- Email: [email protected]
Laser marking system for light guide panel using design of experiment and web-based prototyping
ARTICLE IN PRESS Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
Contents lists available at ScienceDirect
Robotics and Computer-Integrated Manufacturing journal homepage: www.elsevier.com/locate/rcim
Laser marking system for light guide panel using design of experiment and web-based prototyping Hyuk-Jin Kang, Hyung-Jung Kim, Ji-Seok Kim, Woon-Yong Choi, Won-Shik Chu, Sung-Hoon Ahn n School of Mechanical and Aerospace Engineering & Institute of Advanced Machinery and Design, Seoul National University, Seoul 151-744, South Korea
a r t i c l e in fo
abstract
Article history: Received 28 June 2006 Received in revised form 12 January 2010 Accepted 15 April 2010
A light guide panel (LGP) is an element of the liquid crystal display (LCD) back light unit (BLU), which is used for display devices. In this study, the laser marking process is applied to the fabrication of light guide panels as the new fabrication process. In order to obtain a light guide panel which has high luminance and uniformity, four principal parameters such as power, scanning speed, ratio of line gap, and number of line were selected. A web-based design tool was developed to generate patterns of light guide panel via the network, and the tool may assist the designer to develop various prototype patterns. Topcon-BM7 was used for luminance measurement of each specimen with 100 mm 100 mm area. By Taguchi method, optimized levels of each parameter were found, and luminance of 3523 cd/cm2 and uniformity of 92% were achieved using the laser machined BLU. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Back light unit Light guide panel CO2 laser marking Design of experiment Web-based system
1. Introduction Recently, the display industry has been rapidly expanding, and companies have developed new technologies such as electroluminescence, plasma display panel (PDP), and thin-film transistor (TFT)-LCD. Among these, TFT-LCD needs a light source because TFT-LCD does not have its own lighting mechanism. The back light unit is one of the components of an LCD which provides light source from the back side of a TFT-LCD panel [1]. BLU has many components as shown in Fig. 1. High luminance, uniformity, and low current consumption are the key factors to determine BLU’s performance, and pattern design for the LGP is critical. LGP, which is one of the BLU’s main components, is used to transform linear source of light to planar beams by creating patterns for light reflection, and is usually made of PMMA (polymethylmethacrylate). Currently, such technologies are available to manufacture BLU as silk screen printing, injection molding, stamping, and V-cutting [2]. Each technology has the advantages and disadvantages in terms of performance, cost of manufacture, productivity, and reproducibility. Recently, laser marking technology has been used in order to reduce the time for design and development, which includes adjustment of model’s patterns, mold, and stamper [3]. As laser marking employs print-less technology, it does not reflect light by printed patterns. In addition, cutting surface is more specular than that of mechanical V-cutting, resulting in higher
n
Corresponding author. Tel.: + 82 2 880 7110; fax: + 82 2 883 0179. E-mail address: [email protected] (S.-H. Ahn).
0736-5845/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.rcim.2010.04.003
reflection, and thus luminance is higher than that by printed patterns [2]. To cope with customer’ needs and to survive in the highly competitive business environment, companies have to develop effective methods in the product development and configurations of manufacturing organizations. As a result, integrated information technology (IT) models that one enterprise possesses and uses to manage their required IT infrastructure are rapidly changing into distributed manufacturing area offered by specialized IT services [4,5]. In accordance with this trend, many manufacturing enterprises are employing strategies whereby they carry out research on highly applicable technologies in the design and manufacturing processes to save the time and cost. The increasing importance of information in product development processes is leading many researchers to consider a web-based collaborating environment for design and manufacturing [6–8]. The objective of this research is to utilize advantages of laser marking technology, which enables short patterning time and high luminance in the fabrication of BLUs and develop a webbased design system for pattern design. Designers can access the tool via the web browsers and prototype design can be rapidly made. Also, the concept of D.O.E. (design of experiment) was applied in order to obtain uniform and high luminance LGP.
2. Web-based design system of LGP In this paper, a laser-marking process is applied to the patterning of V-groove reflectors in LGP. V-groove is a reflector [9], and the gaps between patterns are important parameters in
ARTICLE IN PRESS 536
H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
Fig. 1. The typical structure of TFT-LCD. Fig. 3. Generated tool path of laser machining for light guide panel (both in x and y directions).
a design tool was developed, and the interface is based on a web browser. Fig. 2(a) shows the web-based pattern design tool for LGP. The design tool may assist designers to design patterns of LGP without using generic design programs like CAD. The design tool uses control points of Bezier curve, so it is able to easily change the gap between patterns (Fig. 2(b)). The gap between patterns governs the difference in the amount of reflected light from the light source. For a BLU with twin lamps at each longitudinal edge, center of the LGP receives less amount of light than sides of LGP. So, it is required to make dense pattern at the center in order to increase reflected amount of light. The control parameter ‘1 refers to the first gap between the one edge of LGP and the first groove, ‘2 refers to the gap between the first groove and the second groove, and ‘n refers to the gap between the n 1th groove and the nth groove. This web based design tool uses the flash (Flash FX) as user interface [10]. Using flash script, the Bezier curve is interactively generated by the control point data. Those generated points are automatically converted into NC (Numerical Control) codes for laser machining by a Matlab code, and the NC codes are transmitted to the designer through the web. The x- and y-directional groove patterns can be reviewed on the web-based viewing tool. Fig. 3 shows the generated tool path of laser machining when the ratio of line gap is 100:30 and size of work piece is 100 mm 100 mm (x and y directions). The sequential procedure of pattern design is described in detail as follows:
Fig. 2. (a) User interface of web-based pattern design tool for light guide panel and (b) definition of the gap between grooves.
