201033654 六、發明說明： 【發明所屬之技術領域】 本發明係關於根據噴墨方式之彩色濾光片之製造方法 。詳言之’本發明係關於噴墨方式之彩色濾光片之製造方 法’係具有由複數噴嘴構成的噴嘴列之記錄頭複數排列之 頭單元來對基板上同時進行列印的場合，補正個記錄頭的 安裝位置之噴嘴排列方向的偏離而進行列印爲特徵之彩色 φ 濾光片之製造方法。 【先前技術】 現在，彩色濾光片係被構築於玻璃透明基板上，由遮 光用的黑矩陣（BM) 、RGB (紅、綠、藍）之三原色畫素 、畫素保護膜、作動電極透明導電膜等構件所構成。其中 ’最基本的構件爲三原色畫素，其分光透過率、色調等色 顯示相關的性能、耐光性、平滑性等顯示面板的構成所需 〇 要的特性，以及耐熱性、耐藥品性、尺寸安定性等顯示面 板的組裝上所必要的特性等，被要求著各種分歧的特性， 而有各種色材（染料、顏料、金屬薄膜）以及濾光片製造 方法（染色法、顏料分散法、電鍍法、印刷法、蒸鍍法） 相關的開發硏究持續進行迄今。 今後’彩色液晶顯示裝置（LCD )亦被期待著往大畫 面化、高精細化、低成本化的方向發展。其中，對於基本 ' 構件之彩色濾光片，其性能要求越來越高。 在市場上爲了對應彩色LCD的大型化、高精細化、 201033654 低成本化，作爲現在彩色濾光片的製造方法，替代主流之 光蝕刻法，已可見到開發改採噴墨方式的製造方法之動向 〇 藉由此噴墨方式製作彩色濾光片時，在形成BM的胞 開口部之瞄準的位置高精度地進行印刷是必須的。 一般而言，噴墨式記錄裝置，把具有複數個噴嘴排列 於副掃描方向之噴嘴列之記錄頭，在主掃描方向並列排列 複數個而成的頭單元’往主掃描方向之去程與回程之至少 一方掃描，同時由複數噴嘴吐出墨水而往被記錄物記錄。 以噴墨方式製造彩色濾光片時，因爲有與鄰接的胞的 開口部混色之危險性，所以關於落下位置被要求非常高的 精度。同時列印複數色時，進而要求更高的精度，必須要 有對應於各色之記錄頭的位置偏離的補正功能。 於專利文獻1，記載著檢測出被形成彩色濾光片的基 板與描繪頭（記錄頭）之間之相對位移與從描繪頭吐出的 墨水之在基板上的落下位置，根據這些檢測結果來配合基 板及記錄頭間之6自由度方向的位置爲特徵之彩色濾光片 製造裝置。 然而，記載於專利文獻1的裝置及方法，也未能充分 達成效率佳地生產彩色濾光片。例如，爲了使複數色之噴 墨之落下位置均能良好的對準位置，必須要補正在同一個 頭單元安裝複數之頭時產生的頭間之副掃描方向之偏離， 使頭之安裝位置爲可移動的話，裝置的構成變得複雜而需 要供製作移動機構的成本。進而，在使頭移動而補正偏離 -6 - 201033654 的作業必須要耗掉很多時間。 〔專利文獻1〕日本專利特開平09-4991 9號公報 【發明內容】 〔發明所欲解決之課題〕 本發明係爲了解決前述問題而爲之發明，課題在於提 供改善相關的問題點，同時活用噴墨方式的特性之廉價且 φ 可信賴性高之彩色濾光片之製造方法。 〔供解決課題之手段〕 本案發明人銳意檢討如何解決前述課題的結果，發現 預先製作各記錄頭之複數噴嘴之全列印圖案，同時使各記 錄頭的兩端之噴嘴之一定數目確保爲移位空間，使用該移 位空間適當偏移實際列印圖案，可以不移動記錄頭之安裝 位置自身，而容易補正該記錄頭的副掃描方向（噴嘴排列 φ 方向）的偏離導致列印偏離，可以有效解決前述課題，而 完成本發明。 亦即’本發明提供以下1 )〜3 )所示之彩色濾光片之 製造方法。 1) 一種彩色濾光片之製造方法，係使具有把複數噴 嘴排列於副掃描方向之噴嘴列的記錄頭排列複數個於主掃 描方向上而構成的頭單元在基板上掃描，使由前述噴嘴吐 出之墨水落於藉由被形成於前述基板上的黑矩陣而區隔的 胞之開口部而構成之根據噴墨方式之彩色濾光片之製造方 -7- 201033654 法，特徵爲包含以下步驟（a)〜（c)之彩色濾光片之製造 方法。 步驟（a):前述各記錄頭之複數噴嘴之中，以落於 開口部的位置之噴嘴爲有效噴嘴，以不落於開口部的位置 的噴嘴爲無效噴嘴，於前述各記錄頭對所有的噴嘴判定是 有效噴嘴或是無效噴嘴的步驟（判定步驟） 步驟（b):作成關於一個開口部之墨水吐出控制用 個別圖案的步驟（個別圖案作成步驟） @ 步驟（c):對在前述步驟（a)被判定爲有效噴嘴的 噴嘴群，***在前述步驟（b )所得到的個別圖案，同時 使前述各記錄頭之複數噴嘴之中該記錄頭的兩端部的噴嘴 之一定數量登錄爲移位空間（shift space)，使用該移位 空間補正個記錄頭之噴嘴排列方向的偏離而作成全列印圖 案的步驟（全列印圖案作成步驟） 2) 如申請專利範圍第1項之彩色濾光片之製造方法 ，其中前述步驟（a)之是有效噴嘴或是無效噴嘴的判定 © ，藉由是否滿足下列式（1 )而進行的。 〈數學式1〉 D<(AxE)%(B + C)<(B-D) …⑴ (數學式（1)之中，「％」表示除算時之餘數，記號 A爲噴嘴編號（1〜N之整數）’B爲開口部長邊之長度’ C爲B Μ寬幅，D爲剩餘寬幅，E爲噴嘴間隔）。 3) 如1)或2)之彩色濾光片之製造方法’其中進行 試印而檢測出前述步驟（c )之各記錄頭之噴嘴排列方向 -8 - 201033654 . 之偏離。 〔發明之效果〕 根據本發明，複數記錄頭間之噴嘴排列方向（副掃描 方向）的記錄位置偏離的補正變得容易，沒有必要移動記 錄頭之安裝位置自身，所以不需浪費很多時間與成本於具 備複雜的移動機夠之裝置或者頭的移動作業等，可以高精 φ 度地製造彩色濾光片。 【實施方式】 (η根據噴墨方式之彩色濾光片之製造方法 本發明之彩色濾光片之製造方法，係使具有把複數噴 嘴排列於副掃描方向之噴嘴列的記錄頭排列複數個於主掃 描方向上而構成的頭單元在基板上掃描，使由前述噴嘴吐 出之墨水落於藉由被形成於前述基板上的黑矩陣而區隔的 〇 胞之開口部而構成之根據噴墨方式之彩色濾光片之製造方 法。 以下，使用圖面詳細說明本發明之實施型態之一例。 又，在本例，顯示由RGB之3色所構成的彩色濾光片之 製造方法，但本發明並不以此例爲限定，例如亦可應用於 黃、洋紅、藍綠、黑（YMCK )之四色所構成之印刷物之 製造方法。 圖1係由噴嘴方向所見之被並列搭載的3個記錄頭（ 頭1，2,3 )所構成的頭單元之槪略圖。記錄投1，2,3分別吐 201033654 出R、G、B之各色（順序未必相同）之墨水。 各記錄頭之噴嘴的排列，亦可爲圖1所示之一列，亦 可爲複數列。複數列的場合，藉由使各噴嘴列於其排列方 向上偏移配置，可以縮小外觀上的噴嘴間隔，可以列印更 爲精細的圖案。 對基板賦予墨水時，使前述頭單元掃描（主掃描）於 掃描方向X(主掃描方向），使墨水吐出至基板上。這個 場合，如圖1所示，各記錄頭之複數噴嘴通常被排列於副 @ 掃描方向，具備該噴嘴列的各記錄頭並列地排列於主掃描 方向。 作爲記錄頭，可以使用可分別設定由各個噴嘴吐出之 墨水吐出量之多滴（multi-drop )方式者。使用這樣的多 滴方式之頭，可使往開口部內之墨水吐出量隨著落下的位 置而改變，可以防止混色。 圖2係顯示被形成黑矩陣（BM)之基板之槪略圖。 藉由格子狀的黑矩陣區隔的胞的開口部有規則地排列著。 馨 於此胞的開口部被賦予各色之墨水而成爲副畫素。一般而 言，開口部大致爲長方形，於此長方形的長邊方向排列的 開口部列（開口部1之列、開口部2之列、開口部3之列 )，分別被賦予相同顏色的墨水。於3色之彩色濾光片的 場合，開口部1，2,3分別作爲R列、G列、B列（順序未 必相同）成爲1組之反覆單位。 作爲使圖1所示之頭單元掃描而對基板上之所有的開 口部賦予預定的墨水的手段，由基板之端部使頭單元掃描 -10- 201033654 . 於一方向（主掃描方向之去程）（主掃描）同時往基板上 之開口部賦予墨水，到達相反的端部時再與掃描方向直交 的方向上（副掃描方向）副掃描相當於頭寬幅之長度。接 著，於與前述主掃描方向相反的方向（回程）再度使頭單 元主掃描同時對基板上之開口部賦予墨水。 藉由反覆這樣的動作，可以使用小的頭單元對大尺寸 之基板全體賦予墨水。此外，即使是種種尺寸的基板，也 φ 不需要交換噴嘴而對全體賦予墨水而製造彩色濾光片。 使頭單元掃描的方向（主掃描方向），最好是被形成 於基板的開口部之短邊方向。此時，以使噴嘴列之排列方 向與開口部之長邊方向成爲一致的方式配置頭單元較佳。 藉由把頭單元與基板如此地配置，設定掃描方向，可以有 效率地使用噴嘴，可以提高生產性。 在前述說明，舉出固定基板而使頭單元掃描之例’但 相反地使基板掃描而藉由固定的頭單元賦予墨水的方法亦 φ 被包含於本發明。 (2)步驟（a):有效噴嘴與無效噴嘴的判定步驟 在本發明，使頭單元掃描而對基板上賦予墨水時，預 先製作而登記各記錄頭之對權噴嘴之列印圖案（墨水之吐 出控制資料）。 一個頭單元之列印圖案（全列印圖案）’可以首先製 作相當於一個開口部的列印圖案（個別圖案）’而藉由將 其適用（***）於該頭單元之所有噴嘴而作成。但是於構 -11 - 201033654 成各記錄頭之噴嘴之中，存在有墨水落下於開口部的位置 之噴嘴，與不落於開口部的位置之噴嘴。 圖3係顯示於一個開口部列上，由一個記錄頭之噴嘴 列吐出的墨水落下的位置。由圖3可知，從所有的噴嘴吐 出墨水的場合，隨著開口部與噴嘴之位置關係不同，有適 切落下於開口部之中的吐出墨水，亦有落下於開口部與 BM之邊界線上或BM上之吐出墨水。 但是，落下於開口部與BM之邊界上或者於BM上會 @ 變成混色的原因，此外會浪費墨水所以應該要避免。因此 ，在本發明’預先於前述個記錄頭對所有的噴嘴，判定各 噴嘴是否是落下於開口部的位置之噴嘴（有效噴嘴），還 是不會落下於開口部的位置之噴嘴，亦即落下於開口部與 BM之邊界線上或BM上的位置之噴嘴（無效噴嘴），而 不使墨水由無效噴嘴吐出。 亦即，本發明之方法，包含：前述各記錄頭之複數噴 嘴之中’以落於開口部的位置之噴嘴爲有效噴嘴，以不落 n 於開口部的位置的噴嘴爲無效噴嘴，於前述各記錄頭對所 有的噴嘴判定是有效噴嘴或是無效噴嘴的步驟（判定步驟 :步驟（a))。 判定之具體方法沒有特別限制，例如可以針對頭（ head )上的N個噴嘴，由其BM寬幅、開口部長邊的長度 而藉由下列數學式（1)來判定是有效噴嘴還是無效噴嘴 〈數學式1〉 -12- 201033654 D<(AxE)%(B + C)<(B-D) ...(1) 其中，前述數學式（1)中，「％」代表除算時之餘數 ，此外記號A〜E代表以下意義。 A:噴嘴編號（1〜N之整數） B:開口部長邊的長度 C : BM寬幅 D :寬裕寬幅 φ E :噴嘴間隔 又，藉由前述數學式（1)判定有效噴嘴或無效噴嘴 時，前提是有必要使顯示噴嘴的位置的座標系原點，與藉 由黑矩陣劃分的開口部長度方向的座標系的原點一致。前 述數學式（1)之A〜E均係以原點（顯示噴嘴的位置之座 標系與黑矩陣所劃分的開口部長度方向之座標系共通之原 點）起算的座標（距離）來表示。 以下，使用圖4，說明藉由前述數學式（1)判定其爲 φ 有效噴嘴或無效噴嘴的方法。 前述數學式（1)之噴嘴編號A係於具有由一端排列 成一列的1〜N個噴嘴的一個記錄頭，表示由該一端起依 序賦予編號的場合之噴嘴編號（又，與原點重叠的噴嘴爲 0號）。亦即，「（AxE)」係表示任意（噴嘴編號A)的噴 嘴之記錄頭內的位置（由一端起算的距離）。此外，「 (B + C)」係代表玻璃基板上之黑矩陣（BM)與藉其劃分的 - 開口部之一個所構成的一組之該開口部長邊方向的長度。 