201825 A6 B6 五、發明説明（1 ) 本發明偽關於液晶顯示裝置之駆動方法。 通常在驅動單純矩陣型液晶顯示裝置時.傈採用一種 電懕平均法。然而，實際上之液晶面板傺由電阻不等於 0之掃描及倍號電極所構成，而且液晶層産生介質之作 用。因此，利用習用之電壓平均法驅動時，其施加於掃 描電棰及信號電極交叉而形成之顯示點上之實效電 壓，依照液晶面板所顯示之文字及園侏之園型而發生變 化。結果使其顯示發生不均勻。 造種問題傺習知之問題。其對策有例如在特開昭62-31825 號，持開昭6 0 - 1 9 1 95號，恃開昭60-19196號中掲示之在 1幀之間，多次的反轉施加於液晶面板之電壓極性之方 法（以後稱線反轉驅動方法）。 另外又有本發明之發明人所申請專利之待願昭63-159914 號之顯示不均勻之改菩方法。 然而，上述線反轉驅動方法只是改善被液晶面板所挾 持之液晶之光學特性因施加霉®之頻率成份而發生變化 所造成之顯示之不均勻稍有功效而已，不能完全消除顯 {請先閲讀tt面之注意事項再填商本頁) 昭 願 待 之 請 申 所 人 明 。發 句之 均明 不發 之本 示 菩 改 之 號 尚 法但 方’ 法 正 修 壓 霣 稱 後 以 不 示 顯 之 下 如 有 勻 均 不 示 顯 善 〇 改菩 的改 箸得 顯獲 可能 然不 雖勻 —均 經濟部中央橾準局印装 之該 莕 。 改容 能内 不示 仍顯 法其 正及 修 1 壓板 笛面 用晶 利液 明示 説表 圓圖 4 4 11 11 β β 照 Ο 參匀 先均 下不 以示 顯 肀 4(210X 297父发） 201825 A6 B6 五、發明説明（2 ) 液晶面板1以一對基H2,3挾持液晶（未園示），在基片 2上形成横向之多個掃描電極Y1〜Y6,在基片3上縱向的 形成多個倍號電極XI〜乂6。掃描電極Y1〜Y6與倍號®極 XI〜X6分別交叉之位置，成為顯示點.該液晶面板1具 有6x6之顯示點，其理由僳為了說明之方便，通常有更 多之點。圖中斜線部份表示點亮之部份（以後，黏亮之 顯示點稱為點亮點，未點亮之顯示黏稱為未點亮點）， 鼷中表示棋盤_案之顳示内容。在其掃描霣極之左方， 施加由本發明之發明人所申請之待願昭63-159 9M號所 掲示之顯示不均勻改善方法中，利用毎間隔一條之顯示 之電壓修正法之依照顯示圖型而變化之掃描霜®波形。 換言之，從選擇某一掃描電極變更成蓮擇另一掃描電極 時，配合某一掃描霣棰上之點亮點數目與S —被選擇之 掃描霣極上之點亮點之數目差工，在非選擇電壓上重璺 修正霣壓。但中所示之顯示内容，因為其差值工經常 為0 ,故非選擇電壓上未施加有修正霄壓。在倍號霣極 上，毎隔一條交替的從上下方施加佶號電壓波。此時，液 晶面板1偽在施加於顯示點之實效電壓增大時變成黑 色而進行所諝之玉片顳示。 寘際上顯示該圔中之顯示圖型時.倍號笛極XI, X3,X5 所形成之顳示點中，愈位於上方之顯示點愈淡，愈位於 下方之點愈黑。相反的，倍號《極X2,X4,X6所形成之顯 (請先KI讀背面之注意事項再填寫本頁) • JTT. .線.201825 A6 B6 V. Description of the invention (1) The present invention is about the method of moving the liquid crystal display device. Usually, when driving a simple matrix type liquid crystal display device, an electric averaging method is used. However, the actual liquid crystal panel is composed of scanning and multiplying electrodes whose resistance is not equal to 0, and the liquid crystal layer generates a medium. Therefore, when the conventional voltage averaging method is used for driving, the effective voltage applied to the display point formed by the crossing of the scanning electrode and the signal electrode changes according to the characters displayed on the liquid crystal panel and the garden shape of the garden. As a result, the display is uneven. The problem of seeding is the problem of conventional knowledge. The countermeasures include, for example, Japanese Patent No. 62-31825, No. 6-0-1 9 1 95, and No. 60-19196, which are shown between 1 frame and multiple inversions are applied to the LCD panel. The method of voltage polarity (hereinafter referred to as line inversion driving method). In addition, there is a method for correcting the display unevenness of No. 63-159914, which is a patent application filed by the inventor of the present invention. However, the above-mentioned line inversion driving method only improves the display unevenness caused by changes in the frequency characteristics of the liquid crystal held by the liquid crystal panel due to the application of mildew®. It is slightly effective and cannot completely eliminate the display {please read ttThe precautions on the face and then fill in this page) Please let me know if you want to wait. If the sentence is clear, the original name of the Bodhisattva is not yet issued, but the method of the Buddha's law is correct. After the law is repaired, if there is no display, if there is evenness, it will not show the goodness. The reformation of the Bodhisattva may be obtained. Even though it is not uniform-the Central Plains Bureau of the Ministry of Economic Affairs should print it. If the capacity is not changed, it will still be displayed and corrected. 1 The flute surface of the pressure plate will be clearly expressed with crystal liquid. 4 4 11 11 β β According to the photos, the uniformity should be displayed first and then not displayed to show 4 (210X 297) ) 201825 A6 B6 V. Description of the invention (2) The liquid crystal panel 1 holds the liquid crystal (not shown) with a pair of substrates H2, 3, forming a plurality of lateral scanning electrodes Y1 ~ Y6 on the substrate 2 on the substrate 3 A plurality of double-numbered electrodes XI ~ Q6 are formed in the longitudinal direction. The positions where the scan electrodes Y1 ~ Y6 and the multiplex poles XI ~ X6 cross respectively become display points. The LCD panel 1 has 6x6 display points. The reason is that there are usually more points for the convenience of explanation. The oblique line in the figure indicates the lighted part (hereinafter, the sticky display point is called the lighted point, and the unlit display point is called the unlighted point), and the middle part shows the temporal content of the chessboard_case. To the left of its scanning pole, the display unevenness improvement method shown in the application No. 63-159 9M applied by the inventor of the present invention is applied. Scanning Frost® waveforms that vary in type. In other words, when changing from selecting one scan electrode to selecting another scan electrode, according to the difference between the number of lighted points on a certain scan electrode and the number of lighted points on the selected scan electrode, the non-selected voltage Heavy weight correction of pressure. However, since the display content shown in the figure is always zero, the correction voltage is not applied to the non-selected voltage. On the double pole, an alternating voltage wave is applied from above to below. At this time, the liquid crystal panel 1 becomes black when the effective voltage applied to the display point increases, and the temporal display of the jade piece is performed. When displaying the display pattern in this circle, the temporal display points formed by the horns XI, X3, X5 of the times, the more the display points above the lighter, the darker the points below. On the contrary, the double number "Pole X2, X4, X6 is formed by the display (please read the notes on the back of KI before filling in this page) • JTT ... line.
經濟部中央標準局印$L 愈 點 之 方 上 於 位 愈 淡 愈 -點 4 示 I 顯 之 方 下 於 位 愈 中 點 示 甲 4(210X 2971'发） 201825 A6 B6 經濟部中央橾準局印裝 五、發明説明（3 ) 黑。換言之，愈靠近施加倍號電壓波形之一側之顯示黏 ，其實際上施加之實效電壓愈小。 本發明之發明人經遇多次實驗及研究該顯示不均後， 發現以下之事實。 以下參照第15圖U)〜（c)説明此事實。 第15圖（a)〜（c)表示第14圖之液晶面板1之各波 形。第15鼷（a)中，以實線表示第14圖之顯示點D31之位 置上之佶號電極X3上之霣壓波形。以虛線表示顯示點D21 之位置之倍號電極X2上之電壓波形。圖中，為了説明之 方便，將實線波形輿虛線波形互相錯開，但實際上成 為一致。第15圚-(b)表示第14團之顯示點D21至D31之位 置之倍號電極XI上之電壓波形。第15園- (c)表示第14圈 之顯示點D31之位置之掃描電極Y1上之電壓波形及倍號 電極X3上之電壓波形之差值。換言之，以實線表示施加 於顯示點D31之霣壓波形。同樣的，第15圖（c)中虛線表 示施加於顯示點D21之堪壓波形。斜線部份表示施加於 點亮點與未點亮點之笛K之差，而非産生顯不均勻之電 K差。_中，70^1^2^3^4^5表示電壓。假設霄懕 V5,V3,V0,V4為第1組之點亮，未點亮，選擇非S擇霣 壓，笛壓V0,V2,V5,V1為第2組之點亮，未點亮，遘擇， 非選擇霣壓，将選擇，非選擇霣壓施加於掃描霣極，而 將點亮，未點亮霣壓施加於信號笛極。（以後，將施加 於掃描罨極之罨壓波形稱為掃描霣壓波形，施加於佶號 一 5 - {請先閱讀背面之注意事項再填寫本頁) •^, .訂. •漆.· f 4(210X 297 /：^) 201825 A6 B6 經滴部中央標準局印^ 五、發明説明（4 ) 電極之波形稱為倍號電壓波形）。第1舆第2霣壓组遇期 性的轉換。本例中，在全部掃描電極Y1〜Y6上施加選擇 電壓後轉換（此週期稱為1個幀，圖中以下1,下2表示）。 第15圖（a)中，因為黏D31與施加倍號電壓波形之端部 之矩離較短，故黏D31上之位置之信號電掻X3上幾乎無 電壓波形之衮減，施加之信號電壓波形完整的施加在上 面。然而，如第15圖- (b)所示，因為點D21與施加信號 電壓波形之端部之距離較長，故點D21位置之信號電極 X2上施加電壓波形發生顯箸之衮減或失真之信號電壓波 形。換言之，因為倍號電極XI〜X6所具有之電胆及以液 晶作為介質之電容量所形成之積分電路，使其電壓波形 産生産生衮減及失真。因此施加於信號電極XI,X3,X5之 點亮電S(未點亮霜壓）轉換成未點亮（點亮電壓）時，在 掃描電極Y1上産生之釘子狀雜訊，大於倍號電極X2,X4, X6之點亮電壓（未點亮電壓）轉換成非黏亮霣壓（點亮電 壓）時之雜訊。因此，信號電極XI,X3,X5之點亮電壓（未 點亮霄壓）轉換成未點亮電壓（點亮電®)而在掃描®極 Y1産生之釘子狀雜訊將成為控制性因素。因此，如第15 圖（c)中之實線波形所示，施加於顯示點D31之實效電壓 減小，而如虛線波形所示，施加於顯示點D21之實效霣 壓加大。 此時，産生於第14鼷之掃描霣極Y6上之雜訊，相反的 ，在倍號霣掻X2,X4,X6所産生之雜訊成為控制性因素， (請先閲請背面之注意事邛再填寫本頁) .蛑. -打_ •姝.The Ministry of Economic Affairs Central Bureau of Standards printed $ L. The more points above, the weaker the position becomes-the point 4 shows I. The places below the position show the middle point. A 4 (210X 2971 'issued) 201825 A6 B6 Central Ministry of Economic Affairs Printed V. Description of Invention (3) Black. In other words, the closer to the display stick on the side of the applied voltage waveform, the smaller the effective voltage actually applied. The inventor of the present invention has found the following facts after many experiments and studies on the display unevenness. This fact will be explained below with reference to Figure 15 U) ~ (c). Figures 15 (a) to (c) show the waveforms of the liquid crystal panel 1 of Figure 14. In the fifteenth circle (a), the solid pressure line shows the pressure waveform on the number electrode X3 at the position of the display point D31 in FIG. 14. The voltage waveform on the double-number electrode X2 showing the position of the point D21 is indicated by a dotted line. In the figure, for the convenience of explanation, the solid waveform and the dotted waveform are staggered from each other, but in fact they are the same. Fifteenth- (b) shows the voltage waveform on the double-numbered electrode XI at the positions of the display points D21 to D31 of the 14th group. 15th circle-(c) shows the difference between the voltage waveform on the scan electrode Y1 at the position of the display point D31 on the 14th circle and the voltage waveform on the multiplier electrode X3. In other words, the solid line indicates the waveform of the pressure applied to the display point D31. Similarly, the dotted line in Fig. 15 (c) shows the pressure waveform applied to the display point D21. The slanted line indicates the difference between the flute K applied to the lit point and the unlit point, rather than producing an uneven electric K difference. In _, 70 ^ 1 ^ 2 ^ 3 ^ 4 ^ 5 represents the voltage. Suppose Xiaoyi V5, V3, V0, V4 is the first group of lights, unlit, select non-S selective pressure, flute pressure V0, V2, V5, V1 is the second group of lights, unlit Select, non-selective pressure, apply the selected, non-selective pressure to the scanning pole, and light, unlit pressure to the signal dipole. (In the future, the waveform of the pressure applied to the scanning pole will be referred to as the scanning pressure waveform, and it will be applied to the number one 5-{please read the precautions on the back before filling this page) • ^, .Order. • Paint. · f 4 (210X 297 /: ^) 201825 A6 B6 Printed by the Central Standards Bureau of the Drip Department ^ V. Description of the invention (4) The waveform of the electrode is called the double-numbered voltage waveform). The 1st and 2nd Yuan pressure group encounters a periodical change. In this example, the selection voltage is applied to all the scan electrodes Y1 to Y6 and then converted (this period is called one frame, which is indicated by the following 1 and 2 in the figure). In Fig. 15 (a), the moment between the stick D31 and the end of the applied double voltage waveform is shorter, so the signal on the stick D31 has almost no voltage waveform reduction on the X3, and the applied signal voltage The waveform is completely applied on it. However, as shown in FIG. 15-(b), because the distance between the point D21 and the end of the applied signal voltage waveform is long, the applied voltage waveform on the signal electrode X2 at the point D21 occurs significant reduction or distortion. Signal voltage waveform. In other words, the integrated circuit formed by the electric bladder of the double-numbered electrodes XI ~ X6 and the capacitance of the liquid crystal as the medium causes the voltage waveform to be reduced and distorted. Therefore, when the lighting electric S (unlit frost voltage) applied to the signal electrodes XI, X3, X5 is converted into unlit (lighting voltage), the nail-shaped noise generated on the scan electrode Y1 is larger than the double-number electrode Noise when X2, X4, X6 lighting voltage (non-lighting voltage) is converted into non-stick bright voltage (lighting voltage). Therefore, the lighting voltage of the signal electrodes XI, X3, X5 (unlighted voltage) is converted into an unlighted voltage (lighting ®) and the nail-shaped noise generated at the scan® pole Y1 will become a controlling factor. Therefore, as shown by the solid-line waveform in FIG. 15 (c), the effective voltage applied to the display point D31 decreases, and as shown by the dotted-line waveform, the effective voltage applied to the display point D21 increases. At this time, the noise generated on the 14th scan electrode Y6, on the contrary, the noise generated by the X2, X4, X6 in the multiplier becomes the controlling factor, (please read the notes on the back first Qiong then fill out this page). 蛑.-打 _ • 姝.
