Chapter 5 Discussions on Issues of LEDs Backlight
5.2 Issues of the 4-in-1 (RGGB) LEDs Backlight
5.2 Issues of the 4-in-1 (RGGB) LEDs Backlight
For the different driving methods of the backlight, the deviations of spectral distribution and chromaticity coordinate can be observed.
The results of the hold type backlight are different from the superposition of the R, G and B field in the positions of the spectral peak, maximum spectral radiance (SR) and brightness. The hold type backlight is composed of R, G and B field where the LEDs are always turned on. Compared with the superposition of the ideal R, G and B field, the hold type backlight has different spectral distributions, chromaticity coordinates and brightness. The results are shown in Table5-1. Fig. 5-1 (a) shows spectral distribution for the hold type backlight. Fig. 5-1 (b) shows superposition of
- 45 -
the spectral distribution for the ideal R, G and B field where the duty cycle of the LEDs is 100%.
Table 5-1 Results of the peak wavelength, chromaticity coordinates and brightness where the duty cycle of the LEDs is 100%
Driving type Peak (nm) Max SR L (nits)
R 650 0.9285
G 532 0.4296 15896
Hold type
B 462 0.6266
R 639 1.3524
G 530 0.4686 17735
Superposition duty cycle = 100%
B 459 0.7718
Fig. 5-1 (a) Spectral distribution for the hold type backlight
- 46 -
Fig. 5-1 (b) Superposition of the spectral distribution for the ideal R, G and B field (duty cycle = 100%)
On the other hand, the characteristics of the spectral distribution of the FSC backlight are similar to the superposition of the ideal R, G and B field. The FSC backlight is composed of R, G and B field where the LEDs use PWM method. The duty cycle of the R, G and B LEDs is 50 % in the sub-frame, in other words, the duty cycle of the R, G and B LEDs is 1/6 in the frame (e.g., ON-time of the LEDs is 2.78ms approximately if the frame rate is operated at 60Hz). Compared with the superposition of the ideal R, G and B field where the duty cycle of the LEDs is 1/6, the FSC backlight has the same spectral distributions, chromaticity coordinates and brightness. The results are shown in Table 5-2. Fig. 5-2 (a) shows spectral distribution for the FSC backlight, and Fig. 5-2 (b) shows superposition of the spectral distribution for the ideal R, G and B field where the duty cycle of the LEDs is 1/6.
- 47 -
Table 5-2 Results of the peak wavelength, chromaticity coordinates and brightness where the duty cycle of the LEDs is 1/6
Driving type Peak (nm) Max SR L (nits)
R 637 0.2292
G 529 0.0815 2934
FSC (60Hz)
B 458 0.1300
R 635 0.2347
G 529 0.0812 2934
Superposition duty cycle = 1/6
B 458 0.1320
Fig. 5-2 (a) Spectral distribution for the FSC backlight
- 48 -
Fig. 5-2 (b) Superposition of the spectral distribution for the ideal R, G and B field (duty cycle = 1/6)
The hold type backlight is 5.4 times the brightness of the FSC backlight.
However, the ideal value is 6 times. The spectral distributions of the hold type backlight are different from the FSC backlight, as shown in Table 5-3.
Table 5-3 Results of the peak wavelength, chromaticity coordinates and brightness for the different driving method of the backlight
Peak wavelength (nm) Chromaticity coordinates Driving
type R G B x y
Brightness (nits)
Hold type 650 532 462 0.2932 0.3125 15896
FSC (60Hz) 637 529 458 0.3314 0.3083 2934
- 49 -
5.3 Discussion for the Issues of the 4-in-1 (RGGB) LEDs Backlight
One of reasons for the issues of the 4-in-1 (RGGB) LEDs backlight is that the junction temperatures increase in the LEDs chip. The peak wavelength, spectral width, and the output power of the LEDs strongly depend on temperature. The changes of the junction temperature are attributable to the ambient temperature, power consumption of the LEDs, and amount of heat sinking material in and around the LED. In this experiment, the ambient temperature and amount of heat sinking material are constant. The power consumption of the LEDs is a variable.
