A Novel Driving Method for Fast Switching of a Cell for Display of Moving Pictures

Transcription

1 Japanese Journal of Applie Physics Vol. 43, No. 4A, 2004, pp #2004 The Japan Society of Applie Physics A Novel Driving Metho for Fast Switching of a Cell for Display of Moving Pictures Seo Hern LEE 1, Jin-Woo KANG 1, Gi-Dong LEE 1, Tae-Hoon YOON 1;2 an Jae Chang KIM 1;2 1 Department of Electronics Engineering, Pusan National University, Pusan , Korea 2 Research Institute of Computer, Information an Communication, Pusan National University, Pusan , Korea (Receive June 24, 2003; accepte December 12, 2003; publishe April 9, 2004) We propose a novel riving metho which is very efficient for the isplay of moving pictures on a cell. In the propose riving metho, each scan pulse is ivie into two parts for fast switching unlike the conventional active matrix riving metho that uses one scan pulse for each frame. By overriving or unerriving the cell using an appropriate ata voltage uring the first scan time an applying a esire ata voltage to the cell uring the secon scan time, we can achieve fast switching between all gray levels within the time between the first an secon scan pulse. [DOI: /JJAP ] KEYWORDS: fast switching, ouble scan metho, cell, overrive, unerrive, gray level, liqui crystal 1. Introuction Recently, there has been an increasing eman for isplaying moving pictures using thin-film-transistor liqui crystal isplays (TFT-LCDs). Fast response is the most important prerequisite for the isplay of moving pictures using the TFT-LCDs. The response times of conventional twiste nematic (TN) an supertwiste nematic (STN) liqui-crystal materials use in most irect-view LCDs are too slow to support them. To obtain fast response time, various attempts such as the metho using low cell-gap TN LCD with high optical anisotropy an low viscosity, 1) ferroelectric liqui crystal evices (FLCDs) with spontaneous polarization, 2) an the new riving scheme base on overrive aressing 3 7) have been propose. However, the metho using low cell-gap TN gives poor yiel uring the practical manufacturing process. The metho using FLCDs has some ifficulties in achieving reliable an goo quality alignment without efects. 8) It is ifficult to obtain a uniform alignment of the liqui crystal molecules ue to the occurrence of alignment-structural efects such as the socalle zig-zag efects that egrae memory capability an contrast ratio. Since the riving scheme using overirve aressing is simple an also easy for improving slow switching spee between gray levels, it has been wiely use. However, because the overrive aressing also requires appropriate low cell gap, yiel shoul be consiere, an obtaining high spee is limite. For a cell, it is known that the response time takes less than 3 ms, which is the fastest switching spee among the LC moes using nematic liqui crystals. Therefore, the cell is consiere as the best LC moe for isplaying moving pictures. However, in orer to isplay moving pictures, the switching spee between gray levels shoul be also as fast as the response time. The switching spee between gray levels of a cell is slow like that of the other LC moes. In this paper, in orer to isplay moving pictures, we propose a novel riving metho which is very efficient for the fast switching of a cell. Unlike conventional TFT-LCDs that use one scan pulse for each frame, the propose riving metho ivies the scan pulse into two parts for fast switching. By applying the propose riving metho to a cell, we can achieve a fast switching spee between all gray levels even in a cell with a large cell gap. aress: jckim@pusan.ac.kr 1416 T s T SD Scan line (V S ) T s +V D 1 frame -V D Data line (V D ) Fig. 1. Scan pulse of the propose riving metho. T s is etermine by the resolution an T SD is the time between the first an secon scan pulses. 2. New Driving Metho Figure 1 shows a scan pulse scheme of the propose ouble-scan riving metho. Each scan pulse is ivie into two parts an compose of the first an secon scan pulses. Data voltage is applie uring the write time, T s, which is etermine by the resolution of the isplay panel. During the first scan time, we apply a higher voltage (overriving) or a lower voltage (unerriving) than the ata voltage for a esire gray level, an then uring the seconas a result, we can achieve fast switching between all gray levels ue to the overrive or unerrive uring the first scan time. scan time, we apply a ata voltage corresponing to the esire gray level. As a result, we can achieve fast switching between all gray levels ue to the overrive or unerrive uring the first scan time. 2.1 Determination of the time between the first an secon scan pulse In this ouble-scan metho, it is important to etermine the time between the first an secon scan time, T SD, since the switching spee between all gray levels shoul be achieve within T SD. T SD epens on the liqui crystal an cell gap. Uner the assumption of normally black moe, T SD represents the time for a cell to switch from full white level to full black one. Therefore, in orer to etermine T SD,