I I -6 甲 4(210X 297 公; 201825 A6 B6 五'發明説明（5) (請先聞請背面之注意事項再填荈本頁) 施加於顯示黏D26之實效電壓減小，而施加於願示點D36 之實效電壓增大。 以下説明一般狀態。假設從上部算起第η鹤掃描電極 為掃描電極Υη,從左方算起第β桓之信號霣極為倍號電 極Χιη,倍號霉極X·舆掃描電極Υη交叉而形成之顯示點為 顯示點Dmn(以後除非另有説明，皆以此方式表示）。假 設從一端（第14圖中之上方）接受倍號電壓波形之信號電 棰之各信號電棰與掃描電極Yn所形成之顯示點，及輿掃 描電棰Υη + 1所形成之顯示點皆成為點亮點之數目為al ,兩者皆為未點亮點之數目為bl,而與掃描電極Yn所形 成之顯示點為點亮點，與捅描電極Υη + l所形成之顯示點 為未點亮點之數目為C1,與掃描電極Υη所形成之顯示點 為未點亮點而與掃描霣極Υη + 1所形成之顯示點為 點亮 點之數目為dl,從另一端（第14圖中之下方）接受信號電 壓波形之倍號笛棰之各倍號霣棰與掃描霣極Yn所形成之 顯示點，及與掃描霣極Υη + 1所形成之顯示點皆為點亮點 之數目為82,兩者皆為未點亮之數目為b2,與掃描II棰 Yn所形成之顯示點為點亮點而舆掃描霜極Yn + 1形成之顯 示點為未點亮點之數目為C2,輿拥描電極Υη所形成之顯 示點為未點亮點而舆拥描霣極Υη + 1所形成之顳示點為點 亮點之數目為d2。 從一端（第14圈中之上方）接受佶《霣壓波形之信《霣 經濟部中央搮準局印裂 亮1點之與 點S®未中極 中 r,4; 中與101號 點 Μ第信 示 為ί各 顯f’目端之 之of數一極 成N1之另電 形為點從號 所目亮，倍 γη數點的之 極之，樣形 m亮中同波 描點點。、 掃未示f壓 與，顯of電 極 Ο之Μ號 SH1成為信 號為形目受 倍目所數接 各數+之} 之之γη點方 極點極亮下 甲 4(210Χ 297 公发） 201825 A6 B6 五、發明説明（6 ) 掃 描 電 極 Yn 1所 形成 之 顯示點 中 * 點亮 點 之 數 目為N 2 ο η , 未 點 亮 之 數 S 為N2 off,與掃描電極Y η + 1 所 形成之顯示 點 之 中 > 點 亮 點之 數 目為M2 on » 未點 亮 點 之 數目為M2off 此 時 可 得 9 HI 0 IV = a 1 + c 1 HI Off = bl dl Ml 0 IV = a 1 f dl Ml 〇iF = bl + cl 假 設 9 數 值 11為 I 1 =c 1 - dl =Η 1 ο η - Mlon 同 樣 的 « H2ov =a 2 + c 2 Ν2ο f F =b2 + d2 Μ2ο n =a2 + d2 Μ2ο n =b2 + c 2 假 設 9 數 值 12 = c 2 - d2 = N2〇iv/ - M2o 另 外 • 設 定 函數 I (k)為 I (k) =f (k ) *11 + f (L -K) * 12 上 式 中 * f (k)為 函 數，俱 k增大時減小之函數。此函 數 f (k) 表 示 各 倍號 霣 極在揷 描 霣 棰上 産 生 之 釘子狀雜訊之 大 小 愈 接 近 接受 倍 號霣麼 波 形 之端 部 (以後稱為驅動 大 愈 訊 雜 其 極 S 描 掃 之 \»/ 端 {請先閲讀背面之注意事項再填寫本頁) •打. 經濟部中央橾準局印製II -6 A4 (210X 297; 201825 A6 B6 Five'Invention Description (5) (please read the precautions on the back before filling this page) The effective voltage applied to the display stick D26 is reduced, and applied to the wish The effective voltage of point D36 is increased. The following describes the general state. It is assumed that the η-th scan electrode from the top is the scan electrode Υη, and the β-th signal from the left is the double-digit signal electrode, the double-digit electrode. The display point formed by the intersection of the X and the scanning electrodes Υη is the display point Dmn (hereinafter, unless otherwise stated, they are all expressed in this way). It is assumed that the signal voltage of the double-numbered voltage waveform is received from one end (upper in FIG. 14) The display dots formed by the signal electrodes and the scan electrode Yn, and the display dots formed by the scan electrode Y + 1 + 1 are both the number of lit points is al, and the number of both unlit points is bl, and The display dot formed with the scan electrode Yn is a lit dot, the display dot formed with the trace electrode Υη + l is an unlit dot, the number is C1, and the display dot formed with the scan electrode Υη is an unlit dot and scanned The display point formed by 饣 极 Υη + 1 is lit The number is dl, from the other end (lower in Figure 14), the display points formed by the multiples of the whistle of the signal voltage waveform and the scanning pole Yn, and the scanning pole Υη + 1 The number of display points formed is 82, the number of unlit points is b2, and the number of unlit points is b2, and the display points formed by scanning II and Yn are lighted points and the display formed by scanning frost pole Yn + 1 The point is the number of unlit points is C2, the display point formed by the electrode Υη is the unlit point and the temporal display point formed by the electrode Υη + 1 is the number of lighted points is d2. From one end ( (Upper in lap 14) Accepted the letter of "Yuan Pressure Waveform" "The Central Bureau of Economics of the Ministry of Economic Affairs printed a bright and bright 1 point and the point S® is not in the middle pole of the middle r, 4; the middle and the 101th point M For each of the f 'eye ends, the number of poles is N1, and the other electric shape is the point of the number, which is brighter than the number of γη points. The shape of the m is bright. Unshown f pressure and, the SH of the electrode Ο, the M number SH1 becomes the signal, and the γη point of the square is extremely bright, and the bottom pole 4 (210Χ 297 Hair) 201825 A6 B6 V. Description of the invention (6) Among the display points formed by the scan electrode Yn 1 * The number of lighted points is N 2 ο η, the number of unlighted S is N2 off, and the scan electrode Y η + 1 Among the formed display points> The number of lit points is M2 on »The number of unlit points is M2off At this time, 9 HI 0 IV = a 1 + c 1 HI Off = bl dl Ml 0 IV = a 1 f dl Ml 〇iF = bl + cl Suppose 9 the value 11 is I 1 = c 1-dl = Η 1 ο η-Mlon the same «H2ov = a 2 + c 2 Ν2ο f F = b2 + d2 Μ2ο n = a2 + d2 Μ2ο n = b2 + c 2 Assumption 9 Value 12 = c 2-d2 = N2〇iv /-M2o In addition • Set the function I (k) to I (k) = f (k) * 11 + f (L -K) * 12 In the above formula, * f (k) is a function, which decreases when k increases. This function f (k) indicates that the size of the nail-shaped noise generated by each multiple number pole on the trace is closer to the end of the waveform receiving the multiple number (hereinafter referred to as driving the larger signal). Scan the \ »/ end (please read the precautions on the back before filling out this page) • Printed. Printed by the Central Bureau of the Ministry of Economic Affairs
VI K VI 11 為 K 數 绝 之 極 笛 描 掃 為 L 8 甲 4(210X 297 公发） 201825 ' A6 _ B6 五、發明説明（7 ) 绝結而言，從選擇掃描電極Yn變成蘧擇捅描霣極Yn + 1 時，在第Κ锢捅描電極Yk上産生配合函數I(k)之數值之 絶對值之釘狀雜訊。換言之，函數I(k)之數值增大時， 産生大雜訊。産生雜訊之方向偽根據函數I(k)之數值之正 負來決定。 換言之，在掃描電極Yk上，配合該函數I(k)之數值之 釘子狀雜訊之電壓方向舆施加在各信號電極之電壓波形 之變化成為同相時，施加於該信號電極與捅描電極Yk所 形成之顯示點之實效電壓減小而顯示變淡，相位相反時 ,則增加而加濃。由於上述結構而殘留有顯示不均勻之 現象。 以下參照第16_及第17圖説明實施霣壓修正法後，仍 存在之不同之顯示不均勻。第16画及第17圈表示 液 晶面板1及各不相同之顯示内容。圆中，液晶面板1之 構造與第14圈中之液晶面板1完金相同。因此以相同表 示而省略其説明。第16圖，第17圖中，於揷描霣極Y1〜 Y6之左倒。利用本發明之發明人申請之待願昭63-15994 號之顯示不均匀改善方法中，以横向拉線改菩顯示不均 之霣壓修正法，施加由於顯示圓型而變化之掃描霣壓波 形。換言之，配合被選擇之捕描笛極上之點亮點數目Z ，在Μ擇®壓上重疊修正《壓。第16及第17圔中之顯示 内容表示將相似之四角形分別顳示在移靠至左麻與右端之位置 (請先閲讀背面之注意事項再填寫本頁) .装_ .訂· .線.VI K VI 11 is the number K, and the extreme whistle sweep is L 8 A 4 (210X 297 public) 201825 'A6 _ B6 V. Description of the invention (7) In terms of extinction, from selecting the scanning electrode Yn to selecting When the trace electrode Yn + 1, a spike noise is generated on the trace electrode Yk of the K-th prong electrode, and the absolute value of the value of the coordination function I (k) is generated. In other words, when the value of the function I (k) increases, large noise is generated. The direction of noise generation is determined by the sign of the function I (k). In other words, when the voltage direction of the nail-shaped noise matching the value of the function I (k) on the scan electrode Yk and the change in the voltage waveform applied to each signal electrode become the same phase, it is applied to the signal electrode and the trace electrode Yk The effective voltage of the formed display points decreases and the display becomes lighter. When the phase is opposite, it increases and becomes thicker. Due to the above structure, the phenomenon of display unevenness remains. In the following, referring to Figure 16 and Figure 17, the difference in display unevenness that still exists after implementing the pressure correction method will be described. Picture 16 and circle 17 show the liquid crystal panel 1 and the different display contents. In the circle, the structure of the liquid crystal panel 1 is the same as that of the liquid crystal panel 1 in the 14th circle. Therefore, the description is omitted based on the same representation. In Figure 16, in Figure 17, the left of Y1 ~ Y6 is drawn down. In the method for improving the display unevenness of the application No. 63-15994 filed by the inventor of the present invention, the method for correcting the pressure of the display unevenness by changing the horizontal pull to the display unevenness is to apply the scanning pressure waveform that changes due to the display circular shape. . In other words, in accordance with the number of lighted points Z on the selected tracing pole, overlap the "pressure" on the Mselect® pressure. The display contents in the 16th and 17th mean that the similar quadrilaterals are displayed at the position moved to the left and right ends respectively (please read the precautions on the back before filling out this page). 装 _. 定 ·. 线.
經濟部中央標準局印$L 相 全 完 則 示 顯 之 示 所 0 7 I 9 身 I 及 圖 6 II 行 進 若 此 因 甲 4(210X 297乂 尨） 201825 A6 B6 五、發明説明（8 ) 同之修正霣壓將施加在第16函及第17鼷之液晶面板1之 各掃描電極Y1〜Y6上。 依照第16園，修正霄K過多，故在顯示之四健角之橫 方向産生較濃之顯示不均勻。相反的，依照第17圏，修 正電壓遇少，而將殘留横向拉線所造成之顯示不均勻， 換言之，産生在顯示之四傾角之横方向之較淡之顯示不 均勻，將繼缠保留，其理由為，構成液晶面板1之各掃 描電棰Y1〜Y6之電阻與顯示點所形成之電容器不能構成 積分電路，以及點亮點所形成之電容器之霣容量大於未 點亮點所形成之電容器之電容量。換言之，使離開施加 掃描笛壓波形之端部（以後稱驅動單）較逋之點亮點之波 形發生失真之影響大於離開較近之點亮點之影響。因此 ，若點亮黏在較速處，則産生較大之失真。因此，若點 亮點在離開各掃描霣極Y1〜Y6之驅動端較逋之位置時， 包括修正《壓之捕描霄壓波形將發生失真。因此，施加 (請先閱碛背面之注意事項再填駕本頁) .¾. •訂. 被棰 , 電 中號 S ί Υ 倍 ~ 各 Υ1之 極中 電ΧΡ 描 ~ U 11 捐 X 個棰 i S 霣 。設號 少假信 減 〇 個 壓況 P 電情輿 效般棰 實一 霣 之明描 點説播 示下之 顯以擇 於選 li1*- 産 將 則 之 成 0 所 I'線 為拉 置向 位横 之由 點之 示 Ζ 顯值 之數 丨 成之 } 形出(ί 所算 q1 P 式 &Σ-1 一下 II :據 。 ,2根勻 1 於均 = W , 合不 Π配示 χί生顯 Ζ δ 經濟部中央梯準局印製 甲 4(210X 297 7 沒） 201825 A6 B6 五、發明説明（9 ) 中 式 上 巨 數 之 極 S 號 信 極 電 描 掃 為 函 時 點 。亮 數黏 凾點 加示 增顯 之之 大置 增位 時 i 上 大ffi 愈棰 S 霪 • 1 數描 變掃 為之 U擇 ^ 0 數被 函當 數 δ 為 成 數較根 ，行 ， 此進此 因點因 亮 點 之 遠 較 。 端值 給數 供之 ο 之目 為極數 成電點 時描亮 占 HP 6 亮開出 點離算 6Σ 未用而 II 在利權 Z 而俗加 ,Z 之 ί值大據 δ 勻 位 均 品 不 示 示。顯 顯勻使 之均此 成不因 造示。 所顯勻 線除均 拉消不 向全示 横完顯 之能生 用不發 習，仍 因法 ， 對正構 Ζ 修結 值壓述 數電上 之之照 出施依 算實 4 91出 59取 7的 63量 昭定 願即 特亦 施， 實時 種法 一 方 供善 提改 係之 的勻 目均 之不 明示 發顯 本之 ,示 。比掲 低因所 降 號 法捅 方與 之置 正位 修一不 之顯 量之 出型 取圖 該示 於顯 應慮 對考 施 ， 實際 而之 性量 則出 規取 之算 容計 内在 示 ， 顯時 的示 步顙 一 晶 進液 此之 藉看 ，觀 你易 關容 置而 位位 之品 端示 動顯 驅高 之提 棰 ， 電 勻 號均 倍不 及示 棰顯 霄菩 描改 之 Η 基 對 1 之 雇 晶 掖 持 挾 在 為 0 待 1Α 之 明 。發 置本 裝 (請先閲讀背面之注意事項再填寫本頁) •良. .打* k. 經濟部中央抹準局印裝 極施 «而 號群 佶棰 成霣 形描 之拥 上於 Η 形 基波 之壓 方霣 一 描 S掃 於加 ，施 群 ， 極上 霣板 描面 拥晶 成液 形之 方群 甲 4(2lOX 297 公发） 201825 A6 B6 五、發明説明（10) 加倍號電壓波形於倍號電極群，以顯示圈像或文字，配 經濟部中央橾準局印裝 波置離開一 受少極 之颺 所 極數之，正 壓裝型離少 接至轚 示置 時 霣點加作修 電示_ 與至 少之一 顯位 。 案 號亮相動形 描顯之置之 至形每 所之 化 圖 倍點差置波 播晶字位置 對波之 板端 變 盤 ，之之裝之 使液文之位 ，壓群 面動 生 棋 慮上目示果 ,之及端之。* 電極 晶驅 發 示 考極數顯結 型化像一端置之號轚 掖之 極 顯 據霣點晶其 變圖之一裝 £ 倍號 ，極 霣 明 根描亮掖於 之生之形之之電與信 盧電 各 説 供掃點使應 字發示波形化正形或 考與 封 先 度之之面對 文方顯 K 波變修波群 給置 置 首。程擇上方行 及 一所電 K 量使® 極 供位 之 。例之 S 極一進 像少板描電之，電電 可示 壓 例施勻被® 要而 匾至面掃號壓為描描 ，顯 電 施實均將描只和 之之晶受倍電擻掃掃 構之 正 實之不，掃，之 示形液接受正特之之 結型 修 1 勺示置之此加 顯波合中接修 2 壓形 之匾。使 第均顯位擇因相 所壓配群中使第霣波 明之壓可 1中不，之 S 。其 板電有極群，之正壓 C 發像霣又 1 明示述端被定算 面號括電極置明修電化本圏正，例説顯所動外決計 晶倍包描電位發有之變照及修外施下之上驅另來面 液或，掃號之本叠方生依字之另實以生如離輿值方 合形中開信方 重一發 文係 [ 發 距目數一 ...................................................汊..............................打……........................终 (請先閔請背面之注意事項再填¾本頁) 甲 4(210X 297 公发） 201825 A6 B6 經濟部中央標準局印裝 五、發明説明（11 ) 即可。以下説明進行造種修正之液晶顯示裝置之具體實 施例。 第1圈表示本實施例之結構。園中，101為液晶單元， 俗由液晶面板201,掃描電極驅動電路205及信號電極驅 動電路213所構成。102為用來控制液晶顯示裝置之動作 之一連之控制信號，包括閂鎖倍號L P ,幀信號F R ,資料 輪入倍號DIN, X驅動移位時鐘信號XSCL及其他所構成。 103為資料倍號，俗用來決定顯示圖型之倍號。資料信 號103在信號XSCL上昇時發生變化，在下降時被輸入液 晶單元101與電壓波形修正電路104中。104為電壓波形 修正電路（以後稱修正電路）。105為電源電路。108為産 生同步於倍號LP之時鐘信號11〇之電路。106為從電源電 路105输出之用來驅動播描電極所需之2組電源（以後稱Y 電源）。107為從霣源105输出之用來驅動倍號鬣極所需 之2組電源（以後稱X電源）。109為用來決定修正電路104 所輪出之修正電壓之量之修正倍號。110為同步於分頻 霄路108所输出之信號LP之倍號（以後稱修正用時鐘）。 以下説明第1圈之各構成要素之具體結構之一實施例 。第2圖表示液晶單元101之具驩結構之一例。圈中， 201為液晶面板，在其挾持著液晶層之一對基板202, 203之一基H202上形成有横向排列之掃描霄極Y1〜Y6, 而在S —基片203上形成有縱向排列之倍號霣極XI〜X6 。倍號電極XI, X3, X5在其上側具有接受倍號霣壓波形 -13- 甲 4(210X 297 公发） (請先閲讀背面之注意事項再填寫本頁) .蛑. ,訂· 201825 A6 B6 五、發明説明（12) 之端子，而X2, X4, X6在下僱具有端子。播描霄極Y1〜 Y6與倍號電極X〗〜X6交叉形成顯示點204。本例中，為 了說明之方便，液晶面板201之構造為6x6點。但本發明 不受其限制。205為掃描霣棰驅動電路（以後稱Y驅動翠） ,其中包括移位暫存器206,由多個位元所構成之暫存 器同時移位之移位暫存器207,閂鎖電路208,計數器 209, —致檢测電路210,轉換電路211及位準移位器212 所構成。位準移位電路212之输出被導入液晶面板210及 各掃描電掻Y1〜Y6中。第3匾表示Y驅動器205之更具體 之例。圓中，206為移位暫存器，在信號LP下降時输入 信號DIH,而在信號LP下降時將倍號DIN依次傳送至移位 暫存器206之各暫存器。倍號DIH以高電位”Η ”為有源”1” ，通常，以掃描霣棰Υ1〜Υ6之數目，或更多之倍號LP之 數目之間隔，每次輪出一個。因此，"1"之資料通遇移 位暫存器206内，而其他則成為非有源”〇”。此時，各暫 存器將其内容作為控制信號C0予以輪出。 207為由具有多緬位元之暫存器所構成之移位暫存器 。本例中，傜由5個位元構成1僮暫存器。該移位暫存器 像由構成修正佶號109之Υ修正移位時鐘（以後稱佶號 YCSCL)之作用而動作。由於倍號YCSCL之作用依次输入 用來決定構成修正倍號109之修正之量之強倍號（本 例中為4位元，10〜13)及用來決定修正《壓之極性之符 (請先K1讀卄面之注意事項再填寫本頁) .装_ 訂. 經濟部中央抹準局印裴 0 F 號 信 號 甲 4(210X 297 公发） 201825 A6 B6 五、發明説明（13) 208為閂鎖電路，由於信號LP之作用而輸入移位暫存 器207之内容。 209為計數霣路，傜具有輿強度信號1〇〜13之位元數 相同位元數之上昇計數器。該計數器209由於修正用時 鐘101之作用而上昇計數。並由於佶號LP之作用而愎置 Ο 210為一致檢澜霣路，各電路210比較閂鋇電路208之 各對應之暫存器與電路209之输出而檢測是否成為一致 。若成為一致時，卽産生成為有源"1”之控制信號C2。 此時，若輸人於暫存器之符號倍號F之内容（稱之為控制 信號C 1 )表示負數（符號信號F為有源” 1 ”）時，則反轉移 位暫存器之各位元而檢測是否與顯示之數值加上1之 數值成為一致。檢測結果一直被保持至下一個佶號LP逹 到為止。 211為轉換電路，像將構成Y霣源106之10個電SV0, V 1 U , V 1 , V 1 L , V 2 , V 3，V 4 U , V 4 , V 4 L , V 5 中，電壓 V 0 ,V4U , V4 , U4L分成第 1¾ ® 組，電壓 V5 , V1L VI, V1U分成為第2電壓組，依照信號FR轉換成該2組電懕中之 一種之轉換轚路。 在此，假設《壓VO, V4U, V4, V4L分別為第1霣壓組 中之選擇堪®，修正堪路（ϋ),非選擇® ffi,修正霣壓 經濟部中央橾準局印裝 (請先聞讀背面之注竟事項再填寫本頁) _訂. (L)。修正堪壓（U)與（L)統稱為修正霣壓。同樣的，霣 壓V5, V1L, VI, V1U分別為第2電壓組中之蘧擇霜 -1 5 - 甲 4(210X 297 公发） 01825 A6 B6 五、發明説明（丄4) 修正電壓（U ),非選擇電壓，修正霄壓（L )。 212為位準移位器，偽由多_4回路1接點之開蘭所構 成。 當控制倍號C0成為”1”時，各開關選擇S1,亦邸選擇 該選擇電壓。 當控制信號C 0成為” 0”而控制倍號C 2成為"1 ”時，各開 關選擇S3。亦即選擇該非選擇電壓。 當控制倍號C 0與C 2皆為” 0 ”而控制信號C 1為” 0 ”時，各 開關選擇S2。若C1為”1”時，則Μ擇S4。 Υ驅動器205具有上述結構。移位暫存器207因強度信 號10〜13輿符號倍號F而成為5位元結構。但亦可藉著增 減強度倍號之位元數而適當的增減該移位暫存器207之 位元數。 構成移位暫存器206及207之暫存器之數目，閂鎖電路 208, —致檢測電路之數目，構成位準移位器之開鼷之 數目僳如圈中所示的相等於第2圔中液晶面板之掃描霄 極Υ1〜Υ6之數目。 以下説明Υ驅動器205之動作。第3圖中，將倍號DIN同 步於倍號LP輸入於移位暫存器206而傳送。結果位準移 位器212與其成為對應的依次輪出蒱擇霣壓。（以後， 將産生選擇霣壓之開關稱為被選擇之開關，而將輪出其 他笛壓之開闋稱為未被選擇之開關）此時，與其成為同 步的，在倍號LP之一傕遇期内，由佶號YCSCL將強度倍 -1 6 - 甲 4(21UX 297 公发） (請先閲讀背面之注意事硕再填寫本頁) 鈐· ，訂. 201825 A6 B6 經濟部中央標準局印裝 五、發明說明（15) 號10〜13,符號倍號輸入移位暫存器207中。未被選擇 之各開鼷産生修正電壓（ϋ)至（L), 一直到由強度倍號10 〜13所表示之數值之絶對值舆計數器209因修正用時鐘 之作用而上昇計數之數值成為一致為止。輸出修正電壓（U )或（L)傜由控制倍號C1所決定。換言之，傺由符號信號 下為”0"或”1”而決定。在與計數器209之數值成為一致 後，未被選擇之各開關産生非選擇電靨。因此，強度信 號10〜13所表示之數值之絶對值愈大，未被選擇之開闋 輪出更長時間之修正電壓。以上説明Υ驅動之結構及動 作。 第2圖中，213為由移位暫存器214,閂鎖電路215, 及位準移位器216所構成之倍號電極驅動堪路（以後稱X 驅動器）。位準位移器216之输出被導入液晶面板201之 各倍號電極XI〜Χ6。 移位暫存器214以信號XSCL作為時鐮輪入決定顯示圖 型之表示點亮或未點亮之資料倍號1〇3(此時，假設點亮 為有源” 1 ·’，未點亮為非有源” 0 ”）。然後，對應於倍號 馆掻XI〜Χ6輸入金部資料倍號103之後，由倍號LP输入 於閂鋇電路215中。然後，位準移位器根據閂鎖霣路215 之内容及倍號FR,産生一定之電壓。換言之，將構成X 霜源]07之4傾霣壓V0,V2,V3,V5中，堪壓V5,V3分成第 1電IS組，笛壓V0,V2分成第2霄壓組，而根據倍號FR採 用此2組電壓中之一棰（在此，將第1組之®壓V5,V3分別 -1 7 - (請先W讀背面之注意事項再填寫本頁) .象. .訂. •線. 甲 4(210X 297 公发） A6 B6 201825 五、發明説明) 稱為第1組之點亮電鼷，未黏亮電S,將第2組之電KV0 ,V2分別稱為第2組之點亮電壓，未點亮電壓）。若閂鋇 電路215之内容為"1”時，則産生由倍號FR決定之一組之 點亮電壓，若為”0”時，則産生未點亮電壓。 液晶單元101具有如上之結構。因此，液晶單元101中 ，同步於倍號DIN, LP依次施加選擇電壓於掃描霄極Y1 〜Y6,而同步於該電壓之施加，在倍號電棰XI〜X6上施 加配合顯示圖型之點亮或未點亮電S,使液晶面板201 進行顯示。此時修正信號109之強度倍號10〜13及長度 ，極性配合符號信號F之修正電壓，取代非選擇電壓施 加於未施加有選擇電壓之掃描電棰Y1〜Y6上。 以上説明液晶單元101之結構及動作。 第1圖中，104為修正電路，傜將第2臞之液晶面板201 之某一掃描電極Yn舆S —掃描電極Yn + 1 (n = 1 , 2----5, ，但η=6時，則變成1而非為η + 1)上之點亮黏204之數 目，分別考慮其離開佶號笛極XI〜Χ6之施加驅動波形之 端部之距離予以計數，根據其差值形成強度信號10〜13 及符號信號F之霣路。以下參照第4圓説明具體之例。第 4圖中，401為撥動正反器（以後稱T-F/F)。402U,402L 為閛電路。403,404為計數器。40511,405!^為乘數産生笛 路。406U.406L為計數器。407U,407L為閂鎮電路。408U ,408L為進行減算之運》霄路。4091K409L為記億元件， 410U,410L為閂鋇II路。411U,411L為進行乘算之蓮算笛 - 1 8 _ 甲 4(210X 297 公发） (請先KI請卄面之注意事邛再填寫本頁) •W-. .訂* 經濟部中央標準局印裝 201825 A6 B6 經濟部中央橾準局印裝 五、發明説明（17) 路。412為進行加算及除算之運算電路。 T-F/F電路401因信號LP之作用而被復置，産生"0", 對閘電路402L成為有源之输入，而對閛402U則成為非有 源輪入。當倍號XSCL所形成之時鐘輸入後，即反轉。因 此，當對應於第2圖之液晶面板201之倍號電棰XI〜X6中 之偶數電棰X2,X4,X6之資料信號進入第4圖之閘極電路 402U及402L時，只有閘電路402L變成有源，而輪出資料 信號。閘電路402U之輸出與資料信號無關的成為非有源 。相反的，當第2圖之對應於奇數電極XI,X3,X5之倍號 進入第4圖之閘極電路402U及402L時，只有閘極電路402U 成為有源而輸出資料倍號。閘極電路402L之輪出與資料 信號無關的成為非有源。換言之，該3偕電路將關連於 從倍號電極之上驅動之電極之資料倍號輿關連於從下驅 動的電極之資料倍號予以分離。