頭之一端起至噴嘴編號A的噴嘴爲止的距離（AxE)以 -13- 201033654 (Β + C)除之時之餘値，成爲原點側之最近的開口部端部起 直到噴嘴編號A的噴嘴爲止的距離F (圖4中之F)( (AxE)%(B + C) = F)。 例如’由圖4中之一端起第14個（噴嘴編號14)之 噴嘴7’之從原點至該噴嘴爲止的距離爲（14χΕ)，圖4之例 的場合，除以（Β + C)的2倍距離之餘値，亦即把（14><£)以 (Β + C)除之之餘値爲F(參照圖4)❶201033654 VI. Description of the Invention: [Technical Field of the Invention] The present invention relates to a method of manufacturing a color filter according to an inkjet method. More specifically, the present invention relates to a method for producing a color filter for an ink jet method, which comprises a head unit in which a plurality of recording heads of a nozzle array composed of a plurality of nozzles are arranged on a substrate to simultaneously print on a substrate, and corrects A method of manufacturing a color φ filter characterized by a deviation in the direction in which the nozzles are arranged at the mounting position of the recording head. [Prior Art] Now, the color filter is constructed on a glass transparent substrate, and the black matrix (BM) for light shielding, the three primary color pixels of RGB (red, green, and blue), the pixel protective film, and the movable electrode are transparent. It is composed of a member such as a conductive film. Among them, the 'most basic component is the three primary color pixels, and its spectral transmittance, color tone and other colors display related performance, light resistance, smoothness, and other characteristics required for the display panel, as well as heat resistance, chemical resistance, and size. Various characteristics such as stability and assembly of the display panel are required, and various characteristics (different dyes, pigments, metal films) and filter manufacturing methods (dyeing method, pigment dispersion method, plating) are required. Law, printing method, vapor deposition method) Related developments continue to date. In the future, the color liquid crystal display (LCD) is expected to be developed in the direction of large-scale, high-definition, and low-cost. Among them, the performance requirements for the color filter of the basic 'components are getting higher and higher. In order to cope with the increase in size and definition of color LCDs and the cost reduction of 201033654 in the market, as a method of manufacturing color filters, instead of the mainstream photolithography method, a manufacturing method of developing an inkjet method has been found. When the color filter is produced by the ink jet method, it is necessary to perform printing with high precision at the position where the cell opening of the BM is formed. In general, an ink jet recording apparatus has a recording head in which a plurality of nozzles are arranged in a nozzle row in the sub-scanning direction, and a head unit in which a plurality of head units are arranged side by side in the main scanning direction to go to the main scanning direction. At least one of the scans is performed, and at the same time, the ink is ejected from the plurality of nozzles and recorded to the recorded object. When the color filter is manufactured by the ink jet method, since there is a risk of color mixing with the opening of the adjacent cell, a very high precision is required regarding the drop position. When a plurality of colors are printed at the same time, a higher precision is required, and a correction function corresponding to the positional deviation of the recording heads of the respective colors is necessary. Patent Document 1 describes detecting a relative displacement between a substrate on which a color filter is formed and a drawing head (recording head) and a drop position of the ink discharged from the drawing head on the substrate, and matching the detection results. A color filter manufacturing apparatus characterized by a position in the direction of 6 degrees of freedom between the substrate and the recording head. However, the apparatus and method described in Patent Document 1 have not sufficiently achieved the efficiency of producing a color filter. For example, in order to make a good position of the falling position of the ink of the plurality of colors, it is necessary to compensate for the deviation of the sub-scanning direction between the heads generated when the same head unit is mounted at the head, so that the mounting position of the head is When moving, the configuration of the device becomes complicated and the cost for manufacturing the moving mechanism is required. Furthermore, it takes a lot of time to move the head to correct the deviation from -6 - 201033654. [Problem to be Solved by the Invention] The present invention has been made to solve the above problems, and an object of the present invention is to provide an improvement in related