TWI897948B - Inkjet printing device and inkjet printing method - Google Patents

Abstract

An inkjet printing device includes a printing original image holder that holds original image data indicating the landing positions of ink droplets on a printing substrate. The device also includes a normalized two-dimensional offset information holder that holds either a two-dimensional offset correction amount or an added value. The two-dimensional offset correction amount is used to correct the two-dimensional offset of the landing position of ink droplets from each nozzle, and the added value is the sum of the two-dimensional offset correction amount and the one-dimensional offset correction amount used to correct the one-dimensional offset of the landing position of the ink droplets at the ejection time of each ink droplet. In addition, it has: a variable delay device for each nozzle, which generates ejection on/off instruction information to change the ejection timing based on the original image data and the added value; and a drive signal selector, which turns on/off the ejection of ink droplets for each nozzle based on the ejection on/off instruction information.

Description

Inkjet printing device and inkjet printing method

This disclosure relates to an inkjet printing device and an inkjet printing method.

Conventionally, there is a known method for forming the organic light-emitting layer of an organic EL (electroluminescence) display panel using an inkjet printing device. Examples of such methods include forming low-molecular-weight organic materials or high-molecular-weight organic materials using a solvent coating method.

A typical method for forming an organic light-emitting layer using a solvent coating method involves using an inkjet printer to eject droplets of ink containing an organic light-emitting material (hereinafter referred to as "ink droplets") onto the pixel areas of a display substrate. The ejected ink droplets contain both the organic light-emitting material and a solvent.

A typical inkjet printer includes an inkjet head with multiple nozzles. The inkjet printer controls the positional relationship between the nozzles and the printed material (workpiece) while ejecting ink droplets from the nozzles, applying the ink droplets to the printed material.

For example, Japanese Patent Publication No. 2003-266669 (hereinafter referred to as "Patent Document 1") discloses an inkjet printing device that deposits ink droplets onto a substrate in a manner that spreads in equal directions within a recessed area called a pixel region, thereby forming pixels with a predetermined line width.

When using the inkjet printing device disclosed in Patent Document 1 to form a plurality of pixels with a predetermined line width, if the ink droplet placement is offset, the required amount of ink droplets cannot be applied to the target location, i.e., the self-pixel area. Consequently, the amount of ink applied to the self-pixel area is insufficient. Furthermore, if ink droplets intended for the self-pixel area are ejected into adjacent pixels, the adjacent pixels will have an excess of ink.

In other words, due to the excess or insufficient amount of ink, the thickness of the formed light-emitting layer may be different in the pixel area and other pixel areas.

Furthermore, if the ink droplets ejected into adjacent pixel regions and the ink droplets ejected into the own pixel region are different luminescent inks, and if any ink droplets land between the own pixel region and the adjacent pixel region, the landed ink droplet will connect the two pixel regions. This can create mixed-color luminescence in the two pixel regions.

Causes of ink droplet landing position deviation include, for example, yawing (swaying in the rotational direction relative to the display substrate's travel direction) and pitching during scanning of the display substrate. Other causes of droplet landing position deviation include nozzle placement deviation (hereinafter referred to as "nozzle placement deviation") and angular deviation of ink droplets ejected from the nozzle relative to the vertical direction (hereinafter referred to as "ejection angle deviation").

Patent Document 2, for example, discloses a method for eliminating the aforementioned deviation in droplet position. Specifically, Patent Document 2 discloses a method for varying the timing at which ink droplets are ejected from the nozzles of an inkjet head according to the degree of deflection generated when scanning a display substrate.

More specifically, for example, Japanese Patent Publication No. 2007-152215 (hereinafter referred to as "Patent Document 2") first places rulers and encoders on the left and right ends of a table (linear motion unit) on which a display substrate is mounted. The position of each table end, obtained by the rulers and encoders, is then used to calculate the amount of deflection. Furthermore, the method discloses correcting the timing of ink droplet ejection from all nozzles to match the calculated amount of deflection.

The following describes the method for eliminating the deviation of the dripping position described in Patent Document 2 using Figures 11 and 12.

Figure 11 (similar to Figure 4 of Patent Document 2) illustrates the positional displacement of the nozzle's discharge orifice caused by substrate deflection, as disclosed in Patent Document 2. Specifically, Figure 11 shows two left and right rulers mounted on the worktable of the movable platform, as well as the positions of ink droplets ejected from the nozzle (nozzle discharge orifice 69 in the figure) when the worktable passes directly under the inkjet head (head 67 in the figure). The two rulers are the p-side ruler, indicated by p0...pn, and the q-side ruler, indicated by q0...qn.

As shown in Figure 11, at time T0, at positions p0 and q0, ink droplets are ejected from each of nozzle orifices 69-1 through 69-n. Furthermore, at time Tn, at positions pn and qn-1, ink droplets are ejected from each of nozzle orifices 69-1 through 69-n.

At this time, the phenomenon of a one-scale (pulse) delay on the p-side ruler shown in Figure 11 is called skew. In other words, the skew amount (degree of skew) can be obtained from the p-side and q-side rulers shown in Figure 11.

Furthermore, Figure 12 (same as Figure 7 of Patent Document 2) shows various components used to correct the timing of ink droplet ejection from the inkjet head.

As shown in Figure 12, ruler reading heads A65-1 and B65-2 respectively read the pulses (positions) of the p-side and q-side rulers shown in Figure 11. Position encoders A91-1 and B91-2 count the number of pulses read for each ruler.

The waveform generator 81 measures the time offset (the delay per scale mark on the p-side scale) based on the number of pulses per scale. It then generates the ejection on/off timings for all nozzles based on the time offset and pattern data (printed image). Furthermore, the waveform generator 81 controls the timing of ink droplet ejection from each nozzle based on the ejection on/off timings. This allows correction of ejection timing that is affected by the offset.