order to improve uniformity of reflected light. The most important role of grooves is to reflect light from the lamp and to make uniform distribution of the reflected light. Grooves near the source of light get sufficient amount of light from the lamp. So, in this location, to reflect the light to the frontface direction, less number of reflectors are necessary. On the other hand, many reflectors are required at distant locations from the source of light. Designer adjusts the interval between each pattern and the number of patterns during design process of LGP. In this research,
1. Define number of required points. 2. Select an axis of pattern. 3. Control a curve using blue and red points. A. Blue points: base point of curve B. Red points: gradient control point 4. Define zero line (base line of the pattern). 5. Click ‘Save value’ button (x, y axis). 6. Click ‘Submit’ button. The laser equipment used in this research recognizes DWG (Drawing) files; therefore, a module was developed to convert NC codes into DWG files. Fig. 4 shows a communication system of the design tool for prototyping of LGP patterns. Such procedures are quickly performed through the web, thereby facilitating the designer to produce the trial product of a prototype in short period of time [11]. In addition, the web communication has an advantage in accessibility. Through the internet, the designer and
ARTICLE IN PRESS H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
537
Fig. 5. Cross sectional dimensions of V-grooves as a function of speed (power: 50 W).
Fig. 4. Communication structure of web-based LGP design and fabrication system.
Table 1 Levels of selected factors. Level
manufacturer from different locations can fabricate various prototypes of LGP.
3. Experimental 3.1. D.O.E (design of experiment) The Taguchi method is one of the good methods in deciding level of factors for luminance and uniformity of the LGP. The quality of the pattern, which is made by laser, is determined by various factors. Thus, determination of the processing conditions is an important part of the experiment which minimizes the number of experiments and can maximize the distinctive value of characteristic pattern, if any unexpected external disturbance comes into effect [12]. SN ratio (signal-to-noise ratio) is used to minimize noise factor and maintain product’s robustness using the robust design concept of the Taguchi method. The SN ratio is defined as the ratio of the magnitudes of the input signals to the signal affected by the noise. Hence, the value which maximizes the SN ratio of each controlled factor becomes robust against noise. The definition of SN ratio varies according to object function, and thus this function becomes special factor. The main objective of this experiment, that is, luminance and uniformity, exhibits largerthe-better characteristics since the higher the values of luminance and uniformity the better the results are. For reducing expected loss, mean squared deviation (MSD) is calculated using Eq. (1) [13]. MSD ¼
n 1X 1 n i ¼ 1 y2i
The SN ratio given in Eq. (2) uses Eq. (1) [7,8]. " # n 1X 1 SN ¼ 10 log n i ¼ 1 y2i
ð1Þ
ð2Þ
¼ 10 log½MSD where n is the number of values of measurement and y the measured value of property. In the case of larger-the-better characteristics, as the expected value decreases, the SN ratio increases, therefore the luminance and uniformity improve.
Factor
1
2
3
Power (W) Scanning speed (mm/s) Ratio of line gap (‘1: ‘1) Number of lines
30 30 100:30 80
40 40 100:40 85
50 50 100:50 90
Table 2 Orthogonal arrays and experimental results. Exp. no.
A
B
C
D
Average luminance (cd/m2)
Uniformity (%)
Time (s)
1 2 3 4 5 6 7 8 9
1 1 1 2 2 2 3 3 3
1 2 3 1 2 3 1 2 3
1 2 3 2 3 1 3 1 2
1 2 3 3 1 2 2 3 1
3339 3415 3425 3523 3353 3442 3452 3326 3280
68.4 72.5 80.4 92 77.9 66.6 85.2 62.2 65.4
56 59 71 49 52 63 62 47 55
3.2. Preliminary experiment In this paper, a CO2 laser (M-330, LVI tech) in TEM00 mode is used, whose power output ranges from 0 to 50 W, and wavelength measures 10.6 mm. Laser ablation is affected by laser power, wavelength, scanning speed, and absorption of light while during polymerization [14]. The most influential factors are laser power and scanning speed [14–16]. For this reason, these two factors were selected for preliminary experiment. The diameter of beam and focus of laser was fixed. The best condition for manufacturing V-groove on PMMA during preliminary experiment was achieved by changing power and scanning speed. Fig. 5 shows the laser machined cross section of V-groove on PMMA. For the case where a constant power was applied, as the scanning speed increased, the depth of V-groove gets reduced, but the width remained almost constant; on the other hand where a constant scanning speed was applied, as the power increased, the depth of V-groove increased, but the width remained almost constant.