2 Jpn. J. Appl. Phys., Vol. 43, No. 4A (2004) S. H. LEE et al V (a) (a) (b) Fig. 2. (a) Triangular-wave form at a frequency of 5 Hz with a bias voltage of 2.5 V, (b) voltage epenence of transmittance in a cell. we have to investigate the electrooptic characteristics of the cell. In this work, we fabricate a cell with a cell gap of 8.9 mm to show that by using the propose riving metho, fast switching can be achieve between all gray levels even in a large cell gap. The liqui crystal mixture ZLI-1557 (Merck) was use. The electroe area of the fabricate cell was 1 1cm 2. The alignment layer was coate on the bottom an top glass substrates by spin coating SE-3140 (Nissan Chemicals Co.), an rubbing was carrie out in the parallel irection. The pretilt angle generate by rubbing was approximately 5. The rubbing irections of the cell were set at 45 between the parallel polarizers. In orer to operate the cell as a isplay evice, the initial splay state was change to a ben state by applying a square wave of 1 khz with a voltage of 12 V rms. 9 12) Figure 2(b) shows the measure transmittance versus voltage characteristics of the cell riven by a triangular wave of 5 Hz with a bias voltage of 2.5 V as shown in Fig. 2(a). A bias voltage of 2.5 V was applie to prevent the ben state from returning to the splay state. From Fig. 2(b), we selecte 2.5 V for gray level zero (full black) an 10 V for final gray level (full white). Figure 3 shows the measure response characteristics of the cell. As shown in Fig. 2(b), in orer to achieve the transition from full white level to full black one, we shoul apply 2.5 V to the cell. However, in the case of such switching, as shown in Fig. 3(a), the transition time is as slow as about 9.3 ms. Therefore, in orer to achieve fast transition from full white level to full black one, we applie 0 V instea of 2.5 V, which makes the slope of transmittance curve steeper. As a result, we fin that the switching time of the cell from full white level to full black one becomes as faster as 4 ms. In practice, in transition from full white level to full black one, the falling time becomes the fastest when we apply 0 V. In the test cell with a cell gap of 8.9 mm, the switching time from full white level to full black one was (b) Fig. 3. Response characteristics of the transmittance with the applie voltage for the cell: (a) Response characteristics for the applie voltage changes from 10 V to 2.5 V an 0 V, (b) enlarge response characteristics for measuring the response time, where the rising an falling times are 0.6 ms an 2.4 ms, respectively. measure as 4 ms. If the cell gap of a test cell is below 8.9 mm, we can expect that T SD can be shorter than 4 ms. However, we have to be careful in applying 0 V to the cell because the ben state may return to a splay state uner a critical voltage. The application of 0 V for 4 ms oes not influence the transition from a ben state to a splay state. As a result, by applying 0 V to the cell, a faster switching time of 4 ms can be achieve for the transition from full white level to full black one, an the ben state is still maintaine. For the cell, the rising an falling times were measure to be 0.6 an 2.4 ms, respectively, as shown in Figs. 3(a) an 3(b). As a result, if we choose T SD to be 4 ms, the switching time from a higher gray level to a lower one may be achieve within 4 ms by unerriving the cell uring the first scan pulse. The switching time from a lower gray level to a higher one may be sufficiently fast within 4 ms since the transition is achieve by an electric fiel. Hence, for the transition between any two gray levels, we can achieve faster switching within 4 ms by overriving or unerriving the cell uring the first scan time. 2.2 Switching between a higher gray level an a lower one Uner the assumption of normally black moe an for a switching from a higher gray level to a lower one, uring the