The following describes a method for correcting a drop position deviation and printing, as described in, for example, Japanese Patent Application Laid-Open No. 2015-138693 (hereinafter referred to as "Patent Document 3"), using Figures 13A to 15.

The method disclosed in Patent Document 3 first ejects ink droplets from all nozzles onto a droplet observation substrate having the same surface area as the display substrate. Next, the total amount of yaw, vertical wobble, nozzle arrangement deviation, and nozzle ejection angle deviation (hereinafter referred to as the "total drop deviation") is pre-measured based on the deviation of the ink droplet's landing position from the reference position. Furthermore, the printed image is deformed (corrected) based on the measured total drop deviation and the deviation in the negative direction (hereinafter referred to as the "total drop deviation correction amount"). Patent Document 3 discloses the above printing method.

Specifically, Figures 13A and 13B (similar to Figure 16 of Patent Document 3) illustrate the relationship between the printed substrate and the inkjet head during printing operations performed by platform movement.

As shown in Figures 13A and 13B, the printing device of Patent Document 3 includes: a printed substrate 1, exemplified by a display substrate having pixel regions 1c formed thereon; an inkjet head 3 having nozzles 3a; and a movable platform 4.

In addition, (a1), (a2), and (a3) of Figure 13A show the state without the deflection of the moving platform 4 and the rotation of the inkjet head 3.

Figure 13B shows a state where the front half of the inkjet head 3 is not rotating, while the rear half of the inkjet head 3 is rotating. Furthermore, Figure 13B (b1) shows a state where the movable platform 4 is rotating counterclockwise. Figure 13B (b2) shows a state where the movable platform 4 is rotating clockwise. Figure 13B (b3) shows a state where the movable platform 4 is scanning while rotating counterclockwise, thereby performing printing.

Figure 14 (similar to Figure 13 in Reference 3) shows the expected droplet positions when the printing operation shown in Figure 13B is performed with uniform pitch within the surface of a droplet observation substrate having the same area as the display substrate. In addition, in each pixel area 1c shown in Figure 14, the expected droplet positions of eight drops are indicated by black dots.

In addition, Figure 15 (same as Figure 14 in Reference 3) shows the original image data created based on the assumed drop positions shown in Figure 14. The original image data shown in Figure 15 is stored in the memory of the printing device.

The printing device first ejects ink from each nozzle of the inkjet head based on the stored original image data. At this point, the printing device corrects for any deviation in the overall droplet distribution, depositing eight drops of ink uniformly within each pixel area. This reduces color mixing and uneven light emission from the display.

However, the method of Patent Document 2 corrects the ejection timing of all nozzles at the ruler portion of the stage. Therefore, if the first deflection observed at the ruler portion differs from the second deflection on the display substrate on the stage, the droplet position will shift according to the difference between the first and second deflections.

In recent years, the trend toward shorter printing lines has been driven by larger and more chamfered display panels and increased production volumes. Consequently, the difference between the primary and secondary deflections tends to increase due to reduced rigidity caused by the increased platform area and increased print scan speeds.

Furthermore, the method of Patent Document 3 reflects the total drop offset correction amount based on the total drop offset (the sum of yaw, vertical yaw, nozzle arrangement offset, and discharge angle offset) across the entire printed image. Therefore, the method of Patent Document 3 can correct for drop position offset.

Furthermore, the total droplet offset is the sum of one-dimensional offset (the sum of nozzle placement offset and ejection angle offset) independent of stage movement, and two-dimensional offset (the sum of yaw and pitch) caused by stage movement. Therefore, the one-dimensional offset varies more frequently and to a greater extent over time than the two-dimensional offset. This is because the precision of a high-mass stage fluctuates smoothly, depending solely on temperature. In contrast, the ejection angle offset of a droplet of several picoliters (picoliters) varies depending on ink concentration or the number of ejections. Therefore, correction of the total droplet offset, which accounts for these different variations, must be performed based on the more frequent one-dimensional offset.

However, as the printed image becomes larger, processing time, such as the time required to reflect the drop aggregate offset correction amount and the time to write the original image data to memory, will increase.

Furthermore, as mentioned above, in recent years, the trend toward larger display panels, more chamfered corners, and increased production volume has led to a trend toward shortening printing production lines. Consequently, it has become difficult to create large printed images and write them into image memory within the time required to meet the required production volume.

This disclosure provides an inkjet printing device and inkjet printing method that can reduce processing time and correct the deviation of the droplet landing position of ink drops.

One aspect of the present disclosure is an inkjet printing device comprising: a moving platform for moving a printed substrate; and an inkjet head for ejecting ink droplets from a plurality of nozzles relative to the printed substrate. The inkjet printing device comprises a printing original image holder for holding original image data, wherein the original image data indicates the drop positions of ink droplets on the printed substrate. The inkjet printing device further comprises a normalized two-dimensional offset information holder for holding either a two-dimensional offset correction amount or an added value, wherein the two-dimensional offset correction amount is used to correct the two-dimensional offset of the drop position of the ink droplet from each nozzle, and the added value is the sum of the one-dimensional offset correction amount for correcting the one-dimensional offset of the drop position of the ink droplet at the ejection time point of each ink droplet and the one-dimensional offset correction amount. The system also includes a variable delay unit for each nozzle, which generates ejection on/off instruction information to change the ejection timing based on the original image data and the added value; and a drive signal selector, which turns ink droplet ejection on/off for each nozzle based on the ejection on/off instruction information. The one-dimensional offset is the sum of the offset in the nozzle configuration and the angular offset of the ejected ink droplets relative to the vertical direction. The two-dimensional offset is the sum of the yaw and vertical oscillations generated when scanning the printed circuit board.