ARTICLE IN PRESS 538
H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
From the test, laser power was selected in the range 30–50 W, and scanning speed was selected from 30 to 50 mm/s at intervals of 10 mm/s.
3.3. Selection of factors From the preliminary experiment, preference is given to laser power, scanning speed, and machined pattern. The levels of selected factors are presented in Table 1. The number of lines implies total number of lines of the pattern. The effective factors (power, scanning speed, ratio of line gap, and number of lines) are selected. According to the result, the possible machining of the minimum width of V-groove is about 210 mm which is due to limitation of spot size depending on CO2 laser’s wavelength. Hence, gap between lines and maximum level of line selected for each V-groove do not overlap in the 100 mm 100 mm range.
3.4. Method of experiment
Fig. 6. Setup for DOE study.
The D.O.E uses orthogonal array, which ensures safety and reproducibility, minimizes size of the experiment, and enables to include many factors and levels. Hence, orthogonal arrays and experimental results are presented in Table 2, and selection of 4 factors and their 3 levels (L934) of orthogonal array were made. Each factor comprises of laser power (A), scanning speed (B), ratio of line gap (‘1:‘2), and (C), number of lines (D). Each factor’s level is shown in Table 2. Laser marking processed on PMMA was based
Fig. 7. Luminance distribution of laser machined light guide panels. (Sub captions represent: laser power, scanning speed, ratio of pattern gap, and number of line, respectively).
ARTICLE IN PRESS H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
on factors in Table 1. The area of PMMA was 100 mm 100 mm and thickness was 8 mm. To measure the luminance of each LGP, reflector sheet was attached to the bottom face, right side, and left side (Fig.6). Two 38 W CCFL (Cold Cathode Fluorescent Lamp) were attached to the two facing sides of the LGP. Diffuser sheet was laid on the front face. Fig. 7 and Table 2 show luminance and uniformity data measured from 25 points on LGP. Luminance was measured by BM-7 (Topcon). 3.5. Results
and {A2, B1, C3, D3}, respectively. In other words, with 40 W power, 30 mm/s scanning speed, 85 lines, and 100:50 ratio, high luminance was observed, while with 40 W power, 30 mm/s scanning speed, 90 lines, and 100:50 ratio, high uniformity was observed. Except for the number of lines, other factors had the same values. From the percentage contributions of four factors, two geometric factors, number of lines and ratio of line gap, have more effect on the quality of fabricated LGP than two laser marking factors, power and scanning speed. This shows that the advantage of proposed design tool to control the geometric parameters of LGP pattern. In detail, the number of lines has the
Table 3 shows SN ratios of luminance and uniformity. The respond tables for SN ratios of the luminance and uniformity are presented in Tables 4 and 5. Deviation implies difference between maximum value and minimum value of respond for the SN ratio. Percentage contribution implies percentage of total levels [12]. Therefore, luminance is most affected by factor D and uniformity by factor C. Table 2 shows that the experiment no. 4 provided the best result. Improvement in luminance was more than 4% compared with the prepared model, and uniformity was more than 8%. It is shown that the luminance and the uniformity are higher than the existing commercial LGP whose luminance is in the range of 1500–3000 cd/m2, and its uniformity is about 85%. From Figs. 8 and 9, optimal pattern conditions for high luminance and high uniformity are selected as {A2, B1, C3, D2} Fig. 8. Main effect plot for average SN ratio for high luminance. Table 3 SN ratio of luminance and uniformity. Exp. no. 1 2 3 4 5 6 7 8 9
SN ratio of luminance
SN ratio of uniformity
70.41 70.62 70.67 70.93 70.48 70.65 70.75 70.34 70.23
36.53 37.07 38.16 39.21 37.69 36.57 38.45 36.01 36.40
Table 4 Respond table for SN ratio of luminance. Factor
A B C D n
Fig. 9. Main effect plot for average SN ratio for high uniformity.
Effect 1
2
3
70.57 70.69 70.47 70.37
70.68 70.48 70.59 70.67
70.44 70.51 70.63 70.64
n
Deviation
PC (%)
0.24 0.21 0.16 0.30
26.4 23.1 17.6 33.0
Deviation
PCn (%)
0.87 1.14 1.73 0.92
18.7 24.5 37.1 19.7
PC: percentage contribution.
Table 5 Respond table for SN ratio of uniformity. Factor
A B C D n
Effect 1
2
3
37.25 38.06 36.27 36.87
37.82 36.92 37.56 37.36
36.95 37.04 38.10 37.79
PC: percentage contribution.
539
Fig. 10. An example of cross sections (50 W, 40 mm/s, 100:30, 90).
ARTICLE IN PRESS 540
H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
greatest effect on luminance, while ratio of line gap has the greatest effect on uniformity. One may expect that higher line density by 100:30 ratio improves uniformity than lower density by 100:50 ratio, but experimental result is contradictory. The possible reason for this is the numerous lines at the center resulting in the concentration of light at the center, and thus uniformity decreases when the ratio is 100:30. Eq. (3) shows the average luminance of SN ratio by the improved condition.