The heat Q dissipated by an LED is approximately equal to the power consumption, as shown in equation (5.1) [17].
Q≈VF ID (5.1) where VF is the LED forward voltage, I is the driving current, and D is the PWM duty factor. The temperature difference ΔTs-j between the LED package slug and the LED junction can be gotten, as shown in equation (5.2).
j
Therefore, according to the equation (5.1) (5.2), ΔTs-j can be accounted for the junction temperature of the LEDs, as shown in equation (5.3).
j
where Ts is the slug temperature. The Fig. 5-3 shows the relation of the junction temperature, thermal resistance and slug temperature in the LEDs chip.
- 50 -
Fig. 5-3 Relation of the junction temperature, thermal resistance and slug temperature in the LEDs chip
For the hold type backlight, the duty cycle of the LEDs is 100% because the LEDs are always turned on. On the other hand, the FSC backlight utilizes PWM method which has the duty cycle smaller than 100%. Based on the equation (5.1) ~ (5.3), the hold type backlight has larger power consumption than the FSC backlight, and higher junction temperature accordingly.
S. Chhajed and Y. Xi proposed extensive discussions to verify that wavelength shifts of LEDs could be estimated accurately by the corresponding junction temperature [18]. Fig. 5-4 illustrates the peak wavelength as a function of junction temperature for the red, green and blue LEDs. Note that the peak wavelength of the red, green and blue LEDs will be shifted with increasing temperature. The wavelength shift of the red LEDs is larger than blue and green LEDs.
- 51 -
Fig. 5-4 Peak wavelength as a function of junction temperature for the red, green and blue LEDs
Moreover, the light output degradation of the LEDs is also determined by the junction temperature. Higher temperature results in reduced light output [19]. In general, the temperature of the semiconducting element increases in the warmer environment and at higher current. Fig. 5-5 shows the light output of the LEDs at the
- 52 -
constant current is a function of its junction temperature. The temperature dependence of InGaN LEDs (e.g., blue, green and white) is much less than one of AlGaInP LEDs (e.g., red and yellow). Therefore, Red and yellow AlGaInP LEDs have larger wavelength shift than blue, green and white InGaN LEDs.
Fig. 5-5 Relative light output of the red, blue and phosphor-converted white LEDs as a function of the junction temperature
The high junction temperature reduces the maximum spectral radiance and brightness. In addition, the spectral distribution and chromaticity coordinates are shifted by the high junction temperature at the same time.
For the hold type backlight, the junction temperatures of the LEDs are rapidly increased because the LEDs are always turned on. The efficiency of the light output and spectral distribution are changed due to the high junction temperature. The reasons for the deviations of the chromaticity coordinate could be the decay of the light output and the wavelength shift. On the other hand, the PWM method is used in the FSC backlight. The junction temperature can be maintained the constant temperature. In such condition, the spectral distribution and light output can be
- 53 -
maintained the original values. Moreover, the chromaticity coordinates are unchanged.
5.4 Summary
For the FSC backlight, the optical features are almost similar to the superposition of the ideal R, G and B field where duty cycle of the LEDs is 1/6.
However, the optical features of the hold type backlight are different from the superposition of the ideal R, G and B field where the LEDs are always turned on.
For the different driving type backlight (hold type backlight and FSC backlight), the deviation of the peak wavelength of R LEDs is the most obvious. The deviation of the peak wavelength of R LEDs is 13nm. In addition, the deviation values of the chromaticity coordinates are 0.0042~0.0382.
One explanation for the issue of the RGB LED backlight is that the junction temperatures increase in the LEDs chip. The peak wavelength, spectral width, and the output power of the LEDs strongly depend on temperature. The high junction temperature reduces the maximum spectral radiance and brightness of the backlight.
Moreover, the spectral distribution and chromaticity coordinates are shifted by the high junction temperature. For the LED applications, heat dissipation is critical issue, which can be alleviated by PWM method.