problems while utilizing A method of manufacturing a color filter which is inexpensive in terms of an ink jet method and has high φ reliability. [Means for Solving the Problem] The inventors of the present invention are keenly reviewing the results of the above-mentioned problems, and found that the full print pattern of the plurality of nozzles of each recording head is prepared in advance, and the number of nozzles at both ends of each recording head is ensured to be shifted. The bit space, using the shift space to appropriately offset the actual print pattern, can easily correct the deviation of the sub-scanning direction (nozzle arrangement φ direction) of the recording head without causing the mounting position of the recording head itself, thereby causing the printing deviation. The present invention has been accomplished by effectively solving the aforementioned problems. That is, the present invention provides a method of producing a color filter shown in the following 1) to 3). 1) A method of manufacturing a color filter, wherein a head unit having a plurality of recording heads arranged in a nozzle row arranged in a sub-scanning direction is arranged in a main scanning direction, and a head unit is scanned on a substrate to cause a nozzle The method of manufacturing the color filter according to the ink jet method, which is constituted by the ink opening of the cell which is separated by the black matrix formed on the substrate, is characterized by the following steps: (a) to (c) a method of producing a color filter. Step (a): among the plurality of nozzles of each of the recording heads, the nozzle that is at the position of the opening is an effective nozzle, and the nozzle that does not fall at the position of the opening is an invalid nozzle, and all of the recording heads are Step of determining whether the nozzle is an effective nozzle or an invalid nozzle (determination step) Step (b): a step of creating an individual pattern for ink discharge control in one opening (individual pattern creation step) @Step (c): Pairing in the aforementioned steps (a) the nozzle group determined to be the effective nozzle is inserted into the individual pattern obtained in the above step (b), and a certain number of nozzles at both end portions of the recording head among the plurality of nozzles of the respective recording heads are registered as Shift space, a step of correcting the deviation of the nozzle arrangement direction of the recording heads by using the shift space to form a full print pattern (full print pattern creation step) 2) Color as in the first item of the patent application A method of manufacturing a filter, wherein the step (a) is a determination of an effective nozzle or an ineffective nozzle, by whether or not the following formula (1) is satisfied. <Mathematical Formula 1> D<(AxE)%(B + C)<(BD) (1) (In the mathematical formula (1), "%" indicates the remainder when the calculation is performed, and the symbol A is the nozzle number (1 to N). The integer) 'B is the length of the opening side' C is B Μ width, D is the remaining width, E is the nozzle spacing). 3) The method of manufacturing a color filter according to 1) or 2) wherein the deviation of the nozzle array direction -8 - 201033654 of each of the recording heads of the aforementioned step (c) is detected. [Effect of the Invention] According to the present invention, it is easy to correct the deviation of the recording position in the nozzle array direction (sub-scanning direction) between the plurality of recording heads, and it is not necessary to move the mounting position of the recording head itself, so that it is not necessary to waste a lot of time and cost. The color filter can be manufactured with high precision and fineness in a mobile device having a complicated mobile device or a head. [Embodiment] (Method for Producing Color Filter According to Ink Jet Method) A method of manufacturing a color filter according to the present invention, wherein a plurality of recording heads having nozzle rows in which a plurality of nozzles are arranged in a sub-scanning direction are arranged in plural The head unit formed in the main scanning direction is scanned on the substrate, and the ink discharged from the nozzle is caused to fall on the opening of the cell which is partitioned by the black matrix formed on the substrate. A method of manufacturing a color filter will be described in detail below with reference to the drawings. In this example, a method of manufacturing a color filter composed of three colors of RGB is described. The invention is not limited to this example, and can be applied to, for example, a method of producing a printed matter composed of four colors of yellow, magenta, cyan, and black (YMCK). Fig. 1 is three stacked in parallel in the nozzle direction. A sketch of the head unit formed by the recording heads (heads 1, 2, 3). Recording shots 1, 2, and 3 respectively spout 201033654. The inks of the respective colors of R, G, and B (the order may not be the same). Arrangement, It can also be one of the columns shown in Fig. 1, or a plurality of columns. In the case of a plurality of columns, by arranging the nozzles in the direction in which they are arranged, the nozzle spacing in appearance can be reduced, and the printing can be finer. When the ink is applied to the substrate, the head unit is scanned (main scanning) in the scanning direction X (main scanning direction) to discharge the ink onto the substrate. In this case, as shown in FIG. 1, the plurality of nozzles of each recording head Usually, the recording heads provided in the nozzle row are arranged side by side in the main scanning direction. As the recording head, a multi-drop which can set the amount of ink discharged from each nozzle can be used. In the case of using such a multi-drop method, the amount of ink discharged into the opening can be changed depending on the position where the ink is dropped, and color mixing can be prevented. Fig. 2 is a schematic view showing a substrate on which a black matrix (BM) is formed. The openings of the cells separated by the lattice-shaped black matrix are regularly arranged. The openings of the cells are provided with ink of respective colors to become sub-pixels. Generally, the openings In the rectangular shape, the row of the openings arranged in the longitudinal direction of the rectangle (the row of the openings 1 , the row of the openings 2 , and the row of the openings 3 ) are respectively provided with ink of the same color. In the case of a light sheet, the openings 1, 2, and 3 are respectively a unit of the R group, the G column, and the B column (the order is not necessarily the same). The head unit shown in Fig. 1 is scanned to be on the substrate. The opening portion is provided with a predetermined ink, and the head unit scans the end portion of the substrate by -10 to 201033654. In one direction (the main scanning direction is performed) (main scanning), the ink is applied to the opening portion on the substrate at the same time. The opposite end is in the direction orthogonal to the scanning direction (sub-scanning direction), and the sub-scan corresponds to the length of the head width. Then, in the opposite direction (backhaul) from the main scanning direction, the head unit main scanning is simultaneously performed simultaneously. The ink is applied to the opening on the substrate. By repeating such an operation, it is possible to apply ink to the entire large-sized substrate using a small head unit. Further, even in the case of substrates of various sizes, φ does not need to exchange nozzles, and ink is applied to the entire body to manufacture a color filter. The direction in which the head unit is scanned (main scanning direction) is preferably formed in the short side direction of the opening portion of the substrate. In this case, it is preferable to arrange the head unit such that the arrangement direction of the nozzle rows coincides with the longitudinal direction of the opening. By arranging the head unit and the substrate in this manner, the scanning direction is set, and the nozzle can be used efficiently, and productivity can be improved. In the above description, the example in which the head unit is scanned by fixing the substrate is described. However, the method of scanning the substrate and applying ink to the fixed head unit is also included in the present invention. (2) Step (a): Determination step of the effective nozzle and the ineffective nozzle In the present invention, when the head unit is scanned to apply ink to the substrate, the printing pattern of the counter nozzle of each recording head is registered in advance (ink Spit control data). A print pattern (full print pattern) of one head unit can be created by first applying a print pattern (individual pattern) corresponding to one opening portion by applying (inserting) it to all the nozzles of the head unit. However, among the nozzles of the respective recording heads, there are nozzles in which ink is dropped at the position of the opening, and nozzles that do not fall at the position of the opening. Fig. 3 is a view showing a position where ink discharged from a nozzle row of a recording head is dropped on an opening row. As can be seen from Fig. 3, when ink is ejected from all the nozzles, depending on the positional relationship between the opening and the nozzle, the discharged ink that is appropriately dropped in the opening may fall on the boundary line between the