3 1418 Jpn. J. Appl. Phys., Vol. 43, No. 4A (2004) S. H. LEE et al. 30 P Transmittance(%) T SD Q From 8.0 V to 0.0 V From 8.0 V to 2.5 V From 8.0 V to 2.0 V Time(ms) Fig. 4. Response characteristics of a cell with a gap of 5 mm simulate for the applicability of the ouble scan metho. Fig. 5. Response characteristics of the cell for the transitions from gray level zero to higher gray levels riven by the conventional riving metho. first scan time, we apply a lower voltage (unerriving) than the ata voltage for a esire gray level. An then, uring the secon scan time, we apply the ata voltage corresponing to the esire gray level. Inevitably, a ata voltage by which a esire gray level can be achieve after T SD is require. Figure 4 shows an example of fast switching using the propose ouble scan metho. LCD Master/2D (Shintech, Inc.) was use to calculate the optical properties. Since the simulation program for a cell has not been evelope yet, a homogeneous cell was use for the simulation. The liqui crystal mixture ZLI-1557 was use. At the wavelength of 550 nm, we esigne the retaration of the LC layer to become =2. The optic axis of the homogeneous cell was set at 45 between the parallel polarizers, which enotes a normally black moe. As shown in Fig. 4, in the switch-off state, the transition time from full white level at 8 V to full black one at 0 V shows the fastest response time, which takes about 6 ms. Therefore, the T SD of the homogeneous cell becomes 6 ms. Now we will show a metho by which the homogeneous cell can switch from full white state to the gray level at 2.5 V within the T SD. As shown in Fig. 4, in the case that a ata voltage of 2.5 V is applie to the cell which is in full white state uring the first scan time, we can fin that it takes more than 15 ms to reach the gray level at 2.5 V by the conventional active-matrix riving metho. In the propose ouble-scan metho, in orer to switch from full white state to the gray level at 2.5 V, the homogeneous cell is unerrive by a ata voltage of about 2 V uring the first scan time. As shown in Fig. 4, the transmittance rops markely in comparison with that obtaine by applying a voltage of 2.5 V. The transmittance rops to the point Q corresponing to the gray level at 2.5 V after approximately 6 ms, which is T SD. Therefore, if we unerrive the cell with about 2 V uring the first scan time, an after T SD, an apply 2.5 V uring the secon scan time, we can achieve fast switching within the T SD. In the case of switching from a lower level to a higher one, uring the first scan time, we apply a higher voltage (overriving) than the ata voltage for a esire gray level. An then, uring the secon scan time, we apply the ata voltage corresponing to the esire gray level. The switching time from a lower gray level to a higher one may be sufficiently fast since the transition is controlle by an electric fiel. 3. Experiments Before we test the propose riving metho, by using the cell fabricate to etermine T SD, first, we teste 8 gray levels of the cell by the conventional active-matrix riving metho for 2 frames, which is shown in Fig. 5. Assuming a frame rate of 60 Hz an a resolution of VGA, the scan pulse with (gate-on time) an frame time are about 34 ms an 16.7 ms, respectively. As shown in Fig. 5, the switching time from gray level zero (full black) to gray level 7 (full white) is the fastest. Hence, in the transition from gray level zero to the esire higher gray level, if the transition level is lower, the switching spee becomes slower. Likewise, the switching time from gray level 7 to gray level zero is the fastest. We fin that in the transition from gray level 7 to the esire lower gray level, if the transition level is lower, the switching spee also becomes slower. Before we apply the propose new riving metho to the cell using T SD obtaine by experiment in the above, we shoul investigate that the ivie scan pulse with of 17 ms is sufficient for the cell to be charge perfectly to the level of the applie ata voltage. Figure 6 shows the equivalent circuit of nematic LC. In Fig. 6, C st is the capacitance ue to the storage capacitor an C LC is the capacitance ue to the TFT V D Fig. 6. V pix i 1 i 2 i 3 C st C LC R LC Liqui Crystal Equivalent circuit of nematic liqui crystal cell.

4 Jpn. J. Appl. Phys., Vol. 43, No. 4A (2004) S. H. LEE et al inuce polarization. The charging current uring the switch-on time, i (on), which flows through the TFT is expresse as i (on) ¼ K½2ðV GS V th Þ V DS ŠV DS ; ð1þ where K is the current capability of TFT, V GS is the gate-tosource voltage, V DS is the rain-to-source voltage, an V th is the threshol voltage of the TFT. The LC currents, i 1, i 2 an i 3, are given by i 1 ðtþ ¼ V DðtÞ ; ð2þ R LC i 2 ðtþ ¼C LC t V DðtÞ; ð3þ i 3 ðtþ ¼C st t V DðtÞ: ð4þ The charging current, i (on), is equal to the sum of LC currents, i 1, i 2 an i 3. i ðtþ ¼i 1 ðtþþi 2 ðtþþi 3 ðtþ: ð5þ Using these equations an after calculations, we obtain the voltage that is applie to the LC layer uring