- 54 -
Chapter 6
Conclusions ＆ Future Works
6.1 Conclusions
With development of education, communication and entertainment in human daily life, LCDs become an important display technology. High brightness, resolution and excellent color rendering are the major concerns. Although several configurations of the hold-type LCDs have been proposed, many issues such as motion blur, optical efficiency and poor color representation have large space to improve. In this investigation, the scanning FSC LCD is proposed to overcome these defects.
The scanning FSC LCD has potential to serve as the new approach in terms of offering better image quality. It can efficiently improve the fuzzy edge of the moving picture, provide higher color gamut without color filter less and higher optical efficiency.
The aim of this thesis is to accomplish the control circuit of the large scale scanning FSC backlight system. The backlight circuit is composed of the hardware and the software. The hardware consists of the control board, LEDs light bars, LED driver IC and backlight module. Then, the software with C programming language is used to control signals of the backlight circuit.
The optical performances of the BLM are measured by the CCD camera, Conoscope and spectrometer. The uniformities of the different color states are 84% to 89%. The light leakage ratio from the operating block to the neighboring divisions is suppressed down to 11.86 %. Compared with the dynamic driving method of the
- 55 -
backlight (FSC backlight and scanning FSC backlight), the spectral distributions of the hold type has an obviously deviation. The deviations of the peak wavelength are 4nm, 3nm and 13nm at the R, G and B field, respectively. The deviation of the peak wavelength of the R field is the most obvious. In addition, the deviations of the chromaticity coordinates are 0.0042~0.0382.
The oscilloscope is utilized to measure the electronic characterizations of the control circuit. The operation of the proposed FSC driving scheme was verified by electronic test. The results are identical with the ideal signals of the scanning FSC backlight.
For the different frame rate of the FSC backlight, the maximum deviations of the brightness are 100 nits, and the deviations of the chromaticity coordinates are 0.0005 to 0.0032, and then the maximum color difference is 1.0722. The deviations are insignificant in FSC driving scheme.
For the FSC backlight, the optical features are almost similar to the superposition of the ideal R, G and B field where duty cycle of the LEDs is 1/6.
However, the optical features of the hold type backlight are different from the superposition of the ideal R, G and B field where the LEDs are always turned on.
Based on the equations (5.1) ~ (5.3), the hold type backlight has larger power consumption than the FSC backlight, and higher junction temperature accordingly.
The high junction temperature reduces the maximum spectral radiance and brightness.
In addition, the spectral distribution and chromaticity coordinates are shifted by the high junction temperature at the same time.
For the LED applications, heat dissipation is critical issue, which can be alleviated by PWM method.
- 56 -
6.2 Future Works
In the future, a backlight circuit for feedback control of the FSC LCD will be achieved to overcome the deviation of the chromaticity coordinates. There are two kinds of the modulation which can solve the deviation of the chromaticity coordinates.
One is the modulation of the driving current; the other is the modulation of the LED duty cycle.
The backlight system of the feedback control is shown in Fig. 6-1. The light output of the LEDs is measured by the color sensor, and the junction temperature is measured by the temperature sensor. The feedback values combine sensors with A/D (analog to digital) ports of the control board. The values of the modulation are based on the feedback values. Finally, the feedback values are gotten to determine the modulation of the driving current or LED duty cycle.
Fig. 6-1 Backlight system of the feedback control
In order to improve the contrast ratio and save the power consumption, the 2D scanning FSC will be proposed. The driving principles of the 2D scanning FSC
- 57 -
backlight are similar to the scanning FSC backlight. Besides, the backlight combines the local dimming technology, as shown in Fig. 6-2.
Fig. 6-2 2D scanning FSC backlight
- 58 -
Reference
[1] N. Ogawa, T. Miyashita, and T. Uchida, “Field-Sequential-Color LCD Using Switched Organic EL backlight,” SID’99 Digest, pp.1098 (1999).
[2] Norio Koma, Tetsuya Miyashita, Tatsuo Uchida, and Nobuhiro Mitani,” Color Field Sequential LCD Using an OCB-TFT-LCD,” SID’00 Digest, pp.632 (2000).