該等被分離之資料倍號 分別稱之為上資料倍號，下資料倍號。 計數器406U及406L在毎一個被分離之資料倍號時，計 數表示點亮之"1”之狀態之數目。 換言之，只在倍號XSCL上昇時閘極402U,L之输出成為 有源時，計數器506U及L進行加算。該加算結果（假’設該 數值為H 1 ο η , Μ 2 ο η )與佶號L P成為同步的分別被输入閂鎖 锺路407U及L。運算笛路4081]及408L分別進行其输入剎 那前之閂鎖電路407U,L之數值（假設此數值為Ν1〇η,Η2〇η )及計數器406U及L之數值Μ1οη,Μ2〇η之減算。假設其運 -1 9 - (請先Κ1讀背面之注意事項再填寫本頁) .蛑. .訂· -線. 甲 4(210Χ 297乂发） 201825 A6 B6 五、發明說明（18 ) 算結果為 11 = N 1 〇 w - Ml 〇 12 = W2ow - M2o 計數器403為只可表示相當於第2圔之掃描電極Y1〜Y6 之數目之狀態之上昇計數器，本例中，從〇上昇計數至5 ，而被信號DIH復置成為0。第4_之計數器403之輸出作 為記億元件（以後稱為記億器）40911及L之位址。位址之 數值在選擇電壓施加於第2液晶面板201之掃描電棰Yn之 時，成為η - 1。 然後，將上述數值II, 12分別寫入記億器409U及L中， 計數器403所表示之位址中。 以下説明具體動作，當選擇笛壓施加於第2圖之掃描 電棰Υη之期間内，計數器406U及L分別計數Υη + 1上之點 亮黏數目Mlon,M2on。在選擇電壓施加於掃描電極Υη + 1 之剎那前，計數保持於閂鎖電路408U及L之數值Nlon, N 2 ο η ,亦即與掃描電極Υ η上之點亮點數目之差11 , I 2 , 分別寫入記億器409ϋ及L之η-〗位址中。此動作反復的進 行至η從〇變成6然後回到0為止。換言之，在記億器409U 及L中，於0位址分別儲存箸描描霣棰Υ1 及Υ2上之點亮 點數目之差11及12。 在1〜5位址分別儲存著掃描笛極丫2與Υ3,Υ3與Υ4,Υ4與 Υ5,Υ5與Υ1上之點亮數目之差11,12。 經濟部中央捃準局印裝 (請先閲讀背面之注意事項再填寫本頁) .訂· .φ. 計數器404為只可表示相當於描描電極Υ1〜Υ6之數目 -20 - 甲 4(210X 297 7^) 201825 A6 B6 五 '發明說明（19) 之 狀 態 之 上 昇計 數 電 路。 本 例 中，從0上昇計數至5 $ 而 由 信 號 LP愎 置為 0 o 输入於計數器404之 時鐘 (此時鐘稱 之 為 信 號 YCSCL) o 只 要是 在 佶 號L P之一 値邇 期 内 發 生 捅 描 電 棰 Yl Y6之 數 S 之變 化 之 時鐘即可 〇本 實 施 例 中 9 以 信 〇上 m XSCL作為 倍 號 YCSCL 〇 該計數器404之 輪 出 作 為 乘 數 産 生 電 路 405U 與 L之位址之用。 405 U 及 L為産生乘數用之電路，像由只讀記億元件（以 後 稱 ROM) t 二極 齷 矩 陣等 所 構 成之數值 表格 ♦ 亦 即 産 生 將 輪 入 之 位 址作 為 變 數之 函 數 之數值之 鼋路 〇 假 設 位 址 為 K時， 40 5 U為形成f (k)之數值之函數表格。 函數f (k) 為 K 愈 大 時 變成 愈 小 之減 小 函 數。函數 之形 式 像 由 實 驗 等 箄 出 〇 在 此， 為 了 說明 時 之 方便，設 定為如下之 發 生 直 線 變 化 之 函 數。 k = 0 時， f (0)= I 5 k = I 時， f (I)= 14 k = 2 時， f (2)= I 3 k = 3 時， f (3)= I 2 k = 4 時， f ( 4 )= II k = 5 時， f (5)= I 0 S- 生 産。 數路 乘霣 -之 的數 樣函 同之 義 路 為The Central Bureau of Standards of the Ministry of Economic Affairs prints $ L. If the phase is completely completed, the display will show 0 7 I 9 Body I and Figure 6 II. If this is the case 4 (210X 297 乂 尨) 201825 A6 B6 V. Description of the invention (8) Same The corrected pressure is applied to each of the scan electrodes Y1 to Y6 of the liquid crystal panel 1 of the 16th and 17th liquid crystals. According to the 16th circle, the correction K is too much, so that the display of the four healthy angles in the horizontal direction produces thicker display unevenness. On the contrary, according to the 17th circle, the correction voltage is low, and the display unevenness caused by the residual lateral pull-out line, in other words, the lighter display unevenness in the horizontal direction of the four inclination angles of the display will remain entangled, The reason is that the resistance of each of the scanning electrodes Y1 to Y6 constituting the liquid crystal panel 1 and the capacitor formed by the display point cannot constitute an integrating circuit, and the capacity of the capacitor formed by the lighting point is greater than that of the capacitor formed by the unlit point capacity. In other words, the effect of distorting the waveform from the end of the applied flute pressure waveform (hereinafter referred to as the drive order) from the brighter lighting point is greater than that from the closer lighting point. Therefore, if the lighting sticks at a faster speed, there will be greater distortion. Therefore, if the bright point is away from the driving end of each scan electrode Y1 ~ Y6, the waveform will be distorted if the correction is included. Therefore, apply (please read the precautions on the back of the moraine first and then fill in this page). ¾. • Order. Quilted, electric medium number S ί Υ times ~ each pole of the Υ1 CLP description ~ U 11 donated X pieces i S 霣. Set the number of less false letters minus 0 pressure status P. The public sentiment is as good as the public. A clear description of the broadcast shows the choice to choose li1 *-the production will be 0. The I 'line is set. The direction of the horizontal direction is indicated by the point. The number of apparent values is formed. (Formulated by q1 P formula & Σ-1: II. According to., 2 roots are equal to 1 = both are equal to W, and the combination is not suitable. Show X sheng Xian δ δ Printed by the Central Bureau of Economic Affairs of the Ministry of Economic Affairs 4 (210X 297 7) 201825 A6 B6 Fifth, the invention description (9) The huge number of poles in the Chinese style, the S-electrode scanning is the function time. When the number of sticky points is increased, the larger the value is, the bigger the position is. The bigger the ffi is, the better the S is. • The number 1 is selected by the sweep. ^ 0. The number is the function. The number δ is a number. This point is far from the bright spot. The end value is given to the data. The purpose is to count the number of points when the pole becomes an electric point. The bright point is 6 HP. The bright point is 6Σ unused and II is used in the right Z. The value of ί is large according to δ. Evenly-balanced products are not shown. The display of the uniformity makes it equal to no cause. The uniformity of the displayed line is not drawn to the full display of the energy of the horizontal display. If you do n’t learn, it ’s still a matter of law. For the normal structure Z, you can read the photos on the digital screen according to the calculation method. 4 91 out 59. Take 7 out of 63. It ’s a good idea to make sure that the uniformity of the eyes is unclear, and it ’s shown. For Kaushi, the actual amount of sex is shown internally in the calculated volume meter. When the display is on, a crystal is poured into the liquid. Take a look at it and watch your easy-to-close content and position. The height of the drive is higher, and the electric uniform number is not as good as that of the display of the change of the Η group on the 1st. The employment of the crystal is held at 0 and waits for 1Α. To publish this equipment (please read the precautions on the back first) (Fill in this page again) • Liang ... hit * k. The Central Approval Bureau of the Ministry of Economic Affairs prints «and the number group is formed into an eye-catching embrace on the pressure of the H-shaped fundamental wave. Plus, Shi Qun, Fang Qun Jia 4 with a crystal shape and a liquid crystal shape (2lOX 297 public) 201825 A6 B6 V. Description of the invention (10) Double the voltage waveform on the double electrode group to display the circle image or text, matching the printed wave of the Central Bureau of Economics of the Ministry of Economic Affairs, away from the number of poles that are under the influence of the young pole, positive pressure When the model is connected to the display, it will be added as a repair indicator _ with at least one display bit. The case number appeared in the dynamic description. The location of the shape of each shape is doubled. The position of the wave broadcast crystal is changed to the plate end of the wave, and the device is placed to make the position of the hydrology. Consider the above results, and the results. * The electrode crystal drive displays the digital display structure of the pole. The pole is displayed at one end. The pole is displayed according to one of the changes of the dotted crystal. The number is doubled, and the pole is clearly rooted to describe the shape of the life. Zhidian and Xinludian each said that the scanning point makes Yingzi send a wave-shaped orthodox or a test and a letter to Feng Xianzhi. The face of the Wenfang K wave modification wave group is set first. Select the upper row and an electrical K to make the ® pole available. For example, the S pole can be traced like a small plate, and the electricity can show pressure. For example, the shining is required. The plaque to the surface scan number is the trace. The sweeping structure is correct, the sweeping shape-receiving fluid accepts Zhengte's knot-shaped repair, 1 scoop, and this display is followed by repairing 2 pressure-shaped plaques. In the matching group pressed by the selective position of the first apparent position, the pressure of the first epoch can be 1 not, S. Its plate electricity has a pole group, and the positive pressure C is like a fan. The display end is set by the number of the face, including the electrode, and the electrode is fixed. According to the application and repairing, the next face-lift or the upper face drive, the number of the scanned party is based on the word, and the letter is the same as the one from the public value. .................................................. . 擊 ...................................................... ......... finally (please pay attention to the matters needing attention on the back and then fill in this page) A4 (210X 297 public issue) 201825 A6 B6 Printed by the Central Bureau of Standards of the Ministry of Economy V. Description of the invention (11) That's it. The following describes a specific embodiment of a liquid crystal display device that performs seed correction. The first circle shows the structure of this embodiment. In the garden, 101 is a liquid crystal cell, which is generally composed of a liquid crystal panel 201, a scan electrode driving circuit 205, and a signal electrode driving circuit 213. 102 is a series of control signals used to control the operation of the liquid crystal display device, including the latch multiplier LP, the frame signal FR, the data rotation multiplier DIN, the X drive shift clock signal XSCL and others. 103 is the data multiple, which is commonly used to determine the multiple of the display pattern. The data signal 103 changes when the signal XSCL rises, and is input to the liquid crystal cell 101 and the voltage waveform correction circuit 104 when it falls. 104 is a voltage waveform correction circuit (hereinafter referred to as a correction circuit). 105 is the power supply circuit. 108 is a circuit that generates a clock signal 110 synchronized with the multiple LP. 106 is two sets of power supply (hereinafter referred to as Y power supply) needed to drive the scanning electrode output from the power supply circuit 105. 107 is the two sets of power supply (hereinafter referred to as X power supply) required to drive the No. 1 pole output from the Enyuan 105. 109 is a correction multiple used to determine the amount of correction voltage that the correction circuit 104 rotates. 110 is a multiple of the signal LP synchronized with the frequency division Xiaolu 108 (hereinafter referred to as the correction clock). An example of the specific structure of each component in the first circle is described below. FIG. 2 shows an example of the structure of the liquid crystal cell 101. In the circle, 201 is a liquid crystal panel, on which a pair of substrates 202, 203 holding a pair of liquid crystal layers is formed on a substrate H202 with horizontally arranged scan electrodes Y1 ~ Y6, and a vertical arrangement is formed on the S-substrate 203 The number of times is XI ~ X6. The double electrodes XI, X3, X5 have on the upper side a wave size that accepts double pressure-13- A 4 (210X 297 public) (please read the precautions on the back before filling in this page). 蛑., Order · 201825 A6 B6 5. Terminals of the invention description (12), and X2, X4, X6 have terminals under it. The display points Y1 ~ Y6 and the double electrodes X〗 ~ X6 cross to form a display point 204. In this example, for convenience of explanation, the structure of the liquid crystal panel 201 is 6x6 dots. However, the present invention is not limited thereto. 205 is a scanning drive circuit (hereinafter referred to as Y drive Cui), which includes a shift register 206, a shift register 207 that is composed of a plurality of bits, and a latch circuit 208 The counter 209 is composed of a detection circuit 210, a conversion circuit 211 and a level shifter 212. The output of the level shift circuit 212 is introduced into the liquid crystal panel 210 and the scanning electrodes Y1 to Y6. The third plaque shows a more specific example of the Y driver 205. In the circle, 206 is a shift register. When the signal LP falls, the signal DIH is input, and when the signal LP falls, the multiple number DIN is sequentially transmitted to each register of the shift register 206. The multiplier DIH takes the high potential "Η" as the active "1". Generally, it is one at a time at intervals of the number of scans Y1 ~ Y6 or the number of more multipliers LP. Therefore, the data of " 1 " passes into the shift register 206, and the other becomes non-active "〇". At this time, each register rounds off its contents as the control signal C0. 