Although the proposed system provides a new idea for fast prototyping of LGP, it should be improved further to be used in production line for mass fabrication. On the other side, as a platform of IT services, the multi-agent system can be alternative of the web-based system to cover wider processes in design and manufacturing in flexible and rapid manner.
m ¼ A2 þB1 þ C 3 þ D2 3T
Acknowledgement
ð3Þ
T is an average of SN ratio representing level, and its value is 70.56. The average of SN ratio by improved condition is 70.99, and average of SN ratio by conditions (A1, B1, C3, and D3) is 70.42, so net improvement is 0.57. In the same way, the uniformity of SN ratio is expected to improve by 3.23 because T is 37.33, average SN ratio is 36.56 (by present conditions) and 39.79 (by improved conditions at A2, B1, C3, and D3). Fig. 10 shows a cross-section of LGP pattern by laser marking (based on orthogonal array). The power and scanning speed of the laser are important parameters in order to control the shape of V-groove and the optical property of LGP. From the experiment, the best condition for making V-groove with fine optical property is 40 W power and 30 mm/s scanning speed.
4. Conclusions The main advantage of web-based laser marking system for LGP is accessibility to the design tool and increase in the efficiency for generating an optimized pattern with fast prototyping process. This system enables a designer to understand the tendency of optical performance according to variation of each factor before conducting complicated optical analysis which demands high cost and expert knowledge. Effective factors for making LGP by laser and DOE for obtaining improved levels for the factors were investigated as follows: 1) Four factors namely, power, scanning speed, gap between pattern lines, and number of lines were selected, and levels were decided accordingly. 2) 40 W, 30 mm/s, and 100:50 ratio provided the highest luminance (with 85 lines) and the highest uniformity (with 90 lines). 3) The number of lines had the greatest effect on luminance, and the ratio of line gap had the greatest effect on uniformity. 4) By Taguchi method, optimized levels of each parameter were found, and luminance of 3523 cd/cm2 and uniformity of 92% were achieved using the laser machined BLU.
This work was supported by Grants from the second stage of Brain Korea 21 of Seoul National University, the Seoul R&BD Program (TR080578), and the KOSEF and the MEST Program (no. 20090081391). References [1] Kim GD, Kang HJ, Ahn SH, Song CK, Back CI, Lee CS. Laser-marking process for liquid-crystal display light guide panel. P I Mech Eng B—J Eng 2005;219: 565–9. [2] Dubey AK, Yadava V. Simultaneous optimization of multiple quality characteristics in laser beam cutting using Taguchi method. Int J Precis Eng Man. 2007;8:10–5. [3] Pinski EF, Witkowski JS, Abramski KM. Diffractive scanning mechanism for laser marker. Opt Laser Technol 2000;32:33–7. [4] Sriram RD, Szykman S, Durham D. Special issue on collaborative engineering. J Comput Inf Sci Eng 2006;6:93–5. [5] Shen W, Chao KM. Special issue on techniques to support collaborative engineering environments. Adv Eng Inf 2007;21:181. [6] Ahn S-H, Kim DS, Chu WS, Jun CS. MIMS: web-based micro machining service. Int J Comput Integ Manuf 2005;18:251–9. [7] Kim H-J, Chu WS, Ahn SH, Kim DS, Jun CS. Web-based design and manufacturing systems for micromachining: comparison of architecture and usability. Comput Appl Eng Educ 2006;14:169–77. [8] Chu CH, Cheng CY, Wu CW. Applications of the web-based collaborative visualization in distributed product development. Comput Ind 2006;57: 272–82. [9] Lin CS, Wu WZ, Lay YL, Chang MW. A digital image-based measurement system for a LCD backlight module. Opt Laser Technol 2001;33:499–505. [10] Adobes Flashs, www.adobe.com (2009). [11] Ahn SH, Sundararajan V, Smith C, Kannan B, D’Souza R, Sun G, et al. CyberCut: an internet-based CAD/CAM system. J Comput Inf Sci Eng 2001;1:52–9. [12] Chen YH, Tam SC, Chen WL, Zheng HY. Application of the Taguchi method in the optimization of laser micro-engraving of photomasks. Int J Mater Prod Technol 1996;11:333–44. [13] Phillip JR. Taguchi techniques for quality engineering: loss function, orthogonal experiments, parameter and tolerance design. New York: McGraw-Hill; 1996. [14] Liu ZQ, Feng Y, Yi XS. Coupling effects of the number of pulses, pulse repetition rate and fluence during laser PMMA ablation. Appl Surf Sci 2000;165:303–8. [15] Wang SC, Lee CY, Chen HP. Thermoplastic microchannel fabrication using carbon dioxide laser ablation. J Chromatogr A 2006;1111:252–7. [16] Zhou BH, Mahdavian SM. Experimental and theoretical analyses of cutting nonmetallic materials by low power CO2-laser. J Mater Process Technol 2004;146:188–92.