One aspect of the present disclosure is an inkjet printing method performed using a moving stage and an inkjet head. The moving stage moves a printing substrate, and the inkjet head ejects ink droplets from a plurality of nozzles relative to the printing substrate. The inkjet printing method stores original image data and either a two-dimensional offset correction or an added value in separate memories. The original image data indicates the droplet locations of ink droplets on the printing substrate. The two-dimensional offset correction is used to correct the two-dimensional offset of the droplet locations for each nozzle. The added value is the sum of the two-dimensional offset correction and the one-dimensional offset correction used to correct the one-dimensional offset of the droplet locations at the time of each ink droplet ejection. Furthermore, based on the original image data and the added value, ejection on/off instruction information is generated to change the ejection timing. Ink droplet ejection is then turned on/off for each nozzle based on this ejection on/off instruction information. Furthermore, one-dimensional offset is the sum of the offset in the nozzle configuration and the angular offset of the ejected ink droplets relative to the vertical direction. Two-dimensional offset is the sum of the yaw and vertical motion generated during scanning of the printed circuit board.

According to the present disclosure, an inkjet printing device and inkjet printing method can be provided that can reduce processing time and correct deviations in the landing positions of ink drops.

1: Printed substrate

1a: Liquid membrane

1b: Next door

1c: Pixel area

1d: Drip tolerance area

1Dzure#2: One-dimensional offset pixel data

1e: Intra-pixel dripping area

2a: Printing and spitting out dot grid

2b: Point grid within the dripping area

2A-2A, 2B-2B, 2C-2C: Line

2Dzure#2: 2D offset pixel data

3: Inkjet Head

3a: Nozzle

4: Mobile Platform

5: Position detector

6: Output time point generator

7: Drive signal generator

8: Normalized one-dimensional offset information preserver

9: Normalized 2D offset information preserver

10: Printing original image holder

11: Adder for offset information of each nozzle

12: Variable delay device for each nozzle

13: Drive signal selector

20: Expected drop location

21: Actual dripping location

30a, 30b, 30c, 30d, 30e, 30f, 30g, 30h: Pixel holders

31a, 31b, 31c, 31d, 31e, 31f, 31g, 31h: Logical multiplier

32a, 32b, 32c, 32d, 32e, 32f, 32g, 32h: Partial Negative Logic Multiplier

33a, 33b, 33c, 33d, 33e, 33f, 33g, 33h: Logical adder

34: Decoder

39:Data entry department

41: Control Department

59: Head discharge control unit

65-1: Ruler reading head A

65-2: Ruler reading head B

67: Head

69, 69-1, 69-2,…69-n: Nozzle discharge hole

81: Waveform Generation Unit

85: Rectangular Wave Generator

87: Waveform Amplification Unit

89: Pattern Waveform Generation Unit

91-1: Position Encoder A

91-2: Position Encoder B

93: Time offset measurement part

95: Reference waveform selection section

97: Pattern Data Generation Department

100: Inkjet printing device

A65-1, B65-2: Ruler reading head

A91-1, B91-2: Position encoder

CLK:CK clock

LnX: Nozzle spacing in scanning direction

Lny, Lpx, Lpy: interval

ewx,ewy,Lwx,Lwy: length

p0, p1, p2, ... pn: p-side ruler

px: Printing resolution in scanning direction

py: Printing resolution in nozzle row direction

q0,q1,q2,…qn:q side ruler

Din#2,SEL#2: Select data

t: Time of discharge (moment)

T0, T1, T2,…Tn: time

X: Scan direction

x2~x8, x10~x16, x18~x24, x26~x35: Scanning direction coordinates

x1,x9,x17,x25:Boundary

Y: Nozzle direction

y1~y20: Coordinates of the nozzle's direction

#1, #2,…#20: Identification numbers of nozzle 3a

*CLR: signal

Figure 1 is a plan view of a display panel substrate according to embodiment 1 of the present disclosure.

Figure 2A is a cross-sectional view taken along line 2A-2A of Figure 1.

Figure 2B is a cross-sectional view taken along line 2B-2B of Figure 1.

Figure 2C is a cross-sectional view taken along line 2C-2C of Figure 1.

Figure 3 shows the relationship between the pixel area and the printed dot grid in embodiment 1.

Figure 4 is a block diagram showing an inkjet printing device according to embodiment 1.

FIG5 is a diagram showing the distribution of the deviation of the droplet position on the droplet droplet observation substrate in Embodiment 1.

Figure 6 shows the original image data of Implementation 1.

Figure 7 is a diagram showing the connections of the main components for performing the correction action of Implementation Form 1.

FIG8 is a diagram showing an example of the internal structure of the variable delay device of nozzle 2 among the variable delay devices of the nozzles in embodiment 1.

Figure 9 shows the internal signal state and the passage of time of the variable delay device of nozzle 2 in embodiment 1.

FIG10 is a diagram showing an example of the tilted state of the inkjet head in Embodiment 3 and Embodiment 4 of the present disclosure.

FIG11 is a diagram illustrating the positional displacement of the nozzle outlet caused by the deflection of the substrate disclosed in Patent Document 2.

Figure 12 is a diagram showing the components for correcting the timing of head ejection disclosed in Patent Document 2.

FIG13A is a diagram showing the relationship between the printing substrate and the inkjet head during the printing operation performed by moving the stage disclosed in Patent Document 3.

FIG13B is a diagram showing the relationship between the printing substrate and the inkjet head during the printing operation performed by moving the stage disclosed in Patent Document 3.

FIG14 is a diagram showing the hypothetical droplet positions when performing in-plane discharge at a uniform pitch on the droplet observation substrate disclosed in Patent Document 3.

FIG15 is a diagram showing printed image data produced based on the assumed drop position disclosed in Patent Document 3.

The following describes the embodiments of the present disclosure with reference to the drawings. Components common to the various figures are denoted by the same reference numerals, and descriptions of these components are omitted where appropriate.

(Implementation Form 1)

The following describes the printed circuit board 1 of embodiment 1 using FIG1 .

FIG1 is a plan view (top view) of a printed circuit board 1. The printed circuit board 1 is, for example, a substrate for a display panel.