207 is a shift register composed of registers with multiple Burmese bits. In this example, Wu consists of 5 bits to form a child register. The shift register is operated by the action of the Y-corrected shift clock (hereinafter referred to as YCSCL) which constitutes correction number 109. Due to the function of the multiplier YCSCL, the strong multiples (in this case, 4 bits, 10 to 13) used to determine the amount of correction that constitutes the correction multiples 109 are entered in sequence and the sign used to determine the correction of the (K1 first read the notes on the noodles before filling out this page). Binding _ Order. Central Ministry of Economic Affairs, Central Approval Bureau printed Pei 0 F signal A 4 (210X 297 public release) 201825 A6 B6 V. Description of invention (13) 208 is The latch circuit inputs the contents of the shift register 207 due to the signal LP. 209 is a counting counter, and it has an up counter with the same number of bits as the intensity signal 10 ~ 13. The counter 209 is counted up by the correction clock 101. And due to the role of the LP number, the 210 is a consistent detection circuit. Each circuit 210 compares the output of the corresponding register of the barium circuit 208 with the output of the circuit 209 to detect whether they are consistent. If they become consistent, a control signal C2 that becomes active " 1 "is generated. At this time, if the content of the symbol multiple F (referred to as the control signal C 1) input to the temporary memory is a negative number (signal signal When F is active "1"), the bits of the shift register are reversed to detect whether it is consistent with the displayed value plus the value of 1. The detection result is kept until the next LP number is reached 211 is the conversion circuit, which will constitute 10 electrical SV0, V 1 U, V 1, V 1 L, V 2, V 3, V 4 U, V 4, V 4 L, V 5 , The voltages V 0, V4U, V4, U4L are divided into the 1¾ ® group, and the voltages V5, V1L VI, V1U are divided into the second voltage group, which is converted into one of the two groups according to the signal FR. ， Assuming that “Press VO, V4U, V4, V4L are the first choice in the first group of pressure, can correct ®, correct choice (ϋ), non-selection ® ffi, correct the pressure of the Ministry of Economic Affairs Central Central Bureau printing (please first Read the notes on the back of the page and fill out this page) _Subscribe. (L). The correction pressure (U) and (L) are collectively referred to as the correction pressure. Similarly, the pressure V5, V1L, VI, V1U are the first 2 voltage group Zhimingshuang-1 5-A4 (210X 297 public) 01825 A6 B6 V. Description of the invention (丄 4) Corrected voltage (U), non-selected voltage, corrected low pressure (L). 212 is the level shift The device is composed of a multi-4 circuit 1 contact open. When the control multiple C0 becomes "1", each switch selects S1 and also selects the selected voltage. When the control signal C 0 becomes "0" and When the control multiple C 2 becomes " 1 ″, each switch selects S3. That is, the non-selected voltage is selected. When the control multiples C 0 and C 2 are both “0” and the control signal C 1 is “0”, each switch selects S2. If C1 is "1", then S4 is selected. The Y driver 205 has the above structure. The shift register 207 has a 5-bit structure due to the intensity signals 10 to 13 and the sign multiple F. However, the number of bits of the shift register 207 can also be appropriately increased or decreased by increasing or decreasing the number of bits of the intensity multiple. The number of registers constituting the shift registers 206 and 207, the number of latch circuits 208, and the number of detection circuits, and the number of openings constituting the level shifter are as shown in the circle equal to the second The number of scanning poles Υ1 ~ Υ6 in the LCD panel. The operation of the Y driver 205 will be described below. In Fig. 3, the double number DIN is input to the shift register 206 in synchronization with the double number LP and transmitted. As a result, the level shifter 212 and the corresponding one sequentially turn out the selective pressure. (In the future, the switch that generates the selected pressure will be called the selected switch, and the switch that turns out the other flute pressure will be called the unselected switch.) At this time, it will be synchronized with it, in one of the multiples LP. During the encounter period, the number YCSCL will increase the intensity by -1 6-A 4 (21UX 297 public issue) (please read the notes on the back and then fill out this page) 钤 ·, order. 201825 A6 B6 Central Standards Bureau of the Ministry of Economic Affairs Printed fifth, invention description (15) No. 10 ~ 13, the sign multiple is input to the shift register 207. Correction voltages (ϋ) to (L) are generated for each unselected reed, until the absolute value of the value represented by the intensity multiplier 10 to 13 and the value of the counter 209 up-counted by the correction clock become consistent until. The output correction voltage (U) or (L) is determined by the control multiple C1. In other words, Ye is determined by "0" or "1" under the sign signal. After the value of the counter 209 becomes the same, each unselected switch generates a non-selected electric charge. Therefore, the intensity signals 10 to 13 indicate The larger the absolute value of the value, the longer the correction voltage of the unselected open caster. The above describes the structure and operation of the Υ drive. In the second figure, 213 is the shift register 214 and the latch circuit 215. , And the double electrode driven by the level shifter 216 (hereinafter referred to as the X driver). The output of the level shifter 216 is introduced into each multiple electrode XI ~ Χ6 of the liquid crystal panel 201. Shift register 214 When the signal XSCL is used, the sickle wheel enters to determine the display pattern, indicating that the data is lit or unlit. The number of times is 103 (at this time, it is assumed that the lighting is active "1 · ', and the non-lighting is not active. "0"). Then, after inputting the number 103 of the gold part data corresponding to the number of halls XI ~ X6, the number LP is input into the barium latch circuit 215. Then, the level shifter generates a certain voltage according to the contents of the latch 215 and the multiple FR. In other words, among the 4th inclination pressures V0, V2, V3, and V5 that constitute the X frost source] 07, V5 and V3 are divided into the first electrical IS group, and the flute pressures V0 and V2 are divided into the second electrical pressure group. No. FR adopts one of these two groups of voltages (here, the first group of the voltage V5, V3 respectively -1 7-(please read the precautions on the back before filling this page). Elephant ... Order. • Line. A 4 (210X 297 public issue) A6 B6 201825 V. Description of invention) It is called the first group of lighted electric reels, the non-stick bright electric S, the second group of electricity KV0, V2 are called the second Group lighting voltage, unlighted voltage). If the content of the barium latch circuit 215 is "1", a set of lighting voltages determined by the multiple FR is generated, and if it is "0", a non-lighting voltage is generated. The liquid crystal cell 101 has the structure as above Therefore, in the liquid crystal cell 101, a selection voltage is sequentially applied to the scanning electrodes Y1 to Y6 in synchronization with the DIN, LP, and in synchronization with the application of the voltage, the coordinate display pattern is applied on the XI to X6. Turn on or off the light S to make the liquid crystal panel 201 display. At this time, the strength multiple of the correction signal 109 is 10 ~ 13 and the length, the polarity matches the correction voltage of the sign signal F, instead of applying the non-selective voltage to the unselected selective Voltage scanning electrodes Y1 ~ Y6. The structure and operation of the liquid crystal cell 101 are described above. In the first figure, 104 is a correction circuit, and a certain scanning electrode Yn and S—scanning electrode of the second liquid crystal panel 201 Yn + 1 (n = 1, 2 ---- 5,, but when η = 6, it becomes 1 instead of η + 1). The number of light sticks 204 on each, considering their departure from the flute pole XI ~ X6 The distance of the end of the driving waveform applied is counted, and an intensity signal 10 is formed according to the difference ~ 13 and the entropy path of the symbol signal F. The specific example will be described with reference to circle 4. In the fourth figure, 401 is the flip flip-flop (hereinafter referred to as TF / F). 402U, 402L are the circuit of the habbit. 403, 404 are the counter .40511,405! ^ Is the multiplier to generate flute circuit. 406U.406L is the counter. 407U, 407L is the latch town circuit. 408U, 408L is the operation of subtraction. Xiaolu. 4091K409L is the billion-element component, 410U, 410L is the latch Barium II. 411U, 411L are multiplication lotus counting flutes-1 8 _ A 4 (210X 297 public issue) (please first pay attention to KI, please fill out this page) • W-.. Order * Printed by the Central Bureau of Standards of the Ministry of Economic Affairs 201825 A6 B6 Printed by the Central Bureau of the Ministry of Economic Affairs Fifth, Invention Description (17) Circuit. 412 is an arithmetic circuit for addition and division. The TF / F circuit 401 is reset due to the function of the signal LP Set to generate " 0 ", the pair gate circuit 402L becomes the active input, and the pair gate 402U becomes the non-active round-robin. When the clock input formed by the multiplier XSCL is reversed, it corresponds to The data signals of the LCD panel 201 in Figure 2 are the even-numbered cells in XI ~ X6, and the data signals of X2, X4, and X6 enter the gate circuits 402U and 4 in Figure 4. At 02L, only the gate circuit 402L becomes active, and the data signal turns out. The output of the gate circuit 402U has nothing to do with the data signal and becomes non-active. On the contrary, when the second figure corresponds to the odd electrodes XI, X3, X5 When the multiple number enters the gate circuits 402U and 402L of FIG. 4, only the gate circuit 402U becomes active and outputs the data multiple number. The round of the gate circuit 402L becomes inactive regardless of the data signal. In other words, the 3-times circuit separates the data multiples associated with the electrode driven from the double electrode and the data multiples associated with the electrode driven from the bottom. These separated data multiples are called upper data multiples and lower data multiples, respectively. The counters 406U and 406L count the number of “1” states that are lit when each data multiple is separated. In other words, only when the output of the gate 402U, L becomes active when the multiple XSCL rises, The counters 506U and L are added. The result of the addition (assuming that the value is H 1 ο η, Μ 2 ο η) is synchronized with the number LP and is input to the latching circuits 407U and L respectively. Operation flute 4081] And 408L respectively perform the subtraction of the values of the latch circuits 407U, L (assuming this value is Ν10〇, Η2〇η) and the values Μ1οη, Μ2〇η of the counter 406U and L before the input moment. 9-(Please read the precautions on the back of Κ1 before filling in this page). Invaders. Order-line. A 4 (210Χ 297 异 发) 201825 A6 B6 5. Description of the invention (18) The calculation result is 11 = N 1 〇w-Ml 〇12 = W2ow-M2o counter 403 is an up-counter that can only indicate the state corresponding to the number of scan electrodes Y1 ~ Y6 of the second circle. In this example, it up counts from 〇 to 5 and is signaled by DIH Reset to 0. The output of the counter 403 in the 4th_ is used as a billion-counting device (hereinafter referred to as a billion-counting device) 40911 The address of L. The value of the address becomes η-1 when the selection voltage is applied to the scanning electrode Yn of the second liquid crystal panel 201. Then, the above-mentioned values II and 12 are written into the megameter 409U and L, respectively , At the address indicated by the counter 403. The specific operation is described below. When the selection of the whistle voltage is applied to the scanning electrode Υη in Figure 2, the counters 406U and L count the number of lighting sticks Mlon on Υη + 1, respectively. M2on. Before the moment when the selection voltage is applied to the scan electrode Υη + 1, the count is maintained at the values Nlon, N 2 ο η of the latch circuit 408U and L, that is, the difference from the number of lighted points on the scan electrode Υ η 11, I 2, written into the η-〗 address of the megameter 409ϋ and L respectively. This action is repeated until η changes from 0 to 6 and then back to 0. In other words, in the megameter 409U and L, at 0 The address stores the difference between the number of lighted points on the traces Υ1 and Υ2 respectively 11 and 12. The addresses 1 to 5 store the scanning flute poles 2 and Υ3, Υ3 and Υ4, Υ4 and Υ5, Υ5 and The difference in the number of lights on Υ1 is 11,12. Printed by the Central Bureau of Economic Affairs of the Ministry of Economic Affairs (please read the precautions on the back before filling in Page). Order · φ. The counter 404 can only indicate the number of equivalents of the scanning electrodes Υ1 ~ Υ6 -20-A 4 (210X 297 7 ^) 201825 A6 B6 Five 'invention state (19) state of the rising count Circuit. In this example, it counts up from 0 to 5 $ and is set to 0 by the signal LP o The clock input to the counter 404 (this clock is called the signal YCSCL) o As long as it occurs during the period of one of the numbers LP The clock that traces the change of the number S of Y1 and Y6 can be used. 9 In this embodiment, the letter X is the upper XSCL as the multiplier YCSCL. The round of the counter 404 is used as the address of the multiplier generation circuit 405U and L . 405 U and L are circuits used to generate multipliers, such as a numerical table composed of read-only memory devices (hereinafter referred to as ROM) t dipolar mash matrix, etc., that is, a function that takes the rounded address as a function of variable The numerical value path assumes that the address is K, 40 5 U is a function table that forms the value of f (k). The function f (k) is a decreasing function that becomes smaller as K becomes larger. The shape of the function is derived from the experiment and so on. Here, for the convenience of explanation, it is set as the following function that changes linearly. When k = 0, f (0) = I 5 When k = I, f (I) = 14 k = 2, when f (2) = I 3 k = 3, f (3) = I 2 k = 4 When f (4) = II k = 5, when f (5) = I 0 S- production. The number of ways to multiply the number of samples is the same as the right way.