opening and the BM or the BM. Spit out the ink. However, falling on the boundary between the opening and the BM or on the BM will cause the color to become mixed, and the ink will be wasted, so it should be avoided. Therefore, in the present invention, it is determined in advance whether or not each of the nozzles is a nozzle (effective nozzle) that is dropped at the position of the opening, or a nozzle that does not fall at the position of the opening, that is, falls. A nozzle (ineffective nozzle) at a position on the boundary line between the opening portion and the BM or on the BM without discharging the ink from the invalid nozzle. That is, the method of the present invention includes: among the plurality of nozzles of the respective recording heads, a nozzle that is at a position falling on the opening is an effective nozzle, and a nozzle that does not fall at a position of the opening is an invalid nozzle. Each recording head determines a valid nozzle or an invalid nozzle for all the nozzles (decision step: step (a)). The specific method of the determination is not particularly limited. For example, it is possible to determine whether it is an effective nozzle or an invalid nozzle by the following formula (1) for the N nozzles on the head from the length of the BM width and the length of the opening side. Mathematical Formula 1> -12- 201033654 D<(AxE)%(B + C)<(BD) (1) wherein, in the above mathematical formula (1), "%" represents the remainder of the division, in addition The symbols A to E represent the following meanings. A: Nozzle number (integer of 1 to N) B: Length of the side of the opening C: BM width D: width and width φ E : Nozzle interval, when the effective nozzle or the invalid nozzle is determined by the above formula (1) The premise is that it is necessary to make the coordinate origin of the position at which the nozzle is displayed coincide with the origin of the coordinate system in the longitudinal direction of the opening divided by the black matrix. A to E of the above mathematical expression (1) are expressed by coordinates (distance) from the origin (the coordinate system showing the coordinate system of the position of the nozzle and the origin of the coordinate system in the longitudinal direction of the opening divided by the black matrix). Hereinafter, a method of determining whether it is a φ effective nozzle or an invalid nozzle by the above formula (1) will be described with reference to Fig. 4 . The nozzle number A of the above formula (1) is a recording head having 1 to N nozzles arranged in a line at one end, and indicates a nozzle number in the case where the number is sequentially given from the one end (again, overlapping with the origin) The nozzle is No. 0). That is, "(AxE)" indicates the position (the distance from one end) in the recording head of the nozzle of any (nozzle No. A). Further, "(B + C)" represents the length of the opening in the direction of the opening of the group of the black matrix (BM) on the glass substrate and the one of the openings defined by the opening. The distance (AxE) from the end of the nozzle to the nozzle No. A is divided by -13-201033654 (Β + C), and becomes the nearest opening end of the origin side until the nozzle number A The distance F from the nozzle (F in Fig. 4) ((AxE)%(B + C) = F). For example, the distance from the origin to the nozzle of the nozzle 7' of the 14th (nozzle number 14) from one end in Fig. 4 is (14 χΕ), and in the case of the example of Fig. 4, it is divided by (Β + C). After the distance of 2 times, that is, (14 >< £) is divided by (Β + C) as F (refer to Figure 4)❶
爲了使由任意的噴嘴（噴嘴編號A)吐出的具有液滴 ⑬ 寬幅的墨水不被賦予於開口部與黑矩陣之邊界線上而著實 落下於開口部內，噴嘴編號A之相當於F的長度，必須比 〇還長，且比開口部長邊的長度B還要短（0<F<B ) 。F 滿足此條件的場合，由該噴嘴吐出的墨水落下於該開口部 內。 但是此處，被賦予基板的墨水的液滴具有大小，所以 僅使由頭的一端至噴嘴編號 A的噴嘴爲止的距離爲， (ΑχΕ)時，F不滿足前述條件「〇<F<B」，也會有使由噴嘴 G 編號A之噴嘴吐出的墨水落下於開口部與黑矩陣之邊界線 上的情形。因此，在數學式（1 )作爲液滴寬幅有必要考 慮「裕度寬幅D」。 考慮此裕度寬幅D的話，爲了使由噴嘴編號A之噴 嘴吐出的具有液滴寬幅的墨水不被賦予於開口部與黑矩陣 之邊界線上而著實落下於開口部內，噴嘴編號A之F ’必 須比原點側之裕度寬幅（D )還要長，且距離原點遠之側 的裕度寬幅比由開口部長邊的長度減去後之値（B-D )還 14- 201033654 . 要短。亦即，任意之噴嘴成爲有效噴嘴之更嚴密的條件， 爲「D<F<(B-D)」。F如前所述爲（AxE)%(B + C) = F，所以 如前述數學式（1 )那樣，「D<(AxE)%(B + C)<(B-D)」爲 有效噴嘴的條件。 例如F成爲比D還小的場合，由噴嘴編號A之噴嘴 吐出的具有液滴寬幅之墨水變成落下於開口部的原點側之 端之正上方或是黑矩陣上。此外，例如F成爲比（B-D)還 φ 大的場合，由噴嘴編號A之噴嘴吐出的具有液滴寬幅之墨 水變成落下於開口部的離原點較遠側之端之正上方或是黑 矩陣上。因而，吐出墨水均由開口部擠出而落下，不能著 實對開口部內賦予墨水，所以該噴嘴編號A之噴嘴不能作 爲有效噴嘴。 如此般，滿足前述數學式（1)的場合，噴嘴編號A 之噴嘴係落下於開口部的適切位置者，被判定爲有效噴嘴 。不能滿足前述數學式（1)的場合，被判定爲無效噴嘴 Φ 。接著，藉由針對噴嘴編號1〜N進行此判定，可以針對 所有的噴嘴來判定其爲有效噴嘴還是無效噴嘴。 前述數學式（1)中，B、C、E可以因應於頭單元的 設計尺寸而適當選擇。此外，D係考慮墨水的液滴大小而 決定之値，只要與實際的墨水滴的尺寸同等程度者即可沒 有特別限制，較佳者爲由20〜5 Ομιη之範圍內選擇。 又，根據前述數學式（1 )之判定，噴嘴間隔爲一定 的場合可以利用，但如果噴嘴間隔有無法容許的程度的差 異的場合，可以藉由試印而正確測定來自各噴嘴的墨水之 -15- 201033654 落下位置，可以針對一個一個噴嘴判定其爲有效噴嘴或無 效噴嘴。 (3 )步驟（b ):個別圖案製作步驟 於本發明，製作對各記錄頭之全噴嘴之列印圖案時， 於前述步驟（a)把各記錄頭之噴嘴群分爲有效噴嘴群與 無效噴嘴群，同時製作供控制關於一個開口部的墨水吐出 之個別圖案（步驟（b):個別圖案製作步驟），於落下 @ 在一個開口部的位置之有效噴嘴群***該個別圖案。 作爲相當於一個開口部的墨水吐出控制資料之個別圖 案，係決定進入一個開口部的噴嘴樹、各噴嘴之墨水吐出 量、應賦予一個開口部之墨水量等而作成的。 又，沒有必要使進入一個開口部的有效噴嘴（個別圖 案內之有效噴嘴）之全部都吐出墨水，個別圖案內之有效 噴嘴之中亦可存在實際吐出墨水的噴嘴（以下稱爲「吐出 噴嘴」）與不吐出墨水的噴嘴（以下稱爲「非吐出噴嘴」 G )。吐出噴嘴的位置及數目，只要考慮噴嘴數目 '各噴嘴 的墨水吐出量、應賦予該開口部的墨水量等而決定即可。 個別圖案內的噴嘴數目，使用具有噴嘴間隔一定的噴 嘴列之記錄頭的場合，可以由噴嘴間隔與開口部長邊的長 度算出。以噴嘴間隔除開口部長邊的長度，所得到的値的 整數部分爲個別圖案內的噴嘴的最大數目。實際之個別圖 案內的吐出噴嘴數目，可以設定爲比此最大數還要小。例 如，進入一個開口部的噴嘴的最大數目爲5的場合’可以 -16- 201033654 . 不使用兩端的噴嘴作爲非吐出噴嘴，而把個吐出噴嘴配置 於5個噴嘴的中央。 作爲個別圖案的具體製作方法，例如對吐出噴嘴指定 「1」’對非吐出噴嘴指定「0」。無效噴嘴當然成爲非吐 出噴嘴’作爲資料保持著「〇」。例如，個別圖案內之噴 嘴數爲5，所有的噴嘴爲吐出噴嘴的場合，資料如「11111 J那樣’表示吐出噴嘴的資料「1」變成排列5個。沒有 ο 必要使所有的噴嘴爲吐出噴嘴的場合，例如可以像「 10101」這樣，僅使5個有效噴嘴之中的3個噴嘴作爲吐 出噴嘴來使用。 此外，使用前述之多滴方式之頭的場合，也可以對吐 . 出噴嘴細微地指定吐出的墨水量。例如在8色階之多滴方 式噴嘴的場合，可以對非吐出噴嘴設定「0」，而對吐出 噴嘴設定「1」〜「7」之7階段的吐出墨水量。與前述同 樣，個別圖案內之噴嘴數爲5個的場合，也可以如「 ® 73 1」那樣，對開口部的中央部賦予較多的墨水，而於 開口部的端部賦予較少量的墨水地進行設定。 此外，即使個別圖案內的噴嘴的最大數目爲5，也可 以不使用兩端而如「363」這樣採3個噴嘴所構成的個別 圖案，配置於5個噴嘴的中央。亦即此與「03 630」同義 (4 )步驟（c ):全列印圖案製作步驟 在本發明，前述步驟（a)之一個記錄頭之對所有噴 -17- 201033654 嘴之有效噴嘴與無噴嘴的判定作業及根據對在前述步驟（ . b)所製作的相當於一個開口部的有效噴嘴群之個別圖案 ，製作對各記錄頭之所有噴嘴&列印圖案。 亦即，各記錄頭之所有噴嘴（噴嘴編號1〜N)之中 ，對有效噴嘴群***前述個別圖案’登錄各噴嘴之墨水吐 出控制資料。無效噴嘴當然***「〇」而登錄爲非吐出噴 嘴。 如此進行針對N個噴嘴全體完成列印圖案’但在本發 @ 明，把N個噴嘴之中，兩端部之噴嘴之一定數目確保爲移 位空間是很重要的。此移位空間’通常作爲非吐出噴嘴來 看待。作爲移位空間，只要確保一部份的噴嘴，或是藉由 噴嘴間隔與對頭單元之頭的安裝精度來適當調整即可。 如此進行’包含兩端部之移位空間’針對N個噴嘴全 體記錄各個之墨水吐出控制資料’藉此完成針對一個記錄 頭之全列印圖案。由全列印圖案’除去兩端部的移位空間 之部分，被稱爲實際列印圖案。 Θ 接下來說明補正被搭載於頭單元的複數記錄頭之針對 噴嘴排列方向之偏離的方法。首先’在使各記錄頭固定於 頭單元的狀態下進行試印。試印亦可使用專用之非記錄媒 體。 藉由試印檢測出來自各記錄頭之基準噴嘴（通常爲左 右之任一之末端噴嘴）之墨水落下位置。由檢測出的位置 與頭單元的座標，求出各記錄頭之基準噴嘴之絕對座標。 由各記錄頭之基準噴嘴之絕對座標’算出針對噴嘴排列方 -18- 201033654 . 向之各記錄頭之偏離量，使用噴嘴間隔値求出各記錄頭之 偏離量。 亦即，有「移位量=偏離量+噴嘴間隔」之關係。 於先前製作之全列印圖案，利用移位空間而使實際列 印圖案偏離所求得的移位量之部分，補正針對各噴嘴之全 列印圖案而使其確定。 例如，各記錄頭間之噴嘴排列方向之偏離量’若是噴 φ 嘴間隔的W2以上，3/2以下的話，使實際列印圖案移位1 個之確保爲移位空間之噴嘴以進行補正。 如此般，根據本發明之方法，可以使各記錄頭間之噴 嘴排列方向之偏離量，補正成爲噴嘴間隔的二分之一以內 〇 又，與噴嘴排列方向直交的方向偏離，亦即在前述之 例之主掃描方向的偏離，可以藉由列印計時的調整而補正 。此於根據噴墨方式之技術，係屬通常方法而廣爲人知。 Φ 〔產業上利用可能性〕 根據本發明，複數記錄頭間之噴嘴排列方向的記錄位 置偏離的補正變得容易，沒有必要移動記錄頭之安裝位置 自身，所以不需浪費很多時間與成本於具備複雜的移動機 夠之裝置或者頭的移動作業等，可以高精度地製造彩色濾 光片。 【圖式簡單說明】 -19- 201033654 圖1係由噴嘴側所見之被搭載於本發明的方法之頭單 元的記錄頭之槪略圖。 圖2係顯示被形成黑矩陣（BM)之基板之槪略圖。 圖3係顯示開口部與墨水之落下位置之槪略圖。 圖4係使用數學式（1)判定其爲有效噴嘴或無效噴 嘴的方法之模式圖。 【主要元件符號說明】 @ 圖中之各符號係以下所示之物。 1:頭單元（head unit) 21 :頭（head ) 1 22 :頭 2 23 :頭 3 3 :噴嘴 41 :開口部1 42 :開口部2 ❹ 43 :開口部3 5 :黑矩陣（BM ) 6 :非吐出噴嘴 7 :吐出噴嘴 7’：噴嘴編號14之噴嘴 X -掃描方向 •20-In order to prevent the ink having the wide width of the liquid droplets 13 discharged from an arbitrary nozzle (nozzle No. A) from being placed on the boundary line between the opening and the black matrix, the ink of the nozzle No. A corresponds to the length of F. It must be longer than 〇 and shorter than the length B of the opening minister (0<F<B). F When this condition is satisfied, the ink discharged from the nozzle falls into the opening. However, since the droplet of the ink to be applied to the substrate has a size, the distance from the one end of the head to the nozzle of the nozzle No. A is (,), F does not satisfy the above condition "〇<F<B Further, there is a case where the ink discharged from the nozzle of the nozzle G No. A is dropped on the boundary line between the opening and the black matrix. Therefore, it is necessary to consider the "margin width D" as the liquid crystal width in the mathematical formula (1). When the margin width D is considered, the ink having the liquid droplet width discharged from the nozzle of the nozzle No. A is not dropped on the boundary line between the opening and the black matrix, and is actually dropped in the opening, and the nozzle number A is F. 'It must be longer than the margin width (D) on the origin side, and the margin width farther from the origin is smaller than the length of the opening side (BD) 14-201033654. Be short. That is, any nozzle is a more stringent condition for the effective nozzle, and is "D<F<(B-D)". F is (AxE)%(B + C) = F as described above, so "D<(AxE)%(B + C)<(BD)" is an effective nozzle as in the above mathematical formula (1) condition. For example, when F is smaller than D, the ink having the liquid droplet width discharged from the nozzle of the nozzle No. A is dropped directly above the origin side of the opening or on the black matrix. Further, for example, when F is larger than (BD), the ink having the liquid droplet width discharged from the nozzle of the nozzle No. A becomes directly above the end of the opening farther from the origin or black. On the matrix. Therefore, the discharged ink is ejected and dropped by the opening portion, and the ink in the opening portion cannot be reliably applied. Therefore, the nozzle of the nozzle No. A cannot be used as an effective nozzle. As described above, when the mathematical expression (1) is satisfied, the nozzle of the nozzle No. A falls to the appropriate