the scan time as t V 1 DðtÞ ¼ ðc LC þ C st Þ ðkv D 2 ð1=r LC þ 2KðV G V th ÞÞV D þ 2KðV G V D V th V D V 2 D =2ÞÞ: ð6þ Figure 7 shows pixel voltage as a function of scan time, where the scan voltage, V G ¼ 20 V, the ata voltage, V D ¼ 6V, V th ¼ 2V, C st ¼ 0:5pF, R LC ¼ 2:6 T, C pix ¼ 1:5pF, respectively, K ¼ 0:25 ma/v 2 for poly-si TFT, an K ¼ 0:02 ma/v 2 for a-si TFT. As shown in Fig. 7, if the scan pulse with is shorter than 17 ms in the case of a-si TFT, the pixel voltage oes not reach the applie ata voltage. However, if we use poly-si TFT with the current capability of 0.25 ma/v 2, we fin that the ouble-scan metho can be applie even in the resolution of Ultra XGA. Figure 8 shows the response characteristics for 8 gray levels in which we apply the propose new riving metho to the cell using T SD of 4 ms obtaine by experiment. As shown in Figs. 8(a) an 8(b), the switching time between all intergray levels is achieve within T SD of 4 ms in the reverse Fig. 8. Response characteristics of the 8 gray levels obtaine by the propose new riving metho with T SD of 4 ms. transition as well as the transition from gray level zero to a higher one. Table I shows overriving ata voltages uring the first scan time an the esire ata voltages uring the secon scan time for the transition from gray level zero to a higher one. From Fig. 9, we also fin that the transition between gray level 4 an a higher gray level is achieve within T SD of 4 ms. Tables II an III show overriving ata voltages uring the first scan time an the esire ata voltages uring the secon scan time for the transition from gray level 4 to a higher gray level an the reverse process, respectively. Table I. Data voltages applie uring the first an secon scan write times for the transition from gray level zero to higher gray levels. (First scan time) (Secon scan time) 0! ! ! ! ! ! ! Pixel voltage(v) K=0.25µA/V 2 K=0.02µA/V x x x x x x10-5 Scan time(s) Fig. 7. Pixel voltage epenence on scan time calculate from the equivalent circuit. K ¼ 0:25 ma/v 2 an K ¼ 0:02 ma/v 2 are the values for the poly-si TFT an a-si TFT, respectively. Fig. 9. Response characteristics for the transitions from gray level 4 to higher gray levels. All switching are also achieve within T SD of 4 ms.

5 1420 Jpn. J. Appl. Phys., Vol. 43, No. 4A (2004) S. H. LEE et al. Table II. Data voltages applie uring the first an secon scan write times for the transition from gray level 4 to higher gray levels. (First scan time) (Secon scan time) 4! ! ! Table III. Data voltages applie uring the first an secon scan write times for the transition from higher gray levels to gray level 4. (First scan time) (Secon scan time) 7! ! ! Conclusion In orer to realize moving pictures, we propose a novel riving metho which can be applie to the fast switching of a cell. The propose riving metho ivies the scan pulse into two parts unlike the conventional active matrix riving metho that uses one scan pulse for each frame. In orer to use the propose riving metho effectively, we nee to select a T SD that is as short as possible. In a cell, we select T SD to be the time uring which the cell oes not return to the splay state from the ben state an maintains the ben state by the applie voltage of 0 V. In the case of a cell with a gap of 8.9 mm use in the experiments, we obtaine T SD of 4 ms. If the cell gap is lower, it is expecte that T SD can be less than 4 ms. Therefore, the switching time between all gray levels can be less than 4 ms. By applying the propose riving metho to a cell, a fast switching time can be achieve between all intergray levels. Acknowlegment This work was performe uner the Avance Backbone IT Technology Development Project supporte by the Ministry of Information & Communication of the Republic of Korea an partly fune by the Pusan National University Supporting Organization. 1) H. Ichinose, S. Nakajima, Y. Sugiyama, H. Numata, S. Naemura an A. Manabe: IDW 00 Dig., 2000, p ) N. A. Clark an S. T. Lagerwall: Appl. Phys. Lett. 36 (1980) ) H. Nakamura, J. Crain an K. Sekiya: IDW 00 Dig., 2000, p ) H. Nakamura: Jpn. J. Appl. Phys. 40 (2001) ) B.-W. Lee, C. W. Park, S. I. Kim, M. B. Jeon, J. Heo, D. Sagong, J. S. Kim an J. H. Souk: SID 01 Dig., 2001, p ) K. Kawabe an T. Furuhashi: SID 01 Dig., 2001, p ) H. Okumura, M. Baba, K. Taira an A. Kinno: SID 02 Dig., 2002, p ) D. C. Ulrich an S. J. Elston: Ferroelectrics 178 (1996) ) T. Miyashita, Y. Yamaguchi an T. Uchia: Jpn. J. Appl. Phys. 34 (1995) L ) C.-L. Kuo, T. Miyashita, M. Suzuki an T. Uchia: Appl. Phys. Lett. 68 (1996) ) M. Noguchi an H. Nakamura: SID 97 Dig., 1997, p ) H. Nakamura an M. Noguchi: Jpn. J. Appl. Phys. 39 (2000) 6368.