極S- 描 拥 從 為 L 定數 K)之 - 1 “減 f目 由 J數 y 産 來卜 /» 用 (請先閲讀背面之注意事項再填寫本頁) 經濟部中央標準局印裝 η Υ 極 霣 描 拥 之 1Χ ο 2 板 面 晶 液 之 圈 0 2 =5u L 於 -加 中施 例壓 施霣 實擇 本選 。在 巨 甲 4(2lOX 297公沒） 201825 A6 B6 五、發哂棼萌（20} 之期間内，進行第4圖之乘算之運算霣路411U及L分別 進行計數器403所示位址（n-1位址）之記億器4 09U與L之 内容與乘數産生電路405U, L之乘算。此時，計數器404 因倍號YCSCL而進行上昇計數，故乘數産生電路405U及L 所産生之數值同步於倍號YCSCL而變化。換言之，運算 電路411U及L之運算結果分別同步於信號YCSCL而成為 f (0) a» I 1 , f (5) •a. 12 f (1 ) «V I 1 , f (4) 12 f (2) I 1 , f (3) % 12 f (3) % I 1 , f (2) Ϊ* 12 f (4) * I 1 , f (1) 12 f (5) I 1 , f (0) Λ, 12 (請先聞讀竹面之注意事項再填寫本頁) 經濟部中央抹準局印緊 然後，以進行加算及除算之運算電路412將其結果相 加，然後以4除其結果，假設其數值為I時，則將 I = { f (0) * Il + f (5) M2) / 4 I = {f(l)*Il+f(4)M2} / 4 I = {f(2)*Il+f(3)*I2) /4 I = { f(3)* II(2)* 12} / 4 I = { f (m Π+f (m 12} / 4 I = {f(5)ill+f(0)^12} /4 依次同步於信號YCSCL输出。以4相除之理由為若第3 ffl 之移位暫存器207中去除符號信號下時，成為4位元結 構，使該數值I在該4位元内。因此，並非本質上決定 -22- *訂. ，線. 甲 4(2_】0X_2g7C^) 201825 A6 B6 五 '發明説明（21) 者。 在選擇第2圈之掖晶面板201之播描電極Υη之期間内 ，將fU-1), f(L-n + l)乘以捅描霣極Υη及Υη + 1上之點亮 點數目之差II, 12後，將其結果相加而成之數值 1= f(n-l)*Il + P(L-n + l)=U2同步於倍號 YCSCL依次予以 輪出。将其結果之大小作為強度倍號1〇〜13，正負作為 符號倍號F,而與信號YCSCL—起作為第1圖之修正信號 109予以输出。以上説明修正電路1〇4之結構及動作。其 結構不受上述實施例之限制。例如在第4圏中，傲將 1= f(n-l)*Il+F(L-n+l)iI2利用 405U及 L, 411U及 L, 412等電路以即時時間予以計算。但亦可利用CPU等預先 予以計算，寫入重新設置之記億器中，將計數器403,404 之輪出作為位址依次讀出作為修正倍號109予以輸出。 第5圏表示第1團所示霣源電路1〇5之具醭結構之一 實施例。圔中，501〜509為串聯霣阻。其兩端施加有霣 ®V0,V5。此等霣阻501〜509形成為分壓霣路。假設産生 於各霣阻501〜509之電壓從上方設定為V0,V1U,V1,V1L, V2，V3,V4U,V4,V4L,V5,則 V = V0-V 1 =VI -V2 =V3-V4 =V4-V5 經濟部中央橾準局印¾. {請先聞讀背面之注意事項再填寫本頁) •訂· .蝝· V2-V3 = aiV ,其中a為常數，其數值大約為1〜50左右 -23- 甲 4(210X 297〇*) 201825 A6 B6 經濟部中央抹準局印製 五、發明説明（22) 。各電阻值501〜509設定為可成立下列4式之數值。 V1U-V1 = V4-V4L V1-V1L= V4U-V4 510為不改變電阻501〜509所形成之各電壓值，只降 低其阻抗之電壓穩定電路。像由運算放大器所形成之電 壓跟隨器，電晶體所形成之射極跟随器等所構成。如圖 中所示，電壓穩定電路510像針對由電阻501〜509分壓 之各電壓而設置。 根據上述結構，將電壓V 0 , V 1 U , V 1 , V 1 L , V 4 ϋ , V.4 , V 4 L , V 5作為Υ電源1 0 6 ,將電壓V 0 , V 2 , V 3 , V 5作為X電源供給 於第1圖之液晶單元101。 第1圖中，108為産生修正用時鐘110之電路，産生同 步於佶號LP之時鐘。該時鐘110亦可例如將信號XSCL分 頻或倍增而形成，亦可由PLL路形成。此時，該時鐘11〇 不一定經常為相同週期，亦可例如其週期同步於倍號LP 而逐漸增大或減小。該時鐘之週期偽例如由實驗算出。 本實施例中，在倍號LP之一値週期内具有16個週期。 以上説明第1圈之液晶顯示裝置之結構。 以下説明實際上顯示棋盤圖案時之本實施例之動作。 第6圔表示第2圖之掖晶面板201所表示之顯示内容。 第6圖中，斜線部份之顯示點表示點亮部份。本例中 表示進行棋盤圖案之顯示。此時，其動作如下。 第1圖之修正電路104分別計數第2圈之液晶面板201 -24- (請先閱讀卄面之注意事項再填寫本頁) •蛑· •訂· •綠· 甲 4(210Χ 297 公沒） A6 B6 201825 五、發明説明（23) (請先聞讀背面之注意事項再填蒋本頁) 之掃描電極Y1〜Y6中，選擇Yn時，掃描電極Yn + 1與驅動 端在上方之信號電棰XI,Χ3,Χ5所形成之顯示點中，點亮 之 顯 示 點 數 巨 Ml 0 IV 輿 驅 動端在 下方之 信 號電極Χ2,Χ4, X6 所 形 成 之 顯 點 中 ♦ 點 亮 之顯示 點數目 M2 0 Ν 〇 然 後 9 在 m 擇 下 — 個 掃描電 棰 Υη + 1 之 剎那前，算 出 該 計 數 之 數 值 與 修 正 電 路 104之保持於第4 圓所示閂鎖 電 路 40 7 U 及 L之掃描電極Υ η與驅動端在上方之信號電極XI, X3 ,Χ5所形成之顯示點中， 點亮之顯示點數目HI on,舆驅 動 端 在 下 方 之 信 0^1 m 電 極 X2 , X4 , Χ6所形 成 之顯示點中 > 點 亮 顯 示 點 數 S N2 on 之 差 值 11，I 2 I 1 = Η 1 on· -Μ 1 〇 w 12 = N2on -Η 2 ο λ/ 將 之 寫 入 記 億 器 409U ,L 之η - 1位址内。 此動作從η為 1至5 • 從 5至1 反 復 的 進 行 0 因此， 在記億 器 409U , L 中， 寫 入 關 於 全 部 掃 描 電 極 Y 1 Υ6之差 11 〜12 〇 在第6圖之顯示 時 • 於 記 億 器 409 内， 依次的從0至5位址寫入-2,2, -2 1 2 . -2 ,2 1 而 於 記 憶 器 409L内依 次寫入 2， -2,2,-2,2, -2 〇 在 記 億 器 4 0 9 U » L之寫入動作之同時， 對該記億器4 0 9 U » L之η -1 位 址 之 數 值 11 ，I 2 以 Κ = 1.2, • · .6依次進行 { f ( k - mi 1 ^ f (L- K + 1 ) 12) / 4之運 算 ，將其結果 依 次 傳 送 至 第 2圖之Y 驅 動 器 内之移 位暫存 器 2 07 〇 若 η = =3時 • 將 位 址 η - 1 = 2位 址之記 億 器40 9U及L 之 數 值 -2 ,2作 為 數 值 11 ,L 2進行上述運算。 如此，在選擇另 經濟部中央橾準局印裝 甲 4(210Χ 297乂 发） A6 B6 01825 五、發明説明（24 ) (請先閱讀背面之注意事項再填寫本頁) 一餾掃描電極Y(n + 1)之剎那前，於移位暫存器207中， 從上方輸入 I = { f (0) * Il + f (5) * 12} / 4 =15 ί (-2)+10*2} / 4 tsy - 3 I = { f (l"Il+f (mi2) / 4 =14 *(-2)2+11*2 ^=5-2 I = (f(2) *Il+f(3)*I2} /4 =13^(-2)+12*2} / 4 与-1 I = {f(3) ”l+f(2)*I2) / 4 =1 2 * ( - 2 )2+ 1 3 本 2 } 1 .訂· I = { f ( 4 ) * 11 + f ( m I 2 } / 4 =11 ¢(-2)+14^2} /4 与 2 I = { f (5) *Il + f (0)* 12} / 4 =10 *(-2)2-15*2} .線· *=? 3 上述各式中，各數值I之小數點以下予以四捨五入。 經濟部中央橾準局印製 當信號LP下降，而灌擇掃描笛棰YnM時，此等數值被 輪入閂鎖路208中。在此同時，計數器209被復置成0。 -26- 甲 4(210X 297 公发） 201825 A6 B6 經濟部中央橾準局印製 五、發明說明（25) 然後，由於修正用時鐘110之作用而上昇計數。 當n=3時，被選擇之掃描霄+ l變成為掃描電棰Y4 。故位準移位器212之各两關從上方選擇S4(修正電壓（L )),S2(修正電壓（U), S4(修正電S(L)) S1 (選擇電壓），S4(修正電壓（L)), S2(修正電壓（U)), 分別供給於液晶面板201之各掃描電掻Y1〜Y6。 選擇S2至S4之各開關保持該狀態.一直到修正用時鐘 110輪入於計數器209中相酋於各對應之閂鋇電路208所 示之數目，而計數器209之輸出分別與對應之閂鎖電路 2 08所示之數目成為一致為止。成為一致之後，各開蘭 選擇S3(非選擇電壓）而將輪出。採用修正電壓（U)或（L) 僳由數值I為正或負來決定。數值I之絶對值愈大，選擇 修正霄壓取代非選擇電壓愈長之時間。 如此，將電壓供給於液晶面板201之掃描電極Y1〜Y6。 此時，若各倍號電極XI〜X6在被選擇之掃描甭極Yn + 1 上形成之顯示點點亮時，X驅動器213將點亮電壓供給 於各倍號電棰 XI〜Χ6,若為未點亮時，將未點亮電壓 供給於各倍號電棰XI〜Χ6。 以上説明本實施例之動作。第7圏（a)〜（g)表示進行第 6圖所示之顯示時施加於液晶面板201之掃描《極Y1〜Y6 及倍號電極XI〜X6之16壓波形。以下説明更詳細之動作 。第7圈U)中，以實線表示第6圔之顯示黏D31, D51之 位置之佶號16極X3,X5之電®波形，以虛線表示顯示點 t請先閱讀计面之注意事項再瑱寫本頁) •蛑. •訂. 線· 甲 4(210X 297 公沒） 201825 A6 B6 五、發明說明（26) D21, D41之位置之倍號電極X2,X4之電壓波形。第7圔（b) 表將 線即 虛示 以表 而線 ，虛 形以 波， 壓形 電波 之壓 Y1電 掻之 電Y1 描極 掃電 於描 加掃 施於 示生 表産 線將 實即 以示 ，示中 中表d) 丨線1(J _ /|\ 圖以 7 , 第形 。波 真壓 失電 形之 波Y1 之極 置電 位描 之掃D3 '1於點 D3加示 點施顯 示示於 顯表生 於線産 生實將 産以即 7 第 0 真 失 形 波 之 置 位 之 以即 線 實 生 産 將 線 實 以 示 表 線 虛 7以 ,J第. 形。 波真 壓失 電形 之波 Y1之 極置 電位 描之 掃33 ΛΖ D DS點 D 力示力 施i施 形 波 壓 霄 之 TX Y 極 tMni 描 掃 於 示中 表e) /IV ^ 0 虛 以 經濟部中央抹準局印焚- 卽將産生於顯示點D34之位置之波形失真。第7圓（f)中 ,以實線表示施加於動作霄極Y1之電壓波形，以虛線表 示即將産生於顯示點D35之位置之波形失真。第7圏中（g) 中，以實線表示施加於掃描電極Y1之電壓波形，以虚線 表示即將産生於顯示點D36之位置之波形失真。由第7圓 (a)〜（g)可知，依照掃描電極Y1,Y2，Y3之順序，位於愈 上方之電極受到第7圖（a)中實線所示之霣壓波形變化之 影堪愈大，而在掃描電極Y1,Y2,Y3上，朝向該電S波形 變化之方向産生配合該影辑之大小之波形失真。同樣的 ，依照位於下方之掃描電極Υ6,Υ5,Υ4之順序，從下方受 到供給倍號電壓波形之倍號電極Χ2,Χ4之之電壓波形之 變化之影響，而發生其影壜之大小之波形失真。如第7 圖（b)〜（g)中*線所示之波形，第2圔中之Υ驅動器205 以方向與即將産生於各掃描堪掻Y1〜Y6之波形失真之方 向相反之修正電K(U)或（L)取代非選擇笛壓，針對即將 産生於各播描霣極〜Y6之波形失典而輸出。輪出之時 (請先聞讀背面之注意事邛再填寫本頁) ，訂. .線. 甲4(210乂297 7泠） 201825 A6 B6 經濟部中央搮準局印裂 五、發明說明（27) 間配合各掃描電極Y1〜Y6上即將産生之波形失真之大小 而增減。換言之，對較大之失真供給較長時間之修正電 壓，對較小之失真供給較短時間之修正電壓。然後，供 給非選擇電壓。 因此，可把即將産生於各掃描電極Y1〜Y6之波形失真 大致上予以消除。因此，施加於液晶面板201之各顯 示點之實效電壓無差異，故可消除顯示之不均勻。 如上所述，在計算被選擇之掃描電極與下一値被選擇 之掃描電極上之點亮數目之差值時考慮顯示園型舆將驅 動電壓波形驅動之端部位置，而且根據該差值使各電掻 上之施加於掃描電極之修正電壓之量，即可消除顯示之 不均勻。 〔實施例 2〕 實施例1中，偽説明其倍號電極為交替的從上下方供給信 號電E波形之液晶面板。但只從一方供給時亦同樣的有 效。例如只從上方供給時，只要如第4圖所示.以經常 為”1”之倍號取代T-F/F401之輸出供給於閘«路402U及 L即可。（此表示與使各電路402,406L〜410L成為非動作 狀態相同。因此，不需要此等霄路。與此附帶的，電路 405L亦不需要，而且411L所産生之輪入經常成為〇,故 又可簡化電路412)。因此，數值I為 I =(11 * 12)* f(k) 根據此數值I進行與實施例1相同之動作，即可得相 -29- (請先聞讀背面之注意事項再填寫本頁) _玫· ，訂· .線. 甲 4(210X 297 公发） 201825 A6 B6 經濟部中央抹準局印装 五、發明説明（28) 同之效果。 〔實施例 3〕 實施例1,2中像説明只從信號霣極之一方供給信號 電壓波形之液晶面板，但從兩端供給時，亦同樣的有效 〇 在實施例1中計算數值I時，只要以函數g(l K-L/2 1 ) 取代函數f(k)進行計算邸可。此時，函數g(X)為數值X 愈大時愈增大之增加函數。 根據該數值I進行與實施例1相同之動作，即可得相 同之效果。 實施例1〜3中，調整修正量之方法，偽使用將修正電 壓舆非選擇電壓之電壓差保持為一定而增滅施加修正電 壓之時間之方法（以後稱之為時間軸修正）。但亦可將施 加修正霣壓之時間保持一定，而改變修正霣壓與非選擇 電IS之差值（此動作即為電壓軸修正）。亦可改變施加修 正霣壓之時間及霣壓（此動作即為時間電K軸修正）。亦 可將波高值配合所需之修正量發生變化之指數函數所表 示之波形及三角形之波形作為修正霣K (此動作即為函 數波形修正）。 〔實施例 4〕 以下説明横向拉線所造成之顯示不均勻之實施例。 如上所述，在掃描®極Y1〜Ys之被灌擇之掃描笛棰中 ，該掃描霣極與各佶號霜極Xp(p = 1,2.....r)所形成之 -30- (請先閲讀背面之注意事項再填寫本页) •蛑. .訂. -線. 甲 4(210X 297公沒） 201825 A6 B6 五、發明説明（29 ) 顯示點之位置為P時，將發生配合由 6 Z ' = Σ q (i)* δ (i) i = 1 所計算之數值Z’之横向拉線所造成之顯示不均勻。換言 之，施加於對象之掃描電極上之顯示點之實效電壓，配 合該數值Z’而變小。因此，只要一方面使液晶顯示裝置 動作而計算該數值Z’，進行配合該數值Z’修正即可。以 下說明其具體實施例。第8圏表示本實施例之結構。第 8圖中，102,103係與第1圖中之控制倍號，資料倍號 相同，故以相同符號表示而省略其説明。801為液晶單 元，俗由液晶面板201,掃描電極驅動電路805及信號霣 極驅動轚路213所構成。804為電壓波形修正笛路（以後 稱修正電路）計算數值Z'而産生只在配合該數值之長度 之時間内成為有源之修正佶號809,805為霣源《路，806 為驅動掃描電棰用之從霜源電路805输出之2組拥描霉掻 用之電源（由選擇電K及非選擇電壓所構成，以後稱Y霄 源）。電源堪路805在修正倍號809為有源時與非有源時 將互不相同之電壓作為Y電源806之選擇電壓予以輪出。 106為用來驅動信號電棰之從電源電路805输出之2組倍 號甭掻用之電源（以後稱）（笛源），與第1圔中之X霣源 107相同。809為用來決定從修正霣路804輪出之修正笛 (請先閱讀背面之注音V事項再填寫本頁) 訂· •線· 經濟部中央橾準局印製 示 表 圈 9 有 0 例 體 具 之 素 要 成 構 各 。之 號圈 倍 8 正第 修明 之説 量下 之以 壓 甲 4(210X 297 公;%*) 201825 A6 B6 經濟部中央抹準局印製 五、發明説明（30) 液晶單元801之具醴結構例。圖中201為液晶面板，與實 施例1之液晶面板相同。805為掃描電極驅動電路（以後 稱Y驅動器），傺由移位暫存器206,轉換電路911,及位 準移位器912所構成。位準移位器912之輸出被導引至液 晶面板201之各掃描電極Y1〜Y6。移位暫存器206之結構 與實施例1相同，故其說明從略。開關電路911從第8圖 之電源電路804所産生之Y電源806之2組電壓中，依照信 號FR選擇1锢組。亦即將構成Y電源806之4値電壓V0’， V 1 , V 4 , V 5 ’中，電壓V 0 | , V 4分成第1電壓組，電壓V 5 ’， VI分成第2電壓組，而依照佶號FR轉換成該2組之遴擇電 壓，非選擇電壓。同樣的，將電壓V5’，VI稱為第2電壓 組之選擇電壓，非選擇電壓。轉換電路911所選擇之2組 電壓中之任一組電壓被供給於位準暫存器912。 912為位準暫存器，偽由多舾2回路1接點之開闋所構 成，而且根據移位暫存器206之内容轉換。 換言之，當其内容為”1”時，各開關選擇該選擇霄壓 ，而將之供给於對應之各掃描電棰Y1〜Y6。 當其内容為”0”時，各開鼷遘擇該非選擇電壓，將之 供給於對應之各掃描笛棰Y1〜Y6。 以上説明Y驅動器905之結構。以下説明其動作。將 倍號DIN同步於佶號LP輸入移位暫存器206並且將之傳送 。結果.位準移位器912中之一値開關與其成為對應的 依次産生S揮霣壓。其餘之閭關則産生非選擇笛壓以上 (請先M讀背面之注意事項再填寫本頁) •装· .訂· •線· 甲 4(210X 297 公发） 經濟部中央抹準局印製 201825 A6 _ B6 五、發明説明（31) 説明Y驅動器之結構及動作。213為倍號電棰騮動電路（ 以後稱X驅動器），因為舆實施例相同，其說明從輅。 以上為掖晶單元801之結構。液晶單元801之掃描霄極 Y1〜Y6上，同步於倍號DIH, LP施加選擇電壓，然後， 與其成為同步的，在倍號電棰XI〜X6上施加配合顯示圏 型之點亮或未點亮電S而使液晶面板201進行顯示。 以上說明液晶單元801之動作。第8圖中，804為修正 電路，在掃描電極Y1〜Y6中，以下一個被選擇之掃描電 極與各信號電極Xp(p=1,2, ...,6)所形成之顯示點之位 置作為P ,計算 6 Z’ = Σ q (i)*S(i) ..... (1) i = 1 然後，在選擇下一個揷描霄極時，將只在配合於該數值 Z’之長度之時間内成為有源之修正信號809同步於倍號 LP予以輸出。第10圖表示具體之詳細結構。第10圔中， 1 0 01為計數器。1 0 0 2為常數産生電路。1 0 0 3為産生表示 常數産生電路1002所輸出之數值之多個輪出之各倍號與 資料信號103之邏輯積之閘極電路。1004為進行加法 之運》笛路。1 005為保持運算笛路1 005之運算結果之第 1閂鎖電路。1006為第2閂鎖電路。1007為將只在配合於 該第2閂鎖笛路之内容之時間内成為有源之修正倍號809 同步於佶號LP予以粬出之修正倍號産生霣路（以後稱産 生電路）。 -33- (請先閲讀背面之注意事項再填寫本頁) .蛑. •訂· •線. 甲 4(210X 297乂发） 經濟部中央抹準局印裝 201825 A6 B6 五、發明説明（32) 計數器1001被信號LP復置為0,而由於倍號XSCL之作 用 進 行 上 昇 計 數 » 其 輪 出被供 給 ‘於常 數 産生 電 路 1 002 〇 常 數 産 生 電 路 1 0 0 2 僳 由ROM及 二極體矩陣等所構成， 而 且 配 合 計 數 器 100 1 之 輸出， 産生不 相 同之 數 值 〇 換言 之 > 計 數 器 1001 之 輸 出 值增大 時，亦 随 著産 生 較 大 之输 出 〇 其 數 值 對 應 於 (1 )式之函數 q ( i ) 〇 其中， i 為 在 計數 器 1001 所 産 生 之 數 值 上 加 1之 數值。 函 數 之 數 值 亦 可 由 實 驗算出 〇 本實 施 例中 » 如 下 的表 示 〇 1 = 1 時. q (1)= 1 i — 2 時， q (2)= 1.1 i = 3 時， q (3 )= 1.2 i = 4 時， q (4)= 1.3 i = 5 時， q (5)= 1.4 i = 6 時， q (6 )= 1.5 蘭 霣 路 1 003 産 生 常 數 産生霣 路1002所 输出 之 數 值 與資 料 佶 號 1 0 3之邏輯積。 換言之， 在某- -信號XSCL下降時 若 資 料 佶 號 為 有 源 "1 1 則輪出 常數産生電路1 002之數 值 * 若 為 非 有 源 "0 ，，時， 則産生 0 〇此 數 值對 應 於 (1)式 中 之 q ( i) ♦ 5 ，ί ) 〇 其 中 ,ί為倍 號LP下降後， 信號XSC之 下 降 次 數 I 相 笛 於 在 計 數器100 1所産 生 之數 值 上 加 1之 數 值 〇 1 00 4 為 運 算 電 路 t 將 閘極霣 路所産 生 之數 值 與 第 1閂 (請先閱讀背面之注意事項再填寫本頁) •蛑· ’訂· •綠. 甲 4(210X 297 父沒） 經濟部中央採準局印欠* 201825 A6 B6 五、發明説明（33) 鎖電路1005之内容同步於XSCL予以相加，並將其結果送 回第1閂鎖電路1 0 0 5。 1005,1006為第1,第2閂鎖電路。第1閂類電路1005被 信號LP復置。被復置之刹那前之閂鎖電路1005之内容對 應於下式之數值Z’。 6 Z’ = Σ q (i)ί δ ( i) .... (1) i = 1 換言之，將下一匍選擇之掃描電極上之點亮點之數目， 配合點亮點之位置予以加權而計數之數值。 其内容在第1閂鋇電路被復置之剎那前，於信號LP之 下降時被輸入第2閂錤電路1 0 0 6中，此時，在倍號L P下 降時選擇下一個掃描電極，故第2閂鎖電路1006所表示 之數值即為被選擇之掃描電極上被加權之點亮點數。 1007為將只在配合於第2閂鋇電路1006所示之數值之 時間長度内成為有源"1”之修正倍號809,同步於倍號LP 予以輪出之産生電路。 該電路1007可由：産生將倍號XSCL直接，或分頻或倍 增之時鐘（以後稱修正用時鐘）之掀通電路1008;由於該 修正用時鐘而進行上昇計數，被信號LP愎置成0之計數 器1009;産生有源”1"之信號，一直到表示計數器1009 之輪出之數值與第2閂鎖電路所表示之數值成為一致為 止之一致檢測電路1010很容易的形成。修正用時鐘之遇 期可不必為相等間隔，例如可由實驗來決定。 以上說明修正笛路之結構。其次，將被遘擇之描描電 -35- (請先聞讀背面之注意事項再填駕本頁) .¾. •訂· •綠· 甲 4(210X 297父沒） 201825 A6 B6 五、發明説明（34) 極上之點亮點數目予以加權而計數，而在下一傾被選擇 之掃描電極選擇只在配合該數值Z'之長度之内成為有源 之修正佶號809時，同步於信號LP予以输出。 以下參照第11圖説明第8圖之電源笛路805。第11圖 表示電源霄路805之具體結構例。圏中，1101〜1107為 串聯電阻。其兩端施加有電壓V0U.V5L。此等電阻1101 〜1107構成分壓電路，瘕設産生於各電阻1101〜1107之 電壓為從上方算起V0U,V0,V1,V2,V3,V4,V5,V5L,則 V ® V0-V1 =V1-V2 =V3-V4 =V4-V5 V2-V3 = a#V,其中a為1〜50左右之常數。 各電阻1101〜1107之電阻值係設定在可成立下式之數值。 V0U-V0 = V5-V5L。 510為第1圖之電壓穩定電路510相同，故其説明從略 〇 1108,1108為轉換電路，在修正倍號809成為有源”1” 經濟部中央揉準局印裝 (請先聞讀背面之注意事項再填寫本頁) 時，邇擇®壓V0U,成為非有源”0"時，遘擇霣壓V0。然 後，將選擇之18®設定為笛壓V0’。同樣的，轉換電路 1109在修正佶號809成有有源”1"時，蒱擇電壓5L,而在 成為非有源"0"時選擇笛KV5。然後，將選擇之霣壓設 定為電壓V 5 ’。在此，將霣K V 0 U , V 5 L稱之為修正笛壓。 甲 4(210X 297 公发） 201825 A6 B6 經濟部中央橾準局印製 五、發明説明（35) 將電壓V0’，V4,V5',V1作為Y電源106,將電源V0，V2, V3,V5作為X電源而將之供給於第8圈之液晶軍元801。 第8圈之液晶顯示裝置具有如上述結構。 以下說明實際上顯示四角形時之本實施例之動作。第 12画及第13圖為表示第9圓之液晶面板201所顯示之顯示 内容之圈。圖中，斜線部份之顯示點表示點亮。本例中 ,第12圖及第13圖表示在靠左及靠右倒分別顯示相同之 四角形。其動作如下。 首先，在進行第12圖或第13圖之顯示時，第8圜之修 正電路804將第12圖或第13圖之液晶面板201之各掃描電 極Y1〜Y6上之黏亮點數目，在點亮點之位置加權而計數 。亦即計算數值Z’。各掃描電極Y1〜Y6之數值Z’如下。 顯示第12圖之顯示時 掃描電棰 Y1..... Z ' = 〇 + 〇 + 〇 + 〇 + 〇 + 〇 =0 掃描電棰 Y2..... Z， = 1 , 0*1 + 1 . 1*1 + 0 + 0 + 0 + 0=2.1 捅描鬣極 Y3..... Z' = 1.0^1+1.1^1+0+0+0+0 =2 . 1 掃描霄極 Y4..... (請先閱訪背面之注意事項再填寫本頁) •襄· 訂· 緣. 甲 4(210X 297 公发） 經濟部中央抹準局印裂 201825 A6 B6 五、發明説明（36) Z ' = 1. 0^1 + 1. 1^1 + 0 + 0 + 0 + 0 =2.1 掃描電極 Y5..... Z ' = 1 . 〇s：l + l . 1*1+0 + 0 + 0 + 0 + 0 =2.1 掃描罨棰 Y6..... Z ' = 〇 + 〇 + 〇 + 〇 + 〇 + 〇 =〇 顯示第13圖之顯示時 掃描雷棰 Y1..... Z ' = 〇 + 〇 + 〇 + 〇 + 〇 + 〇 =0 播描電極 Y2..... z，= 0 + 0 + 0 + 0 + 1,4*1 + 1，5*1 =2.9 掃描霄極 Y3.....Z' = 〇+〇+〇+〇+1.4-1+1.5*1 =2.9 掃描霣極 Y 4.....Z' = 0+0+0+0+1.4^1+1.5^1 =2.9 掃描霄搔 Y5..... z · = 0 + 0 + 0 + 0 + 1 . 4^1 + 1 .5*1 (請先閏讀背面之注意事項再填寫本頁) .打. •線. 甲 4(210X 297 公发） A 6 B6 201825 五、發明説明（37) =2.9 (請先閱讀背面之注意事項再填寫本頁) 掃描電棰 Y6..... Z' = 〇+〇+〇+〇+〇+〇 =0 第8圖之修正霄路804只輸出配合數值Z’之長度之時 間之有源修正倍號8 0 9。 電源電路805只輪出修正信號809成為有源之時間之修 正電壓V0U,V5L作為取代電壓V0,V5之選擇電壓。 •打. 顯示第12圖之顯示時，掃描電極Y2〜Y5之數值Z’小於 顯示第13圖‘之顯示時之數值。因此，顯示第12圖之顯示 時，若正在灌擇掃描〜Y5,則修正電壓V0U,V5L 成為選擇電壓.施加之時間，短小顯示第13圖之顯示時 之時間。第9圖之Y驅動器905利用這種配合數值Z’而變 化之選擇電壓，依次將選擇電壓施加於液晶面板201之 各掃描電極Y1〜Y6。 Y驅動器213之動作舆實施例1相同，其説明從略。 •綠· 經濟部中央橾準局印裝 進行以上之動作時，係如第12圖顯示例所示，若點亮 點在掃描電極Y2〜Y5之驅動端附近，則配合於該掃描® 極上之電壓波形之失宾不太大，將修正電壓之量較少之 蒱擇電壓，供給於被選擇之拂描電極。換言之，將施加 時間較短之笛壓作為灌擇笛®，取代修正箱V0U,V5L為 堪壓V0,V5供給於踅極上。相反的，如第13圖之顯示例 所示，若點亮點離開掃描笛棰Y2〜Y5之驅動端較速時， 甲 4(210X 297公发） A6 B6 201825 五、發明説明（38) (請先聞讀背面之注意事項再填寫本頁) 則配合其掃描電極上之電壓波形之失真較大，將修正電 壓之量較多之選擇電壓供給於被選擇之掃描電極。換·言 之。将施加時間較長之電壓作為選擇電壓，取代修正霣 壓V 0 U , V 5 L成為電壓V 0 , V 5之電壓供給於該電極。因此， 可根據第12圖及第13圖之之顯示位置，大略的修正修正 量之差異。如上所述，在計數各掃描電極上之點亮點數 目時，考慮點亮點之位置而進行加權並予以計數，依據 其結果變更供給於各電棰之選擇電壓。即可消除顯示不 均勻。 〔實施例5〕 訂. 實施例4中係説明只從掃描霜極之一方供給掃描電壓 波形之液晶面板但從兩端供給時亦同樣的有效果。換言 之，實施例4中，在計算數值Z'時，只要以凾數 P(| i -S/2| )取代函數q(i)進行計算即可，函數p(x) 為數值X增大時滅小之减小函數。 根據數值Z'進行如實施例 4所述動作，卽可産生相同 效果。 .線· 實施例4中，修正量之調整方法傺使用保持霣壓V0U, V 0之差，® β V 5與V 5之差為一定而增滅電壓V 0 U , V 5 L成 為選擇霣壓施加之時間之方法，亦即時軸修正方法。其 他亦可使用笛壓軸修正，時間笛壓軸修正，函數波形修 正法。 經濟部中央橾準局印製 實施例1至2,3係改變非選擇ΙΒΚ,而實施例4及5偽改 一 4 0 一 甲 4(210Χ 297 公发） 201825 A6 B6 五、發明說明（39) 此 因 法 方 正 修 之 均 壓不 霄顯 擇 之 0 ^ 變例 施 實 及 1* 例 施 實 用 利 行 遣 時 同 可 置文 裝及 示像 顯圖 晶該 液盧 於考 含時 包同 出 ’ 取時 在則 ，規 明之 發中 本字 照文 依及 , 像 述園 所之 上示 如顯 所 關波 置壓 位電 之號。 端倍勻 極舆均 各形不 之波著 型壓顯 波電善 壓描改 電掃的 動變著 驅改顯 給俱可 供藺此 之該藉 之據， 板根方 面 ，一 晶出少 液取至 輿而之 字係形 置 裝 之 例 施 實 IX 0 之 置 裝 示 頃 晶 液 之 明 發 本 為 圖 • · 一—1 式第 圖 圖第 之 0 構圖 結塊 之方 路構 電結 第動之 0 ) 08 ί 路 0 0 0% S- ^ 1 H 正 力®眵 之掃路V 構示電正 結表修 為 圖 3 第 ο 圖 之 構 結 之 元 單 晶 液 為 圖 2 形結 波之 壓路 Ε 霞 為源 圖電is 第圖 視 透 之 板 面 晶 掖 之 例 之 板 面 晶 掖 之 時 示 顯 之 一圖 之 6 容第 内行 示進 顯為 示g) /|\ 為} 圖(a 6 圖 第 7 ο 負 圖 。 