Contents lists available at ScienceDirect
Robotics and Computer-Integrated Manufacturing journal homepage: www.elsevier.com/locate/rcim
Laser marking system for light guide panel using design of experiment and web-based prototyping Hyuk-Jin Kang, Hyung-Jung Kim, Ji-Seok Kim, Woon-Yong Choi, Won-Shik Chu, Sung-Hoon Ahn n School of Mechanical and Aerospace Engineering & Institute of Advanced Machinery and Design, Seoul National University, Seoul 151-744, South Korea
a r t i c l e in fo
abstract
Article history: Received 28 June 2006 Received in revised form 12 January 2010 Accepted 15 April 2010
A light guide panel (LGP) is an element of the liquid crystal display (LCD) back light unit (BLU), which is used for display devices. In this study, the laser marking process is applied to the fabrication of light guide panels as the new fabrication process. In order to obtain a light guide panel which has high luminance and uniformity, four principal parameters such as power, scanning speed, ratio of line gap, and number of line were selected. A web-based design tool was developed to generate patterns of light guide panel via the network, and the tool may assist the designer to develop various prototype patterns. Topcon-BM7 was used for luminance measurement of each specimen with 100 mm 100 mm area. By Taguchi method, optimized levels of each parameter were found, and luminance of 3523 cd/cm2 and uniformity of 92% were achieved using the laser machined BLU. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Back light unit Light guide panel CO2 laser marking Design of experiment Web-based system
1. Introduction Recently, the display industry has been rapidly expanding, and companies have developed new technologies such as electroluminescence, plasma display panel (PDP), and thin-film transistor (TFT)-LCD. Among these, TFT-LCD needs a light source because TFT-LCD does not have its own lighting mechanism. The back light unit is one of the components of an LCD which provides light source from the back side of a TFT-LCD panel [1]. BLU has many components as shown in Fig. 1. High luminance, uniformity, and low current consumption are the key factors to determine BLU’s performance, and pattern design for the LGP is critical. LGP, which is one of the BLU’s main components, is used to transform linear source of light to planar beams by creating patterns for light reflection, and is usually made of PMMA (polymethylmethacrylate). Currently, such technologies are available to manufacture BLU as silk screen printing, injection molding, stamping, and V-cutting [2]. Each technology has the advantages and disadvantages in terms of performance, cost of manufacture, productivity, and reproducibility. Recently, laser marking technology has been used in order to reduce the time for design and development, which includes adjustment of model’s patterns, mold, and stamper [3]. As laser marking employs print-less technology, it does not reflect light by printed patterns. In addition, cutting surface is more specular than that of mechanical V-cutting, resulting in higher
n
Corresponding author. Tel.: + 82 2 880 7110; fax: + 82 2 883 0179. E-mail address: [email protected] (S.-H. Ahn).
0736-5845/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.rcim.2010.04.003
reflection, and thus luminance is higher than that by printed patterns [2]. To cope with customer’ needs and to survive in the highly competitive business environment, companies have to develop effective methods in the product development and configurations of manufacturing organizations. As a result, integrated information technology (IT) models that one enterprise possesses and uses to manage their required IT infrastructure are rapidly changing into distributed manufacturing area offered by specialized IT services [4,5]. In accordance with this trend, many manufacturing enterprises are employing strategies whereby they carry out research on highly applicable technologies in the design and manufacturing processes to save the time and cost. The increasing importance of information in product development processes is leading many researchers to consider a web-based collaborating environment for design and manufacturing [6–8]. The objective of this research is to utilize advantages of laser marking technology, which enables short patterning time and high luminance in the fabrication of BLUs and develop a webbased design system for pattern design. Designers can access the tool via the web browsers and prototype design can be rapidly made. Also, the concept of D.O.E. (design of experiment) was applied in order to obtain uniform and high luminance LGP.
2. Web-based design system of LGP In this paper, a laser-marking process is applied to the patterning of V-groove reflectors in LGP. V-groove is a reflector [9], and the gaps between patterns are important parameters in
ARTICLE IN PRESS 536
H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
Fig. 1. The typical structure of TFT-LCD. Fig. 3. Generated tool path of laser machining for light guide panel (both in x and y directions).
a design tool was developed, and the interface is based on a web browser. Fig. 2(a) shows the web-based pattern design tool for LGP. The design tool may assist designers to design patterns of LGP without using generic design programs like CAD. The design tool uses control points of Bezier curve, so it is able to easily change the gap between patterns (Fig. 2(b)). The gap between patterns governs the difference in the amount of reflected light from the light source. For a BLU with twin lamps at each longitudinal edge, center of the LGP receives less amount of light than sides of LGP. So, it is required to make dense pattern at the center in order to increase reflected amount of light. The control parameter ‘1 refers to the first gap between the one edge of LGP and the first groove, ‘2 refers to the gap between the first groove and the second groove, and ‘n refers to the gap between the n 1th groove and the nth groove. This web based design tool uses the flash (Flash FX) as user interface [10]. Using flash script, the Bezier curve is interactively generated by the control point data. Those generated points are automatically converted into NC (Numerical Control) codes for laser machining by a Matlab code, and the NC codes are transmitted to the designer through the web. The x- and y-directional groove patterns can be reviewed on the web-based viewing tool. Fig. 3 shows the generated tool path of laser machining when the ratio of line gap is 100:30 and size of work piece is 100 mm 100 mm (x and y directions). The sequential procedure of pattern design is described in detail as follows:
Fig. 2. (a) User interface of web-based pattern design tool for light guide panel and (b) definition of the gap between grooves.