As shown in Figure 1, a printed circuit board 1 of embodiment 1 includes a liquid repellent film 1a formed on the printed circuit board 1, partitions 1b formed on the liquid repellent film 1a, and pixel regions 1c separated by the liquid repellent film 1a and the partitions 1b. The partitions 1b extend in the nozzle row direction Y.

The partition walls 1b separate adjacent pixel areas 1c in the horizontal direction of the figure in the scanning direction X. The spacing Lpx in the scanning direction X is the distance between the centers of adjacent pixel areas 1c in the horizontal direction of the figure (the intersection of the one-point chain lines in the figure; the same applies hereinafter). The spacing Lpy in the nozzle row direction Y is the distance between the centers of adjacent pixel areas 1c in the vertical direction of the figure.

Lwx shown in Figure 1 is the length of pixel area 1c in the scanning direction X. Lwy is the length of pixel area 1c in the nozzle row direction Y. Pixel area 1c is a rectangular area with rounded edges extending from the center of pixel area 1c at ±Lwx/2 in the scanning direction X and ±Lwy/2 in the nozzle row direction Y.

The following describes the cross-sectional structure of the printed circuit board 1 using Figures 2A to 2C. Figures 2A to 2C are cross-sectional views of the printed circuit board 1.

Specifically, Figure 2A shows a cross section taken along line 2A-2A in Figure 1 . The cross section taken along line 2A-2A is a cross section of a pixel region 1c cut out in the scanning direction X. As shown in Figure 2A , the repellent film 1a does not exist in the pixel region 1c between adjacent partitions 1b.

Figure 2B shows a cross section taken along line 2B-2B in Figure 1 . The cross section taken along line 2B-2B is a cross section taken along the scanning direction X, where a region without pixel regions 1c is cut out. As shown in Figure 2B , a repellent film 1a is present in pixel regions 1c between adjacent partitions 1b.

Figure 2C shows a cross section taken along line 2C-2C in Figure 1 . This cross section cuts through pixel area 1c along the nozzle row direction Y. As shown in Figure 2C , pixel area 1c does not have partition walls 1b, but rather partially contains liquid repellent film 1a.

The following describes the relationship between the pixel area 1c and the print dot grid 2a using Figure 3.

FIG3 shows the relationship between the pixel area 1c and the print dot grid 2a. FIG3 also shows one pixel area 1c among the multiple pixel areas 1c shown in FIG1.

As shown in Figure 3, the pixel area 1c includes a drip tolerance portion 1d, an intra-pixel drip area 1e, etc.

The intra-pixel droplet area 1e has a length ewx in the scanning direction X and a length ewy in the nozzle row direction Y, and is the pixel area where ink droplets can drip. Ink droplets that land within a pixel are ejected into the intra-pixel droplet area 1e.

The droplet tolerance area 1d is the area obtained by subtracting the intra-pixel droplet area 1e from the pixel area 1c. The droplet tolerance area 1d is a margin designed to prevent ink droplets from deviating from the nozzle and being ejected from the pixel area 1c.

Figure 3 shows dotted lines extending in the nozzle row direction Y at widths corresponding to the printing resolution px in the scanning direction X (hereinafter referred to as "first dotted lines"). Also shown are dotted lines extending in the scanning direction X at widths corresponding to the printing resolution py in the nozzle row direction Y (hereinafter referred to as "second dotted lines"). The intersection of the first and second dotted lines corresponds to the print dot grid 2a. In other words, the print dot grid 2a can be the expected ink droplet landing positions ("expected droplet positions 20" described below).

The methods for determining the printing resolution px in the scanning direction X (hereinafter referred to as "scanning direction printing resolution px") and the printing resolution py in the nozzle row direction Y (hereinafter referred to as "nozzle row direction printing resolution py") will be described later.

The dot grid 2b within the droplet area is the print dot grid 2a arranged within the droplet area 1e within the pixel.

In the following, Embodiment 1 is explained using the example of a standard eight ink droplets ejected within the intra-pixel landing area 1e. In this case, the scanning direction printing resolution px and the nozzle row direction printing resolution py are determined by the relationship between the eight ink droplets, the intra-pixel landing area 1e, and the nozzle row direction spacing Lny (see Figure 4 ), described later. Specifically, in Embodiment 1, the scanning direction printing resolution px and the nozzle row direction printing resolution py are determined so that within the intra-pixel landing area 1e, a total of 12 dots are ensured: three dots in the scanning direction X and four dots in the nozzle row direction Y (see Figure 3 ).

In Figure 3, among the 12 droplet-area dot grids 2b, the eight droplet-ejected droplets are represented by black dots, while the four droplet-unejected droplets are represented by white circles. Furthermore, the eight droplet-ejected droplet grids 2b are arranged so that their coordinate centers are at the center of the pixel area 1c.

The following describes the structure of the inkjet printing apparatus 100 according to Embodiment 1 using FIG4 . FIG4 is a block diagram showing the structure of the inkjet printing apparatus 100 .

As shown in Figure 4, inkjet printing device 100 includes an inkjet head 3, a movable platform 4, a position detector 5, an ejection timing generator 6, a drive signal generator 7, a normalized one-dimensional offset information holder 8, a normalized two-dimensional offset information holder 9, a print original image holder 10, an offset information adder for each nozzle 11, a variable delay device for each nozzle 12, and a drive signal selector 13. Inkjet printing device 100 prints on a printed substrate 1.

The inkjet head 3 has a plurality of nozzles 3a that eject ink droplets. In Embodiment 1, two inkjet heads 3 are arranged along the nozzle row direction Y. The plurality of nozzles 3a are arranged along the nozzle row direction Y at intervals Lny. This interval Lny is also maintained at the junction of the two inkjet heads 3. In other words, the nozzle row direction printing resolution py shown in Figure 3 is the same as the interval Lny.

