflective micro-bump structure.

By the eq. (3-2), the relation between the exit angle β, incident angle α, and incline angle γ can be subsequently established in Fig. 3-5. In this graph, each curve refers to the light with a specific incline angle. The gray region is the desired range of exit angle. For each particular incident angle, there exists a better corresponding inclined angle to make the exit angle approximately in the gray region.

Fig. 3-5 Exit angles achieved by refraction for different inclined angles and incident

3.5 Simulation Model of FSC Scanning Backlight System

It may be noted that the tandem wedge shaped light guide is connected with each other and congregated into 32 inch panel size. In order to consider the time of simulating ray-tracing, initial design and optimization process are carried out under the single light guide unit as shown in Fig. 3-6. The simulation module consists of the main body of backlight system, i.e. LEDs light-bar, wedge shaped light guide, reflector, and a detector placed in top of light guide.

Fig. 3-6 Simulation module of wedge shaped light guide.

The illuminating property of the light source in simulation, as shown in Fig. 3-7, is set as quasi-lambertian. The spectrum of Red, Green, and Blue were 636nm, 456nm, and 522nm, respectively. The refractive index of the light guide unit is set as 1.49 which is the same as that of PMMA. In addition, the detector is set to detect the angular distribution and the uniformity of the outcoupled light, which serve as the function of the optimization.

3.6 Optimization of FSC Scanning Backlight System

After establishing the simulation model, the next part is to optimize the parameters of the wedge shaped light guide to obtain the desired optical characteristic.

The parameters include the incline angle of the prismatic micro-bump structure, micro-bump width, and micro-bump gap as shown in Fig.3-8. The incline angle affects the angular distribution of the emitted light, and micro-bump width, and the gap determines the uniformity.

(a)

(b)

Fig. 3-7 (a) Radiation characteristic of LEDs emitting (b) Relarive spectral distribution

Fig. 3-8 Parameters of wedge shaped light guide and effected factor.

3.6.1 Optimization of Angular Distribution

Since the refractive index of PMMA is set 1.49, and the angular distribution of the LED is about ±90°, the maximum of theθ , the refractive angle of the LED light _i passing trough the incident surface of light guide, will reach to ±42°. Therefore,

Parameters Effected factor Incline angle Angular distribution Micro-bump Width

Uniformity Gap

approximately with ±30°.

In the beginning, the tilted surface of the wedge shaped light guide can be roughly classified into four parts for different incident angle in order to choose the suitable incline angle as shown in Fig. 3-9. The first part is the refractive angle θ _i from -42° to -10°. Then, the second part is the refractive angle larger than -10° but smaller than 0° and the light which is reflected once by TIR has the refractive angle from 20° to 42°. The third part is the refractive angle approximately with 0° and the light which is reflected once by TIR has the refractive angle from 15° to 20°. The finally region is the distribution of refractive angle as 5° to 15°.

For convenience, the simulation in this part is carried out under a uniform the incident angle α in region (3) is from 70° to 90° as shown in the yellow region in Fig. 3-10. The yellow region and the gray region intersect at the curve γ near 30°.

Therefore, the angular distribution simulates under the various incline angles γ from 28° to 34°. As seen in Fig. 3-11 (a), the angular distribution at incline angle γ = 32°

has the most narrow Full Width at Half Maximum (FWHM) about ±15°, and approximate to the normal of output surface comparing with other angles. By the same steps, the region (2), (3), and (4) are also simulated incline angle from δ= 28° to 34°, and δ= 32° to 38°. According to the limitation of fabrication process, the different incline angle in corresponding regions are chosen by 40°, 30°, 32°, and 35°, respectively.

(a)

(b)

(c)

(d)

Fig. 3-9 The relation between the refractive angle and zone distribution in (a) region 1, (b) region 2, (c) region 3, and (d) region 4.

(b)

Fig. 3-10 (a) corresponding incline angle in region (3).

(a) (b)

(c) (d)

Fig. 3-11 The angular distribution simulated under the various incline angles in (a) region 1, (b) region 2, (c) region 3, and (d) region 4.

3.6.2 Optimization of Uniformity

Due to the light intensity decreases with the distance apart from light source, the dimension of the micro-bump structure should be changed to keep the uniformity.

The basic principle is that control the filling factor as mentioned in the following formula as shown in Fig. 3-8.

Filling factor = Micro-bump width / Micro-bump width + Micro-bump gap (3-3) The filling factor near to the light source should be smaller than the farther one.

In order to simplify the light guide structures, the micro-bump width is fixed, and only adjusts the gap of the micro-bump to achieve uniform distribution. The flow optimization chart for the encoding scheme is summarized in Fig 3-11. First, the entire light guide is equally divided into several areas. Then, fixing the micro-bump width and varying the filling factor in each region try to get the high uniformity. If the uniformity can not be improved by changing any filling factor, then adjust the relative area of each region. Such steps will be repeated several times, and obtain the highest uniformity finally.

Uniformity = Min. luminance / Max. Luminance (3-4)

3.7 Simulations

According to the optimization flow chart in Fig 3-12, the illumination map of the unitary wedge shaped light guide with prismatic micro-bump structures is optimized as shown in Fig. 3-13. The uniformity of the extraction light and the efficiency of the single light guide unit are 93% and 68%, respectively. Furthermore, the directivity of the extraction light is indeed enhanced. The light coupling from light guide has a narrow divergent angle ±15° on the y-z plane, and about ±50° on the x-z plane, which are shown in Fig. 3-14. This directionality is expected to reduce the probability of the light that leak to the region whose state of the LC is still changing.

Fig. 3-13 Luminance map of wedge shared light guide unit.

Fig. 3-14 Rectangular candela distribution plot for divergence of extraction light.

As mentioned above, the tandem light guide will be assembled and connected with each other into a large panel size LCD. Therefore, each connected boundary between light guides is an important issue. The partitions must be not visible from any viewing angle on the panel due to the shade or obviously discontinuous luminance severely affects the display performance. For this reason, except for simulating the

As mentioned above, the tandem light guide will be assembled and connected with each other into a large panel size LCD. Therefore, each connected boundary between light guides is an important issue. The partitions must be not visible from any viewing angle on the panel due to the shade or obviously discontinuous luminance severely affects the display performance. For this reason, except for simulating the