order to improve uniformity of reflected light. The most important role of grooves is to reflect light from the lamp and to make uniform distribution of the reflected light. Grooves near the source of light get sufficient amount of light from the lamp. So, in this location, to reflect the light to the frontface direction, less number of reflectors are necessary. On the other hand, many reflectors are required at distant locations from the source of light. Designer adjusts the interval between each pattern and the number of patterns during design process of LGP. In this research,
1. Define number of required points. 2. Select an axis of pattern. 3. Control a curve using blue and red points. A. Blue points: base point of curve B. Red points: gradient control point 4. Define zero line (base line of the pattern). 5. Click ‘Save value’ button (x, y axis). 6. Click ‘Submit’ button. The laser equipment used in this research recognizes DWG (Drawing) files; therefore, a module was developed to convert NC codes into DWG files. Fig. 4 shows a communication system of the design tool for prototyping of LGP patterns. Such procedures are quickly performed through the web, thereby facilitating the designer to produce the trial product of a prototype in short period of time [11]. In addition, the web communication has an advantage in accessibility. Through the internet, the designer and
ARTICLE IN PRESS H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
537
Fig. 5. Cross sectional dimensions of V-grooves as a function of speed (power: 50 W).
Fig. 4. Communication structure of web-based LGP design and fabrication system.
Table 1 Levels of selected factors. Level
manufacturer from different locations can fabricate various prototypes of LGP.
3. Experimental 3.1. D.O.E (design of experiment) The Taguchi method is one of the good methods in deciding level of factors for luminance and uniformity of the LGP. The quality of the pattern, which is made by laser, is determined by various factors. Thus, determination of the processing conditions is an important part of the experiment which minimizes the number of experiments and can maximize the distinctive value of characteristic pattern, if any unexpected external disturbance comes into effect [12]. SN ratio (signal-to-noise ratio) is used to minimize noise factor and maintain product’s robustness using the robust design concept of the Taguchi method. The SN ratio is defined as the ratio of the magnitudes of the input signals to the signal affected by the noise. Hence, the value which maximizes the SN ratio of each controlled factor becomes robust against noise. The definition of SN ratio varies according to object function, and thus this function becomes special factor. The main objective of this experiment, that is, luminance and uniformity, exhibits largerthe-better characteristics since the higher the values of luminance and uniformity the better the results are. For reducing expected loss, mean squared deviation (MSD) is calculated using Eq. (1) [13]. MSD ¼
n 1X 1 n i ¼ 1 y2i
The SN ratio given in Eq. (2) uses Eq. (1) [7,8]. " # n 1X 1 SN ¼ 10 log n i ¼ 1 y2i
ð1Þ
ð2Þ
¼ 10 log½MSD where n is the number of values of measurement and y the measured value of property. In the case of larger-the-better characteristics, as the expected value decreases, the SN ratio increases, therefore the luminance and uniformity improve.
Factor
1
2
3
Power (W) Scanning speed (mm/s) Ratio of line gap (‘1: ‘1) Number of lines
30 30 100:30 80
40 40 100:40 85
50 50 100:50 90
Table 2 Orthogonal arrays and experimental results. Exp. no.
A
B
C
D
Average luminance (cd/m2)
Uniformity (%)
Time (s)
1 2 3 4 5 6 7 8 9
1 1 1 2 2 2 3 3 3
1 2 3 1 2 3 1 2 3
1 2 3 2 3 1 3 1 2
1 2 3 3 1 2 2 3 1
3339 3415 3425 3523 3353 3442 3452 3326 3280
68.4 72.5 80.4 92 77.9 66.6 85.2 62.2 65.4
56 59 71 49 52 63 62 47 55
3.2. Preliminary experiment In this paper, a CO2 laser (M-330, LVI tech) in TEM00 mode is used, whose power output ranges from 0 to 50 W, and wavelength measures 10.6 mm. Laser ablation is affected by laser power, wavelength, scanning speed, and absorption of light while during polymerization [14]. The most influential factors are laser power and scanning speed [14–16]. For this reason, these two factors were selected for preliminary experiment. The diameter of beam and focus of laser was fixed. The best condition for manufacturing V-groove on PMMA during preliminary experiment was achieved by changing power and scanning speed. Fig. 5 shows the laser machined cross section of V-groove on PMMA. For the case where a constant power was applied, as the scanning speed increased, the depth of V-groove gets reduced, but the width remained almost constant; on the other hand where a constant scanning speed was applied, as the power increased, the depth of V-groove increased, but the width remained almost constant.