The moving platform 4 moves the mounted printed circuit board 1.

The position detector 5 generates a position information pulse signal. This position information pulse signal is obtained by converting the position information of the movable platform 4 on which the printed circuit board 1 is mounted. Furthermore, the position detector 5 outputs the generated position information pulse signal to the output timing generator 6.

The ejection timing generator 6 generates an ejection timing signal by frequency-dividing the position information pulse signal from the position detector 5 according to the preset scanning direction printing resolution px. Furthermore, the ejection timing generator 6 outputs the generated ejection timing signal to the drive signal generator 7, the normalized one-dimensional offset information holder 8, the print original image holder 10, the offset information adder 11 for each nozzle, and the variable delay device 12 for each nozzle. The ejection timing signal specifies the timing of generating the voltage waveform that drives the nozzles 3a of the inkjet head 3.

Furthermore, the discharge timing generator 6 begins outputting the discharge timing signal in accordance with the discharge start position information stored in the discharge timing generator 6. Furthermore, the discharge timing generator 6 outputs the discharge timing signal for the number of times indicated by the discharge count information stored in the discharge timing generator 6. Thereafter, the discharge timing generator 6 stops generating and outputting the discharge timing signal.

Furthermore, the ejection timing generator 6 generates a split timing signal and outputs it to the normalized two-dimensional offset information holder 9. Specifically, the split timing signal generated by the ejection timing generator 6 is a signal that counts and eliminates ejection timings based on the split width frequency information stored within the ejection timing generator 6. Furthermore, the split timing signal specifies the time at which the normalized two-dimensional offset is output to each nozzle offset information adder 11.

The drive signal generator 7 generates a drive waveform signal for ejecting ink droplets from the nozzles 3a of the inkjet head 3 based on the ejection timing signal from the ejection timing generator 6. Furthermore, the drive signal generator 7 outputs the generated drive waveform signal to the drive signal selector 13.

Normalized one-dimensional offset information holder 8 is a memory that stores normalized one-dimensional offset pixel data. Normalized one-dimensional offset information holder 8 outputs the stored normalized one-dimensional offset pixel data to each nozzle offset information adder 11. The normalized one-dimensional offset pixel data will be described later.

Normalized two-dimensional offset information holder 9 is a memory that stores normalized two-dimensional offset pixel data. Based on the segmentation timing signal from the output timing generator 6, normalized two-dimensional offset information holder 9 outputs the stored normalized two-dimensional offset pixel data to each nozzle offset information adder 11. The normalized two-dimensional offset pixel data will be described later.

The print original image holder 10 is a memory that stores original image data (print source data). The print original image holder 10 outputs the stored original image data to each nozzle variable delay device 12. The original image data instructs the ejection of ink droplets relative to the print dot grid 2a defined on the print substrate 1.

Each nozzle offset information adder 11 adds the following data for each output time point signal: the normalized one-dimensional offset pixel data from the normalized one-dimensional offset information holder 8 and the normalized two-dimensional offset pixel data from the normalized two-dimensional offset information holder 9. Furthermore, each nozzle offset information adder 11 outputs the addition result, i.e., the obtained offset information indicating the offset (delay) of each nozzle 3a, to each nozzle variable delay device 12.

Each nozzle variable delay device 12 generates ejection on/off instruction information based on the original image data from the print original image holder 10 and the offset information from the nozzle offset information adder 11, and outputs it to the drive signal selector 13. The ejection on/off instruction information indicates, for each nozzle 3a, whether ejection is on/off (active/inactive) for each dot grid 2a of the print ejection pattern.

The drive signal selector 13 outputs the drive waveform signal from the drive signal generator 7 based on the ejection on/off instruction information from each nozzle's variable delay device 12, turning each nozzle 3a on/off. This controls the ejection of ink droplets from each nozzle 3a.

Here, before printing on a printed circuit board 1, the inkjet printing apparatus 100 with the above-described configuration first deposits an ink droplet onto one side of a droplet observation substrate (not shown) of the same size as the printed circuit board 1. The inkjet printing apparatus 100 then measures the offset between the actual drop position and the target drop position on the droplet observation substrate. Furthermore, the inkjet printing apparatus 100 determines one-dimensional drop offset information and two-dimensional drop offset information based on the measured offset.

The following describes the deviation of the ink droplet landing position on the droplet landing observation substrate using Figure 5.

Figure 5 is a graph showing the distribution of the deviation of the landing position of ink droplets on the droplet landing observation substrate.

As shown in Figure 5 , the printing area on the droplet observation substrate is divided into four equal parts in the scanning direction X. Specifically, the boundaries of the printing area in the scanning direction X are, for example, x1, x9, x17, and x25. Furthermore, the interval between each division in the scanning direction X (hereinafter referred to as the "division interval BKW") is, for example, 40.0 μm. The inkjet printing device 100 stores the positions of these boundaries or the division interval BKW. Furthermore, the division interval BKW is stored in the ejection timing generator 6 as the division interval pixel BKGSW (described later) divided by the scanning direction printing resolution px.

As shown in FIG5 , the inkjet head 3 has, for example, 20 nozzles 3a, which are arranged at a printing resolution py in the nozzle row direction and are indicated by y1 to y20.

The inkjet printing device 100 ejects ink droplets from the nozzles 3a toward the pre-set expected drop locations 20. The expected drop locations 20 are set along the nozzle row direction Y, within the boundaries x1, x9, x17, and x25 of the printing area in the scanning direction X.

Here, the white circle in Figure 5 indicates the expected droplet location 20. Furthermore, the black dot in Figure 5 indicates the location 21 where the ejected ink droplet actually lands (hereinafter referred to as the "actual droplet location 21").

As shown in FIG5 , there is an offset between the expected drop position 20 and the actual drop position 21. Furthermore, in FIG5 , the arrow between the expected drop position 20 and the actual drop position 21 represents the offset.