position of the opening, and is determined to be an effective nozzle. When the above mathematical formula (1) cannot be satisfied, it is determined as the invalid nozzle Φ. Then, by performing this determination for the nozzle numbers 1 to N, it is possible to determine whether it is an effective nozzle or an invalid nozzle for all the nozzles. In the above mathematical formula (1), B, C, and E can be appropriately selected depending on the design size of the head unit. Further, D is determined in consideration of the droplet size of the ink, and is not particularly limited as long as it is equal to the size of the actual ink droplet, and is preferably selected from the range of 20 to 5 Ο μιη. Further, according to the judgment of the above formula (1), the nozzle interval can be used. However, if the nozzle interval is unacceptably different, the ink from each nozzle can be accurately measured by trial printing - 15- 201033654 Drop position, which can be judged as a valid nozzle or an invalid nozzle for one nozzle. (3) Step (b): In the individual pattern forming step, when the printing pattern for the entire nozzle of each recording head is produced, the nozzle group of each recording head is divided into the effective nozzle group and the invalid in the above step (a). In the nozzle group, an individual pattern for controlling ink discharge for one opening portion is simultaneously produced (step (b): individual pattern forming step), and the individual pattern is inserted in the effective nozzle group at the position of one opening portion. The individual pattern of the ink discharge control data corresponding to one opening is determined by the nozzle tree entering one opening, the amount of ink discharged from each nozzle, the amount of ink to be supplied to one opening, and the like. Further, it is not necessary to discharge the ink from all of the effective nozzles (the effective nozzles in the individual patterns) that enter one opening, and the nozzle that actually discharges the ink may be present in the effective nozzles in the individual patterns (hereinafter referred to as "discharge nozzle"). ) A nozzle that does not discharge ink (hereinafter referred to as "non-discharge nozzle" G). The position and the number of the discharge nozzles may be determined in consideration of the number of nozzles, the amount of ink discharged from each nozzle, the amount of ink to be applied to the opening, and the like. When the number of nozzles in the individual pattern is used as a recording head having a nozzle row having a constant nozzle interval, the nozzle interval and the length of the opening side can be calculated. The length of the open side of the opening is divided by the nozzle interval, and the resulting integer portion of the flaw is the maximum number of nozzles in the individual pattern. The number of spout nozzles in the actual individual pattern can be set to be smaller than this maximum number. For example, when the maximum number of nozzles entering one opening is 5, it can be -16-201033654. The nozzles at both ends are not used as the non-discharge nozzles, and one discharge nozzle is disposed at the center of the five nozzles. As a specific production method of the individual patterns, for example, "1" is designated for the discharge nozzle, and "0" is designated for the non-discharge nozzle. The ineffective nozzle is of course a non-discharge nozzle, and "〇" is held as data. For example, when the number of nozzles in the individual pattern is 5, and all the nozzles are the discharge nozzles, the data "11111 J" indicates that the data "1" of the discharge nozzle has become five. When it is necessary to make all the nozzles as the discharge nozzles, for example, as in "10101", only three of the five effective nozzles can be used as the discharge nozzles. Further, when the head of the above-described multi-drop method is used, the amount of ink discharged can be finely specified for the spout nozzle. For example, in the case of a multi-drop nozzle of 8 gradation, "0" can be set for the non-discharge nozzle, and the amount of discharged ink of the seven stages of "1" to "7" can be set for the discharge nozzle. In the same manner as described above, when the number of nozzles in the individual pattern is five, a large amount of ink may be applied to the central portion of the opening as in "® 73 1", and a smaller amount may be applied to the end portion of the opening. Set the ink. Further, even if the maximum number of nozzles in the individual patterns is 5, an individual pattern composed of three nozzles such as "363" can be used without using both ends, and can be disposed at the center of the five nozzles. That is, synonymous with "03 630" (4) Step (c): The full