[3] T. Yoshihara, T. Makino, and H. Inoue, “A 254-ppi Full-color Video Rate TFT-LCD Based on Field Sequential Color and FLC Display ,” SID’00 Digest, pp.1176 (2000).
[4] Fumiaki Yamada, Hajime Nakamura, Yoshitami Sakaguchi, and Yoichi Taira,
“Sequential-Color LCD Based on OCB with an LED Backlight,” Journal of the SID, vol.10, no.1, pp.81-85 (2002).
[5] M. Mori, “Mechanism of Color Breakup on Field-Sequential Color Projectors,”
SID’99 Digest, pp.350-353 (1999).
[6] M. Ogata, K. Ukai, and T. Kawai, “Visual fatigue in congenital nystagmus caused by viewing images of color sequential projectors,” IEEE’05, pp. 314-320 (2005).
[7] J. B. Eichenlaub, “Develop and preliminary evaluation of field-sequential color LCD free of color breakup,” SID’94 Digest, pp.293-296 (1994).
[8] O. Wada, J. Nakamura, K. Ishikawa, and T. Hatada, “Analysis of Color Breakup in Filed-Se1uential Color Projection System for Large Area Display,” IDW’99, pp.993-996 (1999).
[9] J. Lee, T. Jun, J. Lee, J. Han, and J. H. Souk, “Noble measurement method for
- 59 -
color breakup artifact in FPDs,” IMID/IDMC’06, pp.92-97 (2006).
[10] Fang Jin Yoo, Jong Hoon Woo, Hyun Ho Shin and Chang Ryong Seo, “Side Light Type Field Sequential Color LCD Using Divided Light Guide Plates,”
IDRC 03, pp.180 (2003).
[11] K. Sekiya, T. Kishimoto, K. Wako, S. Nakano, H. Ishigami, K. Käläntär, K.
Shimabukuro, D. Kunioka, T. Miyashita, and T. Uchida, “Spatio-Temporal Scanning LED Backlight for Large Size Field Sequential Color LCD,”
IDW’05, pp.1261-1264 (2005).
[12] K. Käläntär, Tadashi Kishimoto, Kazuo Sekiya, Tesuya Miyashita and Tatsuo Uchida, “Spatio-Temporal Scanning Backlight for Color-Field Sequential Optically Compensated Bend Liquid-Crystal Display,” SID’05 Digest, pp.1316–1319 (2005).
[13] K. Käläntär, Tadashi Kishimoto, Kazuo Sekiya, Tesuya Miyashita and Tatsuo Uchida, “Spatio-Temporal Scanning Backlight Mode for Field-Sequential-Color Optically-Compensated-Bend Liquid-Crystal Display,”
Journal of the SID, Vol.14, pp.151-159 (2006).
[14] Ming-Chin Chien, Chung-Hao Tien, Cho-Chih Chen and Yen-Hsing Lu,“LED Light Lit for Field-Sequential-Color Backlight System,” SID’07 Digest, pp.441-444 (2007).
[15] Chung-Hao Tien, Yen-Hsing Lu, and Yuan-Jung Yao, “Tandem Light-Guides With Micro-Line-Prism Arrays for Field-Sequential-Color Scanning Backlight Module,” Journal of Display Technology, Vol.4, No.2, pp.147-152 (2008).
[16] M. D. Fairchild, “Color Appearance Models,” (2nd Edition).
[17] Kwong Man and Ian Ashdown, “Accurate Colorimetric Feedback for RGB LED Clusters,” 6th Conference on Solid State Lighting, Proceedings of SPIE,
- 60 -
Vol. 6337, 633702 (2006).
[18] S. Chhajed, Y. Xi, Y.-L. Li, Th. Gessmann and E. F. Schubert, “Influence of junction temperature on chromaticity and color-rendering properties of trichromatic white-light sources based on light-emitting diodes,” Journal of Applied Physics 97, 054506 (2005).
[19] “LED Lighting System,” National Lighting Product Information Program, Vol.7, Issues 3 (2003).