The following describes the in-plane scanning direction offset of Implementation 1 in detail using Table 1A.

[Table 1A]

Table 1A shows a list of in-plane scanning direction deviations. The in-plane scanning direction deviations are the deviations between the expected drop position 20 and the actual drop position 21 shown in Figure 5. The unit is μm.

In Table 1A, column x1 (hereinafter referred to as "x1") indicates the boundary x1 of the first row in Figure 5 (hereinafter referred to as "row 1"), and column x2 (hereinafter referred to as "x2") indicates the boundary x9 of the second row in Figure 5 (hereinafter referred to as "row 2"). Furthermore, column x3 (hereinafter referred to as "x3") indicates the boundary x17 of the third row in Figure 5 (hereinafter referred to as "row 3"), and column x4 (hereinafter referred to as "x4") indicates the boundary x25 of the fourth row in Figure 5 (hereinafter referred to as "row 4"). The same applies to column y in Table 1A. The above representation also applies to Tables 1B and 1C.

The -14.2 μm in x1 and y15 shown in Table 1A is the minimum offset (1DZRMIN) among the offsets in the boundary x1 of the first row shown in Figure 5.

Next, the nozzle offset values for Implementation 1, based on row 1, are specifically described using Table 1B.

Table 1B shows a list of the baseline nozzle offset values for row 1. The baseline nozzle offset values for row 1 are the differences between the in-plane scanning direction offset values in row 1 of Table 1A and the in-plane scanning direction offset values for rows 2 through 4 of Table 1A. The unit is μm.

For example, the -6.8 μm in x2 and y2 shown in Table 1B is the value obtained by subtracting the 4.1 μm in x1 and y2 shown in Table 1A from the -2.7 μm in x2 and y2 shown in Table 1A.

Furthermore, the -8.0 μm in x4 and y17 shown in Table 1B is the minimum difference (2DZRMIN) among all the values shown in Table 1B.

Next, we will use Table 1C to provide a detailed explanation of the two-dimensional offset in Implementation 1.

Table 1C shows a list of two-dimensional offsets (hereinafter referred to as the "two-dimensional offset table"). The two-dimensional offsets are normalized to positive values by subtracting the minimum two-dimensional offset difference (2DZRMIN) shown in Table 1B from each of the differences (the baseline nozzle offsets in the first row) shown in Table 1B. The unit is μm.

For example, the 1.2 μm in x2 and y2 shown in Table 1C is the value obtained by subtracting the two-dimensional offset minimum difference (2DZRMIN), which is -8.0 μm, from the -6.8 μm in x2 and y2 shown in Table 1B.

Next, we will use Table 1D to provide a detailed explanation of the one-dimensional offset in Implementation 1.

Table 1D shows a list of one-dimensional offsets (hereinafter referred to as the "one-dimensional offset table"). The one-dimensional offsets are normalized to positive values by subtracting the minimum one-dimensional offset difference (1DZRMIN) from each of the in-plane scanning direction offsets in the first row of Table 1A. The unit is μm.

For example, the 18.3 μm in y2 shown in Table 1D is the value obtained by subtracting the one-dimensional offset minimum amount (1DZRMIN), which is -14.2 μm, from the 4.1 μm in x1 and y2 shown in Table 1A.

Next, the ejection start position correction amount for Implementation 1 is specifically described using Table 1E.

Table 1E shows the ejection start position correction amount. The ejection start position correction amount is the sum of the one-dimensional offset minimum value (1DZRMIN) and the two-dimensional offset minimum difference value (2DZRMIN). The unit is μm.

For example, the -22.2 μm shown in Table 1E is the sum of the one-dimensional offset minimum deviation (1DZRMIN), which is -14.2 μm, shown in Table 1A, and the two-dimensional offset minimum difference (2DZRMIN), which is -8.0 μm, shown in Table 1B.

The reason for setting the values in the one-dimensional offset table and the two-dimensional offset table to positive values is to set the offset values that can be used by the variable delay device 12 of each nozzle (details will be described later).

<Procedure>

The following describes the printing process of the inkjet printing device 100.

Specifically, the printing operation procedure of the inkjet printing device 100 is divided into (1) data saving step and (2) ejection step for explanation.

(1) Data saving steps

First, save the various data mentioned above.

Specifically, the original image data is stored in the print original image holder 10. The normalized one-dimensional offset information holder 8 stores the normalized one-dimensional offset pixel data. The normalized two-dimensional offset information holder 9 stores the normalized two-dimensional offset pixel data.

Below, we will explain an example of original image data using Figure 6. Figure 6 shows an example of original image data.

As shown in Figure 6, the original image data is divided into a grid pattern using a print resolution of px in the scanning direction and a print resolution of py in the nozzle row direction, with 20 nozzles and 31 scans. This grid pattern corresponds to the print dot grid 2a shown in Figure 3.

As shown in Figure 6, multiple pixel regions 1c are defined in the original image data. Each pixel region 1c is identical to the pixel region 1c shown in Figure 3. The black dots shown within the droplet region 1e within each pixel represent the dot grid 2b within the droplet region, indicating the eight drops of ink ejected, similar to Figure 3.

Next, we will use Table 2A to illustrate an example of two-dimensional offset pixel data.

[Table 2A]

Table 2A shows an example of two-dimensional offset pixel data. The two-dimensional offset pixel data shown in Table 2A is an example of a two-dimensional offset correction value.

The 2D offset pixel count data is converted to pixel count data by dividing each value in the 2D offset table shown in Table 1C by the scanning direction resolution px (e.g., 5.0μm).

For example, the value of 0.24 pixels at x2 and y2 shown in Table 2A is the value obtained by dividing 1.2 μm at x2 and y2 shown in Table 1C by 5.0 μm. Furthermore, the value of 3.44 pixels at x4 and y1 shown in Table 2A is the maximum value (2DZRGSMAX) among all the values shown in Table 2A.