構圖 δ 圖 圖 9 形第 波 。 壓圖 電之 加構 施結 第 為 置 裝 之 例 施 實 4 0 第圖 明構 發結 本之 示元 表單 為晶 圖液 _15 其 修 為 圖 ο 11 0 第圖 3 ο 1 圖第 構。 結圔 之視 路透 電之 源板 電面 為晶 圖液 11之 第例 。之 圈容 塊内 方示 之顯 構他 結其 路為 電圖 正12 圖 。 14圖 第視 。透 圖板 視面 透晶 板液 面之 晶例 液施 之實 例一 一 之 之容 容内 内示 示顯 顯之 一 術 9技 示用 表習 為為 (請先閱讀背面之注意事項再填駕本頁) k. .訂· _線· 經濟部中央橾準局印裝 於一之 加之例 施容一 上内之 際示容 實顯内 ,一 示 時9顯 示為一 顯圖另 之16為 圖第圖 4 7 1Χ ο 11 第圖第 行之 。 進形圖 為波視 C)壓透 ^ ffi- ^ )~之 1B (a板晶 圖面液 15晶之 第液例 甲 4(210X 297 公发） 201825 A6 B6 五、發明説明（40) 液晶面板透視圖。 101,108...液晶單元。102...控制信號。103...資料 倍號。104,804...電壓波形修正電路（修正電路）。 8 05...電源電路。1 06,806.. .Y 電源。107...X 電源。 108,..産生同步於倍號LP之時鐘信號之電路。109... 修正信號。110...時鐘信號。 (請先聞讀背面之注意事項再填寫本页) .故· •訂. .線. 經濟部中央橾準局印裝 42- 甲 4(210X 297 公发）Pole S- Describe the obedience is L fixed number K)-1 "The minus f item is produced by the J number y /» Use (please read the precautions on the back before filling in this page) Printed by the Central Bureau of Standards of the Ministry of Economy η Υ 饣 霣 demonstrative 1Χ ο 2 plate crystal liquid circle 0 2 = 5u L in-plus in the middle of the implementation of the pressure to choose the preferred choice. In the giant armor 4 (2lOX 297 public) 201825 A6 B6 Fifth, 哂During the period of Mengmeng (20}, the multiplication of Figure 4 is performed. 411U and L respectively perform the content and multiplication of the 4 billion UU and L at the address (n-1 address) indicated by the counter 403. The multiplication of the number generation circuit 405U, L. At this time, the counter 404 counts up due to the multiple number YCSCL, so the value generated by the multiplier generation circuit 405U and L changes in synchronization with the multiple number YCSCL. In other words, the arithmetic circuit 411U and The operation results of L are synchronized with the signal YCSCL to become f (0) a »I 1, f (5) • a. 12 f (1)« VI 1, f (4) 12 f (2) I 1, f ( 3)% 12 f (3)% I 1, f (2) Ϊ * 12 f (4) * I 1, f (1) 12 f (5) I 1, f (0) Λ, 12 (please listen first (Notes for reading the bamboo noodles and then fill out this page) Printed by the Central Wiring Bureau of the Ministry of Economic Affairs Then, the arithmetic circuit 412 for addition and division adds the results, and then divides the result by 4. If the value is I, then I = {f (0) * Il + f (5) M2) / 4 I = {f (l) * Il + f (4) M2} / 4 I = {f (2) * Il + f (3) * I2) / 4 I = {f (3) * II (2 ) * 12} / 4 I = {f (m Π + f (m 12} / 4 I = {f (5) ill + f (0) ^ 12} / 4 in turn synchronized to the YCSCL signal output. Divided by 4 The reason is that if the sign signal is removed from the shift register 207 of the third ffl, it becomes a 4-bit structure, so that the value I is within the 4-bit. Therefore, it is not essentially determined -22- * order. , Line. A 4 (2_] 0X_2g7C ^) 201825 A6 B6 Five'invention description (21). During the period of selecting the broadcast electrode Υη of the second crystal panel 201, change fU-1), f ( Ln + l) is multiplied by the difference between the number of lighting points on the poles Υη and Υη + 1, II, 12, and the result is added to the value 1 = f (nl) * Il + P (Ln + l ) = U2 synchronizes with the YCSCL multiple times and sequentially turns out. The magnitude of the result is an intensity multiplier of 10 to 13, positive and negative as a symbol multiplier F, and together with the signal YCSCL, it is output as the correction signal 109 of FIG. The structure and operation of the correction circuit 104 are described above. Its structure is not limited by the above embodiment. For example, in the fourth ring, Ao takes 1 = f (n-l) * Il + F (L-n + l) iI2 using 405U and L, 411U and L, 412 and other circuits to calculate in real time. However, it can also be calculated in advance by the CPU, etc., and written into the reset megameter, and the rounds of the counters 403 and 404 are sequentially read out as the address and output as the correction multiple number 109. The fifth ring represents an embodiment of the structure of the high-frequency source circuit 105 shown in the first group. In the meantime, 501 ~ 509 are series resistance. On both ends, 霣 ®V0, V5 is applied. These resistances 501 to 509 are formed as partial pressure paths. Assuming that the voltages generated from the respective resistances 501 to 509 are set to V0, V1U, V1, V1L, V2, V3, V4U, V4, V4L, V5 from above, then V = V0-V 1 = VI -V2 = V3-V4 = V4-V5 Printed by the Central Bureau of Economic Affairs of the Ministry of Economy ¾. (Please read the precautions on the back before filling out this page) • Order ·. 蝝 · V2-V3 = aiV, where a is a constant and its value is about 1 ~ Around 50-23- A4 (210X 297〇 *) 201825 A6 B6 Printed by the Central Approval Bureau of the Ministry of Economic Affairs V. Description of Invention (22). Each resistance value 501 to 509 is set to a value that can satisfy the following four formulas. V1U-V1 = V4-V4L V1-V1L = V4U-V4 510 is a voltage stabilizing circuit that does not change the voltage values formed by resistors 501 ~ 509 and only reduces its impedance. Like voltage followers formed by operational amplifiers, emitter followers formed by transistors, etc. As shown in the figure, the voltage stabilizing circuit 510 is provided for each voltage divided by the resistors 501 to 509. According to the above structure, the voltages V 0, V 1 U, V 1, V 1 L, V 4 ϋ, V.4, V 4 L, V 5 are used as the power supply 1 0 6, and the voltages V 0, V 2, V 3. V 5 is supplied as X power to the liquid crystal cell 101 in FIG. 1. In Fig. 1, 108 is a circuit that generates a correction clock 110, and generates a clock synchronized with the LP number. The clock 110 may be formed by dividing or multiplying the signal XSCL, for example, or may be formed by a PLL circuit. At this time, the clock 110 does not always have to have the same period, and for example, its period may gradually increase or decrease in synchronization with the multiple LP. The period pseudo of the clock is calculated by experiment, for example. In this embodiment, there are 16 cycles in one value cycle of the multiple LP. The structure of the liquid crystal display device of the first circle is explained above. The following describes the operation of this embodiment when the checkerboard pattern is actually displayed. The sixth frame shows the display content indicated by the yoke panel 201 in FIG. 2. In Fig. 6, the display points of the oblique lines indicate the lit parts. In this example, it means to display the checkerboard pattern. At this time, the operation is as follows. The correction circuit 104 in Figure 1 counts the LCD panels 201 -24- of the second circle respectively (please read the precautions before filling in this page) • Insect · • Order · • Green · A 4 (210Χ 297 public) A6 B6 201825 5. Description of the invention (23) (please read the precautions on the back and fill in this page). When Yn is selected in scan electrodes Y1 ~ Y6, the scan electrode Yn + 1 and the drive terminal are above the signal power. Among the display points formed by XI, Χ3, Χ5, the number of illuminated display points is huge Ml 0 IV The display points formed by the signal electrodes Χ2, Χ4, X6 with the drive terminal at the bottom ♦ The number of illuminated display points M2 0 Ν 〇 and 9 before the selection of m-a scanning electrode Υη + 1, the count value is calculated and the correction circuit 104 keeps the scanning electrode Υ held by the latch circuit 40 7 U and L shown in the fourth circle η and the driving end are in the display points formed by the upper signal electrodes XI, X3, and X5, the number of lit display points is HI on, and the driving end is at the bottom of the signal 0 ^ 1 m formed by the electrodes X2, X4, and Χ6 Displaying points > Light up and display points S N2 on Difference value 11, I 2 I 1 = Η 1 on · -Μ 1 billion w 12 = N2on -Η 2 ο λ / write control of the memories 409U, η L - A 1 address. This action is from η to 1 to 5 • Repeatedly from 5 to 1 0 Therefore, in the megameter 409U, L, write the difference 11 to 12 for all the scan electrodes Y 1 Υ6. When shown in Figure 6 • In the memory 409, write -2,2, -2 1 2 from 0 to 5 in order. -2, 2 1 and write 2, -2,2,-in memory 409L 2,2, -2 ○ At the same time of writing to the register 4 0 9 U »L, for the value 11 of the η -1 address of the register 4 0 9 U» L, I 2 with K = 1.2, • · .6 Perform {f (k-mi 1 ^ f (L- K + 1) 12) / 4 operations in sequence, and transfer the results to the shift register in the Y driver in Figure 2 2 07 〇 If η = = 3 • Use the value of mnometer 40 9U at address η-1 = 2 and the value -2, 2 of L as the value 11, and perform the above calculation. In this way, after selecting another Central Ministry of Economic Affairs, the Ministry of Economic Affairs printed armor 4 (210Χ 297 异 发) A6 B6 01825 V. Description of the invention (24) (please read the precautions on the back before filling in this page) One scan electrode Y (n Before the moment of + 1), enter I = {f (0) * Il + f (5) * 12} / 4 = 15 ί (-2) + 10 * 2} from the shift register 207 / 4 tsy-3 I = {f (l " Il + f (mi2) / 4 = 14 * (-2) 2 + 11 * 2 ^ = 5-2 I = (f (2) * Il + f (3 ) * I2} / 4 = 13 ^ (-2) + 12 * 2} / 4 and -1 I = {f (3) ”l + f (2) * I2) / 4 = 1 2 * (-2) 2+ 1 3 Ben 2} 1. Order · I = {f (4) * 11 + f (m I 2} / 4 = 11 ¢ (-2) + 14 ^ 2} / 4 and 2 I = {f ( 5) * Il + f (0) * 12} / 4 = 10 * (-2) 2-15 * 2}. Line · * =? 3 In the above formulas, each value I is rounded below the decimal point. Economy When the signal LP drops and the scan flute YnM is selected, these values are rounded into the latch circuit 208. At the same time, the counter 209 is reset to 0. -26- A4 (210X 297 public announcement) 201825 A6 B6 Printed by the Central Bureau of Economic Affairs of the Ministry of Economy V. Description of Invention (25) Then, due to the function of the correction clock 110 When n = 3, the selected scan voltage + l becomes the scan voltage Y4. Therefore, the two levels of the level shifter 212 select S4 (correction voltage (L)) and S2 (correction voltage ( U), S4 (correction voltage S (L)), S1 (selection voltage), S4 (correction voltage (L)), S2 (correction voltage (U)), which are respectively supplied to the scanning electrodes Y1 ~ Y6 of the LCD panel 201 Select each switch of S2 to S4 to maintain this state. Until the correction clock 110 is cycled in the counter 209, corresponding to the number shown by the corresponding barium circuit 208, and the output of the counter 209 is respectively latched Until the numbers shown in circuit 2 08 become coincident. After becoming coincident, each Kailan selects S3 (non-selected voltage) and will take turns. The correction voltage (U) or (L) is used. The value is determined by whether the value I is positive or negative. The larger the absolute value of the value I, the longer the time for selecting the corrected voltage instead of the non-selecting voltage. In this way, the voltage is supplied to the scanning electrodes Y1 ~ Y6 of the liquid crystal panel 201. At this time, if the multiple-number electrodes XI ~ X6 are in When the display point formed on the selected scanning electrode Yn + 1 is lit, the X driver 213 supplies the lighting voltage to each multiple XI~Χ6, if unlit, the non-lighting voltage is supplied to each of the electrical problem, No. times XI~Χ6. The operation of this embodiment has been described above. The seventh coils (a) to (g) show the 16-voltage waveforms of the scan electrodes Y1 to Y6 and the multiplier electrodes XI to X6 applied to the liquid crystal panel 201 when the display shown in FIG. 6 is performed. More detailed operations are explained below. In circle 7), the solid line represents the 6th pole X3, X5 electrical waveform of the 6th pole of the display sticky position D31, D51, and the dotted line represents the display point t. Please read the notes before counting. (Write this page) • Insect. • Order. Line • A 4 (210X 297 public) 201825 A6 B6 V. Description of the invention (26) Voltage waveforms of the electrodes X2, X4 at the position of D21, D41. 