ARTICLE IN PRESS 538
H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
From the test, laser power was selected in the range 30–50 W, and scanning speed was selected from 30 to 50 mm/s at intervals of 10 mm/s.
3.3. Selection of factors From the preliminary experiment, preference is given to laser power, scanning speed, and machined pattern. The levels of selected factors are presented in Table 1. The number of lines implies total number of lines of the pattern. The effective factors (power, scanning speed, ratio of line gap, and number of lines) are selected. According to the result, the possible machining of the minimum width of V-groove is about 210 mm which is due to limitation of spot size depending on CO2 laser’s wavelength. Hence, gap between lines and maximum level of line selected for each V-groove do not overlap in the 100 mm 100 mm range.
3.4. Method of experiment
Fig. 6. Setup for DOE study.
The D.O.E uses orthogonal array, which ensures safety and reproducibility, minimizes size of the experiment, and enables to include many factors and levels. Hence, orthogonal arrays and experimental results are presented in Table 2, and selection of 4 factors and their 3 levels (L934) of orthogonal array were made. Each factor comprises of laser power (A), scanning speed (B), ratio of line gap (‘1:‘2), and (C), number of lines (D). Each factor’s level is shown in Table 2. Laser marking processed on PMMA was based
Fig. 7. Luminance distribution of laser machined light guide panels. (Sub captions represent: laser power, scanning speed, ratio of pattern gap, and number of line, respectively).
ARTICLE IN PRESS H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
on factors in Table 1. The area of PMMA was 100 mm 100 mm and thickness was 8 mm. To measure the luminance of each LGP, reflector sheet was attached to the bottom face, right side, and left side (Fig.6). Two 38 W CCFL (Cold Cathode Fluorescent Lamp) were attached to the two facing sides of the LGP. Diffuser sheet was laid on the front face. Fig. 7 and Table 2 show luminance and uniformity data measured from 25 points on LGP. Luminance was measured by BM-7 (Topcon). 3.5. Results
and {A2, B1, C3, D3}, respectively. In other words, with 40 W power, 30 mm/s scanning speed, 85 lines, and 100:50 ratio, high luminance was observed, while with 40 W power, 30 mm/s scanning speed, 90 lines, and 100:50 ratio, high uniformity was observed. Except for the number of lines, other factors had the same values. From the percentage contributions of four factors, two geometric factors, number of lines and ratio of line gap, have more effect on the quality of fabricated LGP than two laser marking factors, power and scanning speed. This shows that the advantage of proposed design tool to control the geometric parameters of LGP pattern. In detail, the number of lines has the
Table 3 shows SN ratios of luminance and uniformity. The respond tables for SN ratios of the luminance and uniformity are presented in Tables 4 and 5. Deviation implies difference between maximum value and minimum value of respond for the SN ratio. Percentage contribution implies percentage of total levels [12]. Therefore, luminance is most affected by factor D and uniformity by factor C. Table 2 shows that the experiment no. 4 provided the best result. Improvement in luminance was more than 4% compared with the prepared model, and uniformity was more than 8%. It is shown that the luminance and the uniformity are higher than the existing commercial LGP whose luminance is in the range of 1500–3000 cd/m2, and its uniformity is about 85%. From Figs. 8 and 9, optimal pattern conditions for high luminance and high uniformity are selected as {A2, B1, C3, D2} Fig. 8. Main effect plot for average SN ratio for high luminance. Table 3 SN ratio of luminance and uniformity. Exp. no. 1 2 3 4 5 6 7 8 9
SN ratio of luminance
SN ratio of uniformity
70.41 70.62 70.67 70.93 70.48 70.65 70.75 70.34 70.23
36.53 37.07 38.16 39.21 37.69 36.57 38.45 36.01 36.40
Table 4 Respond table for SN ratio of luminance. Factor
A B C D n
Fig. 9. Main effect plot for average SN ratio for high uniformity.
Effect 1
2
3
70.57 70.69 70.47 70.37
70.68 70.48 70.59 70.67
70.44 70.51 70.63 70.64
n
Deviation
PC (%)
0.24 0.21 0.16 0.30
26.4 23.1 17.6 33.0
Deviation
PCn (%)
0.87 1.14 1.73 0.92
18.7 24.5 37.1 19.7
PC: percentage contribution.
Table 5 Respond table for SN ratio of uniformity. Factor
A B C D n
Effect 1
2
3
37.25 38.06 36.27 36.87
37.82 36.92 37.56 37.36
36.95 37.04 38.10 37.79
PC: percentage contribution.
539
Fig. 10. An example of cross sections (50 W, 40 mm/s, 100:30, 90).
ARTICLE IN PRESS 540
H.-J. Kang et al. / Robotics and Computer-Integrated Manufacturing 26 (2010) 535–540
greatest effect on luminance, while ratio of line gap has the greatest effect on uniformity. One may expect that higher line density by 100:30 ratio improves uniformity than lower density by 100:50 ratio, but experimental result is contradictory. The possible reason for this is the numerous lines at the center resulting in the concentration of light at the center, and thus uniformity decreases when the ratio is 100:30. Eq. (3) shows the average luminance of SN ratio by the improved condition.