Next, we will use Table 2B to illustrate an example of one-dimensional offset pixel data.

Table 2B shows an example of one-dimensional offset pixel data. The one-dimensional offset pixel data shown in Table 2B is an example of a one-dimensional offset correction value.

The one-dimensional offset pixel count data is converted to pixel count data by dividing each value in the one-dimensional offset table shown in Table 1D by the scanning direction resolution px (e.g., 5.0μm).

For example, the 3.66 pixels in y2 shown in Table 2B is the value obtained by dividing the 18.3 μm in y2 shown in Table 1D by 5.0 μm. Furthermore, the 4.8 pixels in y12 shown in Table 2B is the maximum value (1DZRGSMAX) among all the values shown in Table 2B.

Next, we will use Table 2C to explain an example of correcting the number of pixels at the end of ejection.

Table 2C shows an example of the number of pixels used to correct the ejection end position. The 8.24 pixels shown in Table 2C are the sum of the maximum value (2DZRGSMAX) of 3.44 pixels shown in Table 2A and the maximum value (1DZRGSMAX) of 4.8 pixels shown in Table 2B.

Here, the sum of the number of pixels used to correct the output end position and the number of scans of the original image data is stored as output count information in the output timing generator 6. For example, the sum of 2DZRGSMAX (3.44 pixels) and 1DZRGSMAX (4.8 pixels) is 8.24 pixels, and the number of scans of the original image data is 31. Therefore, the total sum is 39.24. The resulting total sum is rounded to an integer and stored in the output timing generator 6 as output count information "40 pixels."

Furthermore, segmentation interval pixels BKGSW are obtained. This is the output interval of normalized two-dimensional offset pixel data for each segmentation pixel, converted from the interval (μm) of each segmentation in the scanning direction. The obtained segmentation interval pixels BKGSW are stored as segmentation interval information in the output timing generator 6. For example, as described above, since segmentation interval BKW = 40.0 μm and scanning direction print resolution px = 5.0 μm, segmentation interval pixels BKGSW = 8.

In addition, the ejection start position correction amount (pixels) generated by the ejection timing signal can be obtained. The ejection timing signal is obtained by converting the sum (μm) of the one-dimensional offset minimum difference (1DZRMIN) and the two-dimensional offset minimum difference (2DZRMIN) into minimum difference pixels.

The ejection timing signal starts when the beginning of the original image reaches the nozzle position. Therefore, the ejection start position correction value of the ejection timing signal is used to correct the ejection start position. Since the minimum one-dimensional offset (1DZRMIN) is -14.2μm, the minimum two-dimensional offset difference (2DZRMIN) is -8.0μm, and px = 5.0μm, the ejection start position correction value (pixels) is -4.44 pixels (see Table 2D).

The following summarizes the terms used regarding offsets in Implementation 1.

Drop offset (μm) is the offset between the expected drop position 20 and the actual drop position 21 (for example, refer to Table 1A).

1D Drop Deflection (μm) is the drop deflection value for row 1 (e.g., row x=1 in Table 1A).

The two-dimensional drop offset (μm) is the difference between the one-dimensional drop offset and the drop offset for each row (e.g., rows with x=2 to 4 in Table 1A) (see Table 1B, for example).

Normalized 2D offset (μm) is the offset obtained by normalizing the 2D drop offset to a positive value (e.g., see Table 1C).

Normalized 1D offset (μm) is the offset obtained by normalizing the 1D drop offset to a positive value (e.g., see Table 1D).

Normalized 2D offset (pixel) is data obtained by normalizing the normalized 2D offset using the scanning direction printing resolution px (for example, see Table 2A).

Normalized one-dimensional offset (pixel) is data obtained by normalizing the normalized one-dimensional offset using the scanning direction printing resolution px (for example, refer to Table 2B).

The output count information (times) indicates the number of times required to output the entire original image after offset correction.

The segmentation interval pixel (pixel) is the interval for extracting normalized two-dimensional offset pixel data.

The output end position correction amount (pixels) is the correction amount (positive value) by which the output end position is offset in order to output the total offset-corrected portion of the original image by adding the absolute amount of the negative offset portion of the offset-corrected original image to the positive offset portion (see Table 2C).

The output start position correction amount (pixels) is the amount (negative value) by which the output start position is offset to compensate for the negative offset of the offset-corrected original image (see Table 2D).

(2) Discharging step

Next, printing is performed by ejecting ink droplets.

Below, we first explain the overall printing process, and then provide a detailed explanation of the correction process.

Here, the ejection timing generator 6 is configured to generate an ejection timing signal at the ejection start position corrected by the ejection start position correction amount (pixels) = -4.44 pixels, with a scanning direction print resolution of px and ejection count information (times) = 40. Furthermore, the ejection timing generator 6 is configured to generate a split timing signal each time the ejection timing signal is generated with a split interval of 8 pixels (pixels).

The following is a detailed explanation of the overall printing process.

First, when the movable platform 4, which carries the printed circuit board 1, moves relative to the inkjet head 3, position information is generated in chronological order.

The generated position information is input to the position detector 5. The position detector 5 generates a position information pulse signal by organizing the position information into a pulse signal, and outputs it to the output timing generator 6.

The ejection timing generator 6 divides the position information pulse signal from the position detector 5 according to the preset scanning direction printing resolution px to generate an ejection timing signal, which is then output to the drive signal generator 7. The ejection timing signal specifies the timing of the voltage waveform that drives the nozzles 3a of the inkjet head 3, as described above. The ejection timing signal is also output to the print original image holder 10, the nozzle offset information adder 11, and the nozzle variable delay device 12. Furthermore, the divided timing signal is output to the normalized two-dimensional offset information holder 9.

The drive signal generator 7 generates a drive waveform signal based on the ejection timing signal from the ejection timing generator 6 and outputs it to the drive signal selector 13. The drive waveform signal is a signal used to eject ink droplets from the nozzles 3a of the inkjet head 3 as described above.