7th (b) The table shows that the line is imaginary, and the line is imaginary, and the wave is imaginary, and the pressure is Y1. That is to say, in the middle table d) 丨 Line 1 (J _ / | \ Figure 7 with the 7th shape. The wave Y1 of the true voltage loss-shaped wave Y1 is polarized with the potential sweep D3 '1 added at point D3 The dotted display is shown on the display surface. The real line will produce a 7th 0th true missing wave. The set real line will produce a real 7th line. The voltage-shaping wave of the voltage-shaping wave Y1 is polar 33. The ΔZ D DS point D shows the force and the force is applied. The TX Y pole tMni of the shape-forming wave is scanned in the table. E) / IV ^ 0 Printing and burning by the Central Bureau of Correction-the distortion of the waveform generated at the position of the display point D34. In the seventh circle (f), the solid line represents the voltage waveform applied to the action pole Y1, and the broken line represents the waveform distortion that will occur at the position of the display point D35. In the seventh circle (g), the solid line represents the voltage waveform applied to the scan electrode Y1, and the broken line represents the waveform distortion to be generated at the position of the display point D36. From the 7th circle (a) to (g), it can be seen that in the order of the scan electrodes Y1, Y2, Y3, the electrode located above is more affected by the change in the waveform of the swell pressure shown by the solid line in FIG. 7 (a). Large, and on the scanning electrodes Y1, Y2, Y3, a waveform distortion matching the size of the album is generated in the direction of the change of the electrical S waveform. Similarly, in accordance with the sequence of the scan electrodes Υ6, Υ5, Υ4 located below, the waveform of the size of the shadow is generated from below by the influence of the change of the voltage waveform of the multiplier electrodes Χ2, Χ4 supplied to the doubling voltage waveform distortion. As shown by the waveform shown by the * line in Figures 7 (b) ~ (g), the Y driver 205 in the 2nd sine is corrected in the direction opposite to the direction of the waveform distortion that will be generated in each scan Y1 ~ Y6. (U) or (L) replaces the non-selected flute pressure, and outputs it for the waveform code that will be generated at each broadcast scan pole ~ Y6. At the time of the round (please read the precautions on the back and fill in this page), order. Line. A 4 (210 ~ 297 7 Ling) 201825 A6 B6 The Central Bureau of Economic Affairs of the Ministry of Economic Affairs has printed five. The description of the invention ( 27) Increase and decrease according to the magnitude of the waveform distortion to be generated on each scan electrode Y1 ~ Y6. In other words, a longer time correction voltage is supplied to a larger distortion, and a shorter time correction voltage is supplied to a smaller distortion. Then, the non-selected voltage is supplied. Therefore, it is possible to substantially eliminate the distortion of the waveform to be generated in each of the scan electrodes Y1 to Y6. Therefore, there is no difference in the effective voltage applied to each display point of the liquid crystal panel 201, so that the display unevenness can be eliminated. As described above, when calculating the difference between the number of lights on the selected scan electrode and the next selected scan electrode, the position of the end portion where the display voltage is driven by the driving voltage waveform is considered, and according to the difference The amount of correction voltage applied to the scan electrodes on each electrode can eliminate display unevenness. [Embodiment 2] In Embodiment 1, a liquid crystal panel in which the double-numbered electrodes are alternately supplied with signal electric E waveforms from above and below is assumed. But it is also effective when supplied from only one party. For example, when supplying from the top only, as shown in Figure 4, it is sufficient to replace the output of T-F / F401 with the multiple of "1", which is always supplied to gates 402U and L. (This means the same as making each circuit 402, 406L to 410L in a non-operating state. Therefore, such a small road is not required. Incidentally, the circuit 405L is also unnecessary, and the round-off generated by 411L often becomes 0, so again The circuit 412 can be simplified. Therefore, the value I is I = (11 * 12) * f (k) According to this value I, perform the same operation as in Example 1, you can get the phase -29- (please read the precautions on the back before filling this page ) _ 梅 · ， 定 ·. 线. A4 (210X 297 public issue) 201825 A6 B6 Printed by the Central Approval Bureau of the Ministry of Economic Affairs Fifth, the invention description (28) The same effect. [Embodiment 3] In Embodiment 1, the liquid crystal panel in which the signal voltage waveform is supplied from only one side of the signal electrode is described, but it is also effective when supplied from both ends. When calculating the value I in Embodiment 1, As long as the function g (l KL / 2 1) is substituted for the function f (k), the calculation can be performed. At this time, the function g (X) is an increasing function that increases as the value X increases. By performing the same operation as in the first embodiment based on the numerical value I, the same effect can be obtained. In Embodiments 1 to 3, the method of adjusting the correction amount uses a method of keeping the voltage difference between the correction voltage and the non-selected voltage constant to increase or decrease the time for applying the correction voltage (hereinafter referred to as time axis correction). However, it is also possible to keep the time for applying the correction pressure constant, and change the difference between the correction pressure and the non-selected electrical IS (this action is voltage axis correction). You can also change the time and pressure of applying the correct pressure (this action is the time K-axis correction). You can also use the wave height value and the waveform represented by the exponential function that changes the required correction amount and the triangle waveform as the correction K (this action is the function waveform correction). [Embodiment 4] The following describes an embodiment of display unevenness caused by a horizontal cable. As mentioned above, in the selected scanning flute of the scanning poles Y1 ~ Ys, the scanning pole and the frost poles Xp (p = 1,2 ... -(Please read the precautions on the back before filling out this page) • Crazy ... Order. -Line. A 4 (210X 297 public) 201825 A6 B6 V. Description of invention (29) When the position of the display point is P, change The display unevenness caused by the horizontal pull of the value Z 'calculated by 6 Z' = Σ q (i) * δ (i) i = 1 occurs. In other words, the effective voltage applied to the display point on the scan electrode of the object becomes smaller in accordance with the value Z '. Therefore, it is only necessary to operate the liquid crystal display device to calculate the numerical value Z 'and perform correction in accordance with the numerical value Z'. The specific examples are described below. The eighth ring shows the structure of this embodiment. In Figure 8, 102 and 103 are the same as the control multiple number in Figure 1, and the data multiple number is the same, so the same symbol is used to omit its description. 801 is a liquid crystal cell, which is generally composed of a liquid crystal panel 201, a scanning electrode driving circuit 805, and a signal electrode driving circuit 213. 804 is a voltage waveform correction flute circuit (hereinafter referred to as a correction circuit) that calculates the value Z 'and generates a correction number that becomes active only during the time that matches the length of the value. The 2 sets of power supply (from the selection of K and non-selection voltage, output from the frost source circuit 805, which are used for the description of the mold) are called Y Xiaoyuan. When the correction power supply 805 is active and non-active, the power supply 805 selects a voltage different from each other as the selection voltage of the Y power supply 806 and turns it out. 106 is the power supply (hereinafter referred to as) (the flute source) used to drive the signal output from the power circuit 805, which is the same as the X yuan source 107 in the first one. 809 is the correction flute used to determine the round from the correction of the road 804 (please read the phonetic V matters on the back before filling in this page). Order · Line · Central Ministry of Economic Affairs of the Ministry of Economic Affairs printed the bezel 9 with 0 examples It must be structured. The number of circles is 8 times. According to the first revision, the pressure is reduced by 4 (210X 297 g;% *) 201825 A6 B6 Printed by the Central Approval Bureau of the Ministry of Economic Affairs 5. Description of the invention (30) The liquid crystal unit 801 has a structure example. 201 in the figure is a liquid crystal panel, which is the same as the liquid crystal panel of the first embodiment. 805 is a scan electrode driving circuit (hereinafter referred to as a Y driver), and Ye is composed of a shift register 206, a conversion circuit 911, and a level shifter 912. The output of the level shifter 912 is directed to each scan electrode Y1 to Y6 of the liquid crystal panel 201. The structure of the shift register 206 is the same as that of the first embodiment, so its explanation will be omitted. The switching circuit 911 selects one set from the two sets of voltages of the Y power supply 806 generated by the power supply circuit 804 of FIG. 8 according to the signal FR. The voltage V0 ', V1, V4, and V5' that constitute the Y power supply 806 will soon be divided into the first voltage group, the voltage V5 ', and the voltage V5', and VI into the second voltage group, and The FR is converted into the selection voltage of the 2 groups according to the serial number FR, and the non-selection voltage. Similarly, the voltages V5 ', VI are referred to as the selection voltage of the second voltage group, and the non-selection voltage. Any one of the two sets of voltages selected by the conversion circuit 911 is supplied to the level register 912. 912 is the level register, which is composed of the opening of the multi-circuit 2 loop 1 contact, and is converted according to the content of the shift register 206. In other words, when the content is "1", each switch selects the selection voltage and supplies it to the corresponding scanning electrodes Y1 ~ Y6. When the content is "0", each open reel selects the non-selected voltage and supplies it to the corresponding scanning flutes Y1 ~ Y6. The structure of the Y driver 905 has been described above. The operation will be described below. Synchronize the double number DIN with the LP input shift register 206 and transfer it. As a result, one of the value switches in the level shifter 912 and its corresponding one sequentially generate S swing pressure. The rest of the Luguan Pass will produce non-selected flute pressure (please read the notes on the back before filling in this page) • Install ·. Order · • Line · A 4 (210X 297 Public Issue) Printed by the Central Warrant Bureau of the Ministry of Economic Affairs 201825 A6 _ B6 V. Description of the invention (31) Explain the structure and operation of the Y driver. Reference numeral 213 is a double-number electrooscillation circuit (hereinafter referred to as an X driver). Since the embodiments are the same, the description will follow. The above is the structure of the crystal unit 801. On the scanning poles Y1 ~ Y6 of the liquid crystal unit 801, a selection voltage is applied synchronously with the DIH and LP times, and then, in synchronization with it, the lighting or non-point of the display coil type is applied to the times XI ~ X6. The S is turned on to cause the liquid crystal panel 201 to display. The operation of the liquid crystal cell 801 has been described above. In Fig. 8, 804 is a correction circuit. Among scan electrodes Y1 to Y6, the position of the display point formed by the next selected scan electrode and each signal electrode Xp (p = 1,2, ..., 6) As P, calculate 6 Z '= Σ q (i) * S (i) ..... (1) i = 1 Then, when selecting the next tracing pole, it will only match the value Z' The correction signal 809 that becomes active for a length of time is output in synchronization with the multiple LP. Fig. 10 shows a concrete detailed structure. In the tenth circle, 1 0 01 is the counter. 1 0 0 2 is a constant generation circuit. 1 0 0 3 is a gate circuit that generates a logical product of multiple times and data signals 103 representing multiple values of the value output by the constant generation circuit 1002. 1004 is the Luck Road for Addition. 1 005 is the first latch circuit that maintains the operation result of the operation flute circuit 1 005. 1006 is the second latch circuit. 1007 is a correction circuit that will become active only within the time period corresponding to the content of the second latch flute circuit 809. The correction circuit will be generated by synchronizing with the correction number issued by the LP (hereinafter referred to as the production circuit). -33- (Please read the precautions on the back before filling in this page). Insects. • Ordered • • Line. A 4 (210X 297 彂 发) Printed by the Central Bureau of Economic Development of the Ministry of Economic Affairs 201825 A6 B6 V. Description of invention (32 ) The counter 1001 is reset to 0 by the signal LP, and it is counted up by the action of the multiplier XSCL »Its round is supplied to the constant generation circuit 1 002 〇 The constant generation circuit 1 0 0 2 The ROM and diode matrix And the output of the counter 100 1 produces a different value. In other words, when the output value of the counter 1001 increases, it also generates a larger output. The value corresponds to the function q of formula (1) (i) ○ where i is the value added by 1 to the value generated by the counter 1001. The value of the function can also be calculated by experiment. In this embodiment »The following expression is when 〇1 = 1. When q (1) = 1 i — 2, when q (2) = 1.1 When i = 3, q (3) = 1.2 When i = 4, q (4) = 1.3 When i = 5, q (5) = 1.4 When i = 6, q (6) = 1.5 Lan Yuanlu 1 003 Generate constants Generate values and data output by Jiaolu 1002 The logical product of the number 1 0 3. In other words, if the data number is active " 1 1 when the XSCL signal drops, then the value of the constant generating circuit 1 002 will be generated * If it is not active " 0, then 0 will be generated. Corresponds to q (i) (5) in (1) (5, ί)) where ί is the number of times the signal XSC has fallen after the number LP has dropped, the phase flute adds a value of 1 to the value generated by the counter 100 1 〇1 00 4 is the arithmetic circuit t. The value generated by the gate gate circuit and the first latch (please read the precautions on the back and then fill out this page) • Chong · 'order · • green. A 4 (210X 297 father did not ) Printed by the Central Acquisition Bureau of the Ministry of Economic Affairs * 201825 A6 B6 V. Description of the invention (33) The content of the lock circuit 1005 is added to the XSCL synchronously, and the result is sent back to the first latch circuit 100 5. 