Although the proposed system provides a new idea for fast prototyping of LGP, it should be improved further to be used in production line for mass fabrication. On the other side, as a platform of IT services, the multi-agent system can be alternative of the web-based system to cover wider processes in design and manufacturing in flexible and rapid manner.
m ¼ A2 þB1 þ C 3 þ D2 3T
Acknowledgement
ð3Þ
T is an average of SN ratio representing level, and its value is 70.56. The average of SN ratio by improved condition is 70.99, and average of SN ratio by conditions (A1, B1, C3, and D3) is 70.42, so net improvement is 0.57. In the same way, the uniformity of SN ratio is expected to improve by 3.23 because T is 37.33, average SN ratio is 36.56 (by present conditions) and 39.79 (by improved conditions at A2, B1, C3, and D3). Fig. 10 shows a cross-section of LGP pattern by laser marking (based on orthogonal array). The power and scanning speed of the laser are important parameters in order to control the shape of V-groove and the optical property of LGP. From the experiment, the best condition for making V-groove with fine optical property is 40 W power and 30 mm/s scanning speed.
4. Conclusions The main advantage of web-based laser marking system for LGP is accessibility to the design tool and increase in the efficiency for generating an optimized pattern with fast prototyping process. This system enables a designer to understand the tendency of optical performance according to variation of each factor before conducting complicated optical analysis which demands high cost and expert knowledge. Effective factors for making LGP by laser and DOE for obtaining improved levels for the factors were investigated as follows: 1) Four factors namely, power, scanning speed, gap between pattern lines, and number of lines were selected, and levels were decided accordingly. 2) 40 W, 30 mm/s, and 100:50 ratio provided the highest luminance (with 85 lines) and the highest uniformity (with 90 lines). 3) The number of lines had the greatest effect on luminance, and the ratio of line gap had the greatest effect on uniformity. 4) By Taguchi method, optimized levels of each parameter were found, and luminance of 3523 cd/cm2 and uniformity of 92% were achieved using the laser machined BLU.
This work was supported by Grants from the second stage of Brain Korea 21 of Seoul National University, the Seoul R&BD Program (TR080578), and the KOSEF and the MEST Program (no. 20090081391). References [1] Kim GD, Kang HJ, Ahn SH, Song CK, Back CI, Lee CS. Laser-marking process for liquid-crystal display light guide panel. P I Mech Eng B—J Eng 2005;219: 565–9. [2] Dubey AK, Yadava V. Simultaneous optimization of multiple quality characteristics in laser beam cutting using Taguchi method. Int J Precis Eng Man. 2007;8:10–5. [3] Pinski EF, Witkowski JS, Abramski KM. Diffractive scanning mechanism for laser marker. Opt Laser Technol 2000;32:33–7. [4] Sriram RD, Szykman S, Durham D. Special issue on collaborative engineering. J Comput Inf Sci Eng 2006;6:93–5. [5] Shen W, Chao KM. Special issue on techniques to support collaborative engineering environments. Adv Eng Inf 2007;21:181. [6] Ahn S-H, Kim DS, Chu WS, Jun CS. MIMS: web-based micro machining service. Int J Comput Integ Manuf 2005;18:251–9. [7] Kim H-J, Chu WS, Ahn SH, Kim DS, Jun CS. Web-based design and manufacturing systems for micromachining: comparison of architecture and usability. Comput Appl Eng Educ 2006;14:169–77. [8] Chu CH, Cheng CY, Wu CW. Applications of the web-based collaborative visualization in distributed product development. Comput Ind 2006;57: 272–82. [9] Lin CS, Wu WZ, Lay YL, Chang MW. A digital image-based measurement system for a LCD backlight module. Opt Laser Technol 2001;33:499–505. [10] Adobes Flashs, www.adobe.com (2009). [11] Ahn SH, Sundararajan V, Smith C, Kannan B, D’Souza R, Sun G, et al. CyberCut: an internet-based CAD/CAM system. J Comput Inf Sci Eng 2001;1:52–9. [12] Chen YH, Tam SC, Chen WL, Zheng HY. Application of the Taguchi method in the optimization of laser micro-engraving of photomasks. Int J Mater Prod Technol 1996;11:333–44. [13] Phillip JR. Taguchi techniques for quality engineering: loss function, orthogonal experiments, parameter and tolerance design. New York: McGraw-Hill; 1996. [14] Liu ZQ, Feng Y, Yi XS. Coupling effects of the number of pulses, pulse repetition rate and fluence during laser PMMA ablation. Appl Surf Sci 2000;165:303–8. [15] Wang SC, Lee CY, Chen HP. Thermoplastic microchannel fabrication using carbon dioxide laser ablation. J Chromatogr A 2006;1111:252–7. [16] Zhou BH, Mahdavian SM. Experimental and theoretical analyses of cutting nonmetallic materials by low power CO2-laser. J Mater Process Technol 2004;146:188–92.