print pattern making step is in the present invention, and the effective nozzle of one of the recording heads of the above step (a) for all nozzles -17-201033654 is not The nozzles are determined and the print patterns for all the nozzles of each of the recording heads are created based on the individual patterns of the effective nozzle groups corresponding to one opening made in the above step (.b). In other words, among all the nozzles (nozzle numbers 1 to N) of the respective recording heads, the ink discharge control data of each nozzle is registered in the effective nozzle group. Of course, the invalid nozzle is inserted into "〇" and registered as a non-discharge nozzle. In this way, the printing pattern is completed for all of the N nozzles. However, in the present invention, it is important to ensure that a certain number of nozzles at both ends of the N nozzles are the displacement spaces. This displacement space 'is usually viewed as a non-discharge nozzle. As the displacement space, it is only necessary to ensure a part of the nozzles or to appropriately adjust the mounting accuracy of the nozzles and the head of the head unit. Thus, the "displacement space including both end portions" is performed to collectively record the respective ink discharge control data for the N nozzles, thereby completing the full print pattern for one recording head. The portion of the displacement space at both end portions removed by the full print pattern ' is referred to as an actual print pattern. Θ Next, a method of correcting the deviation of the nozzle array direction of the complex recording head mounted on the head unit will be described. First, a trial print is performed in a state where each recording head is fixed to the head unit. A dedicated non-recording medium can also be used for trial printing. The ink drop position from the reference nozzle of each recording head (usually the end nozzle of either of the left and right) is detected by trial printing. From the detected position and the coordinates of the head unit, the absolute coordinates of the reference nozzles of the respective recording heads are obtained. From the absolute coordinates of the reference nozzles of the respective recording heads, the amount of deviation from each of the recording heads was calculated using the nozzle spacing 针对 for the nozzle array -18 - 201033654. That is, there is a relationship of "shift amount = deviation amount + nozzle interval". In the previously produced full print pattern, the actual print pattern is shifted from the portion of the obtained shift amount by the shift space, and the full print pattern for each nozzle is corrected to be determined. For example, if the amount of deviation in the direction in which the nozzles are arranged between the recording heads is equal to or greater than W2 of the nozzle φ, and 3/2 or less, the actual print pattern is shifted by one to ensure that the nozzles of the shift space are corrected. In this way, according to the method of the present invention, the amount of deviation of the nozzle arrangement direction between the recording heads can be corrected to become one-half of the nozzle interval, and the direction orthogonal to the nozzle arrangement direction is deviated, that is, in the foregoing The deviation of the main scanning direction of the example can be corrected by the adjustment of the printing timing. This is well known in the art according to the technique of the ink jet method. Φ [Industrial Applicability] According to the present invention, it is easy to correct the deviation of the recording position in the nozzle array direction between the plurality of recording heads, and it is not necessary to move the mounting position of the recording head itself, so that it is not necessary to waste a lot of time and cost. The color filter can be manufactured with high precision by a complicated mobile device or a moving operation of the head. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic diagram of a recording head mounted on a nozzle side and mounted on a head unit of the method of the present invention. Fig. 2 is a schematic view showing a substrate on which a black matrix (BM) is formed. Fig. 3 is a schematic view showing the position where the opening portion and the ink are dropped. Fig. 4 is a schematic view showing a method of determining whether it is an effective nozzle or an invalid nozzle using Mathematical Formula (1). [Explanation of main component symbols] @ Each symbol in the figure is the following. 1: head unit 21: head 1 2 : head 2 23 : head 3 3 : nozzle 41 : opening 1 42 : opening 2 ❹ 43 : opening 3 5 : black matrix (BM ) 6 : Non-discharge nozzle 7 : Discharge nozzle 7': Nozzle No. 14 nozzle X - Scan direction • 20-