1005 and 1006 are the first and second latch circuits. The first latch type circuit 1005 is reset by the signal LP. The content of the latch circuit 1005 immediately before reset corresponds to the value Z 'in the following formula. 6 Z '= Σ q (i) ί δ (i) .... (1) i = 1 In other words, the number of lighted points on the scan electrode selected next is weighted according to the position of the lighted point and counted Of the value. The content is input into the second latch circuit 1 0 0 6 when the first latch barium circuit is reset when the signal LP falls. At this time, the next scan electrode is selected when the multiple LP falls, so The value indicated by the second latch circuit 1006 is the number of weighted lighting points on the selected scan electrode. 1007 is a generation circuit that will become active " 1 "correction times 809 only within the time period matching the value shown in the second barium circuit 1006, and be synchronized with the times LP to be rounded out. The circuit 1007 may be : Generate a turn-on circuit 1008 that directly multiplies the XSCL, or divides or multiplies the clock (hereinafter referred to as the correction clock); the counter 1009 is set to 0 by the signal LP due to the upward counting of the correction clock; The active "1" signal is easily formed until the coincidence detection circuit 1010 indicating that the value of the counter 1009 in turn matches the value indicated by the second latch circuit. The period of the correction clock need not be at equal intervals, for example, it can be determined by experiment. The above describes the structure of the modified flute road. Secondly, will be selected for tracing -35- (please read the precautions on the back and then fill in this page). ¾. • Order · • Green · A 4 (210X 297 father did not) 201825 A6 B6 V. DESCRIPTION OF THE INVENTION (34) The number of lighted points on the pole is weighted and counted, and the scan electrode selected at the next inclination is only synchronized with the signal LP when it becomes the active correction number 809 within the length matching the value Z ′. To be output. The power flute 805 of FIG. 8 will be described below with reference to FIG. 11. Fig. 11 shows a specific configuration example of the power supply road 805. In the ring, 1101 ~ 1107 are series resistance. The voltage V0U.V5L is applied across it. These resistors 1101 to 1107 form a voltage divider circuit. The voltage generated by each resistor 1101 to 1107 is V0U, V0, V1, V2, V3, V4, V5, V5L from the top, then V ® V0-V1 = V1-V2 = V3-V4 = V4-V5 V2-V3 = a # V, where a is a constant of about 1 ~ 50. The resistance value of each resistor 1101 to 1107 is set to a value that can establish the following formula. V0U-V0 = V5-V5L. 510 is the same as the voltage stabilization circuit 510 in Figure 1, so its explanation is omitted. 1108, 1108 are conversion circuits, and become active "1" at the correction multiple 809. Printed by the Central Bureau of Economic Development of the Ministry of Economic Affairs (please read the back page first Please pay attention to this page and then fill in this page), when you select V0U and become non-active "0 ", you select V0. Then, set the selected 18® to the flute voltage V0 '. Similarly, the conversion circuit In 1109, when the amended number 809 is active with "1", the selection voltage is 5L, and when it becomes non-active with "0", the flute KV5 is selected. Then, the selected voltage is set to the voltage V 5 '. Here, K K 0 U and V 5 L are called modified flute pressures. A4 (210X 297 public issue) 201825 A6 B6 Printed by the Central Bureau of Economic Affairs of the Ministry of Economy V. Description of Invention (35) Use voltage V0 ', V4, V5', V1 as Y power supply 106, and power supply V0, V2, V3, As the X power supply, V5 is supplied to the LCD 801 in the eighth circle. The liquid crystal display device of the eighth circle has the structure as described above. The following describes the operation of this embodiment when the quadrangular shape is actually displayed. Figures 12 and 13 are circles showing the display contents displayed on the liquid crystal panel 201 of the ninth circle. In the figure, the dots in the diagonal line indicate lighting. In this example, Figure 12 and Figure 13 show that the same quadrangle is displayed on the left and right, respectively. The action is as follows. First, when the display in FIG. 12 or FIG. 13 is performed, the eighth correction circuit 804 divides the number of sticky bright spots on the scan electrodes Y1 to Y6 of the liquid crystal panel 201 in FIG. 12 or FIG. The positions are weighted and counted. That is, the value Z 'is calculated. The numerical value Z 'of each scan electrode Y1 to Y6 is as follows. Scanning power Y1 ..... Z '= 〇 + 〇 + 〇 + 〇 + 〇 + 〇 = 0 scanning power Y2 ... Z, = 1, 0 * 1 + 1. 1 * 1 + 0 + 0 + 0 + 0 = 2.1 Poke the pole Y3 ..... Z '= 1.0 ^ 1 + 1.1 ^ 1 + 0 + 0 + 0 + 0 = 2. 1 Scan the pole Y4 ..... (please read the precautions on the back before filling in this page) • Xiang · Order · Fate. A 4 (210X 297 Public Issue) Printed by the Central Approval Bureau of the Ministry of Economic Affairs 201825 A6 B6 V. Description of Invention (36) Z '= 1. 0 ^ 1 + 1. 1 ^ 1 + 0 + 0 + 0 + 0 = 2.1 Scanning electrode Y5 ... Z' = 1.. S: l + l. 1 * 1 +0 + 0 + 0 + 0 + 0 = 2.1 Scanning Y6 ... Z '= 〇 + 〇 + 〇 + 〇 + 〇 + 〇 = 〇 When displaying the display in Figure 13, scan Lei Y1 .. ... Z '= 〇 + 〇 + 〇 + 〇 + 〇 + 〇 = 0 Broadcast electrode Y2 ... z, = 0 + 0 + 0 + 0 + 1, 4 * 1 + 1, 5 * 1 = 2.9 Scanning Y3 ... Z '= 〇 + 〇 + 〇 + 〇 + 1.4-1 + 1.5 * 1 = 2.9 Scanning Y 4 .... Z' = 0 + 0 + 0 + 0 + 1.4 ^ 1 + 1.5 ^ 1 = 2.9 Scanning Y5 ..... z · = 0 + 0 + 0 + 0 + 1. 4 ^ 1 + 1.5 * 1 (please read the note on the back first Please fill in this page again for details). Hit. • Line. A 4 (210X 297 public) A 6 B6 201825 5. Description of the invention (37) = 2.9 (please read the precautions on the back before filling in this page) Scanning power Y6 ... Z '= 〇 + 〇 + 〇 + 〇 + 〇 + 〇 = 0 8th The corrected Xiaolu 804 in the figure only outputs the active correction multiple number 8 0 9 that matches the length of the value Z '. The power supply circuit 805 only rounds out the correction voltages V0U and V5L at the time when the correction signal 809 becomes active as the selection voltage to replace the voltages V0 and V5. • Hit. When the display in Figure 12 is displayed, the value Z 'of the scan electrodes Y2 ~ Y5 is less than the value when the display in Figure 13 is displayed. Therefore, when displaying the display in Figure 12, if the scan ~ Y5 is being selected, the correction voltages V0U and V5L become the selection voltages. The time of application is shorter than the display time in Figure 13. The Y driver 905 of Fig. 9 applies the selection voltage that varies according to the value Z ', and sequentially applies the selection voltage to the scanning electrodes Y1 to Y6 of the liquid crystal panel 201. The operation of the Y driver 213 is the same as that of the first embodiment, and its description is omitted. • Green · The Ministry of Economic Affairs Central Bureau of Precinct Printing performs the above operations, as shown in the example shown in Figure 12. If the lighting point is near the drive terminals of scan electrodes Y2 ~ Y5, the voltage on the scan® pole is matched The loss of the waveform is not too large, and a selective voltage with a small amount of correction voltage is supplied to the selected brush trace electrode. In other words, the flute pressure with a shorter application time is used as the filling flute®, instead of the correction box V0U, V5L being the pressure V0, V5 supplied to the pole. On the contrary, as shown in the display example in Figure 13, if the lighting point is faster than the drive end of the scan flute Y2 ~ Y5, A4 (210X 297 public) A6 B6 201825 V. Description of invention (38) (please Read the precautions on the back first and then fill out this page.) In conjunction with the larger distortion of the voltage waveform on the scan electrode, the selected voltage with a larger amount of correction voltage is supplied to the selected scan electrode. In other words. The voltage applied for a long time is used as the selection voltage, and instead of the modified voltage V 0 U, V 5 L becomes the voltage V 0, and the voltage of V 5 is supplied to the electrode. Therefore, the difference between the correction amounts can be roughly corrected based on the display positions in Figure 12 and Figure 13. As described above, when counting the number of lighting points on each scan electrode, the position of the lighting point is weighted and counted, and the selection voltage supplied to each electrode is changed according to the result. This eliminates display unevenness. [Embodiment 5] Order. Embodiment 4 shows that the liquid crystal panel in which the scanning voltage waveform is supplied from only one side of the scanning frost electrode has the same effect when supplied from both ends. In other words, in Example 4, when calculating the value Z ′, it is sufficient to substitute the function q (i) for the value P (| i -S / 2 |), and the function p (x) is when the value X increases The reduction function of extinction. Performing the operation as described in Embodiment 4 according to the value Z 'can produce the same effect. . Line · In Example 4, the adjustment method for the correction amount is to maintain the difference between the voltage V0U and V 0, the difference between the β V 5 and V 5 is constant, and the voltage V 0 U and V 5 L are selected as the choice. The method of pressure application time is also the method of real-time shaft correction. Others can also use flute pressure axis correction, time flute pressure axis correction, and function waveform correction method. The Central Bureau of Economic Affairs of the Ministry of Economic Affairs printed Examples 1 to 2, 3 to change the non-selective ΙΒΚ, while Examples 4 and 5 were pseudo-modified 4 0 1 A 4 (210Χ 297 public release) 201825 A6 B6 V. Description of the invention (39 ) This is due to the fact that the average pressure of the French side is not small. 0 ^ Variations and 1 * Examples of practical benefits can be placed in the same costume and display when the utility is dispatched. The liquid Lu is included in the package when tested. When the time is taken, the original word in the stipulated hair according to the text, as shown on the Shuyuan, shows the number of the voltage setting voltage as shown in the display. Duan Peiyun, Ji Yu, and Yu Ji, all kinds of wave-shaped pressure-displayed waves, good voltages, and changes in the scanning and scanning, and the changes and changes in the drive are all available as evidence for this. The liquid is taken to the example of the zigzag shape installation. The installation of IX 0 shows the clear version of the crystal liquid. The first version of the figure is the first figure of the first figure. Knot moving 0) 08 ί Road 0 0 0% S- ^ 1 H Zhengli® Eye Sweep V The structure of the electric positive junction table is repaired as shown in Figure 3, the structure of the elementary single crystal liquid is shown in Figure 2 The pressure path of the knot wave Ε Xia is the source picture electricity is the first example of the plate crystal crystal through the chart. The first one is shown in the 6th row of the figure. The display is shown as g) / | \ Is} Figure (a 6 Figure 7 ο Negative picture. Composition δ Figure 9 shape wave. Compression of the electric structure plus the construction of the first example of the installation and implementation of 4 0 The figure shows the structure of the statement The meta-form is Jingtu Liquid_15 and its modification is picture ο 11 0 No. 3 ο 1 Picture No. structure. The electrical surface of the source board of the lucid electricity is the No. 11 of Jingtu Liquid The display structure shown inside the circle block is the electrogram. Figure 12 is the first view. Figure 14 is the view. The view of the transparent plate is through the crystal surface of the crystal plate. The internal display shows one of the techniques used in the skill 9 (please read the precautions on the back before filling in this page) k.. Order · _ Line · Printed by the Central Bureau of Economic Affairs of the Ministry of Economic Affairs on one plus For example, when the volume is displayed on the top, the content is displayed on the display, 9 is displayed as a display, and 16 is the image. The image is shown in Figure 4 7 1Χ ο 11 in the first line. Through ^ ffi- ^) ~ 1B (a plate crystal surface liquid 15 crystals of the first liquid example A 4 (210X 297 public release) 201825 A6 B6 V. Invention description (40) perspective view of the liquid crystal panel. 101,108 ... liquid crystal Unit. 102 ... Control signal. 103 ... Data multiple. 104,804 ... Voltage waveform correction circuit (correction circuit). 8 05 ... Power circuit. 1 06,806 ... Y power supply. 107 ... X power supply. 108, .. a circuit that generates a clock signal synchronized with the multiple LP. 109 ... correction signal. 110 ... clock signal. (Please read the precautions on the back before filling this page) . Therefore • Order. Line. Printed by the Central Bureau of the Ministry of Economic Affairs 42- A 4 (210X 297 public)