Meanwhile, the print original image holder 10 synchronizes the logical print bitmap data, or original image data, with the ejection timing signal and outputs it to each nozzle's variable delay device 12. Furthermore, the original image data is used to instruct the ejection of ink droplets relative to the print dot grid 2a defined on the print substrate 1, as described above.

Furthermore, the normalized one-dimensional offset information holder 8 synchronizes the normalized one-dimensional offset data (pixel) of each nozzle 3a with the ejection time point signal and outputs it to each nozzle offset information adder 11.

Furthermore, the normalized two-dimensional offset information holder 9 synchronizes the normalized two-dimensional offset data (pixel) of each nozzle 3a with the division time point signal and outputs it to the nozzle offset information adder 11.

Each nozzle offset information adder 11 adds the following data at each ejection time point: the normalized one-dimensional offset data (pixel) for each nozzle 3a from the normalized one-dimensional offset information holder 8, and the normalized two-dimensional offset data (pixel) for each nozzle 3a from the normalized two-dimensional offset information holder 9. Furthermore, each nozzle offset information adder 11 outputs the added result, i.e., the offset information indicating the offset (delay) for each nozzle, to each nozzle variable delay device 12.

Each nozzle variable delay device 12 generates ejection on/off instruction information based on the original image data from the print original image holder 10 and the offset information from the nozzle offset information adder 11, and outputs it to the drive signal selector 13. As described above, the ejection on/off instruction information indicates whether to turn on/off (activate/inactivate) the ejection of ink droplets for each print dot grid 2a for each nozzle 3a.

The drive signal selector 13 turns the drive waveform signal from the drive signal generator 7 on and off for each nozzle 3a based on the ejection on/off instruction information from the variable delay device 12. This controls the ejection of ink droplets from each nozzle 3a.

The above-described operation is repeated while the movable platform 4 passes beneath the inkjet head 3 at a predetermined speed. This allows ink droplets to be ejected onto the printed circuit board 1 according to the original image data. Once the movable platform 4 has passed beneath the inkjet head 3, the ejection of ink droplets according to the original image data is complete.

As described above, the inkjet printing device 100 can now perform printing operations.

Next, the characteristic feature of Implementation 1, namely the correction action, will be explained using Figure 7.

In this correction operation, the original image data and drop offset information are pre-set to each holder shown in Figure 4. This corrects the drop offset of each nozzle 3a and the drop offset within the printed circuit board 1 caused by the deflection of the moving stage 4.

FIG7 is a diagram showing the connections of the main components that perform the correction operation among the components of the inkjet printing device 100 shown in FIG4 .

As shown in Figure 7, the normalized one-dimensional offset information holder 8, the normalized two-dimensional offset information holder 9, the original print image holder 10, the nozzle offset information adder 11, and the nozzle variable delay device 12 are connected for each nozzle 3a. In Figure 7, #1 to #20 indicate the identification numbers of the 20 nozzles 3a.

As shown in FIG7 , each nozzle variable delay device 12 has 20 variable delay devices (from nozzle 1 variable delay device to nozzle 20 variable delay device).

Next, as an example of the various nozzle variable delay devices 12 described above, the internal structure of the variable delay device for nozzle 2 will be described using Figure 8.

FIG8 is a diagram showing an example of the internal structure of the variable retarder for nozzle 2 among the variable retarders 12 for each nozzle shown in FIG4 and FIG7.

As shown in Figure 8, the variable delay device of nozzle 2 is composed of pixel holders 30a to 30h, logic multipliers 31a to 31h, partial negative logic multipliers 32a to 32h, logic adders 33a to 33h, and a decoder 34.

Pixel holders 30a through 30h are composed of, for example, 1-bit flip-flops. Each pixel holder holds the D input value at Q when the CK clock (CLK) signal is activated. Furthermore, Q is cleared to 0 (zero) by the /*CLR signal.

Logical multipliers 31a to 31h are composed of, for example, AND gates. Each logical multiplier outputs a logical AND of two inputs.

The partial negative logic multipliers 32a to 32h are composed of, for example, 1-input negative logic AND gates. Each partial negative logic multiplier outputs the logical AND of a positive logic 1 input and a negative logic 1 input.

Logical adders 33a to 33h are composed of, for example, OR gates. Each logical adder outputs the logical sum (OR) of its two inputs.

Decoder 34 outputs positive logic bits to S1 through S8 according to the input value of 1 to 8. For example, when the input value to decoder 34 is 1, S1 becomes 1, and S2 through S8 become 0. When the input value is 2, S1 and S2 become 1, respectively, and S3 through S8 become 0. When the input value is 3, S1 through S3 become 1, respectively, and S4 through S8 become 0. When the input value is 4, S1 through S4 become 1, respectively, and S5 through S8 become 0. When the input value is 5, S1 through S5 become 1, respectively, and S6 through S8 become 0. When the input value is 6, S1 through S6 become 1, respectively, and S7 and S8 become 0. When the input value is 7, S1 through S7 become 1, respectively, and S8 becomes 0. When the input value = 8, all S1 to S8 become 1.

Next, the signal status inside the variable delay device of nozzle 2 will be explained using Figure 9.

FIG9 is a diagram showing the internal signal status of the variable delay device of the nozzle 2 shown in FIG8 along with the passage of time.

Below, we will use nozzle #2 (hereinafter referred to as "the second nozzle") as an example to explain the correction operation during printing.

As shown in Figure 9, the normalized one-dimensional offset information holder 8 shown in Figure 7 holds 3.66 (unit: pixel) as the one-dimensional offset pixel data for the second nozzle (1Dzure#2). The normalized two-dimensional offset information holder 9 shown in Figure 7 holds 1.6, 0.24, 1.04, and 2.44 (unit: pixel) as the two-dimensional offset pixel data for the second nozzle (2Dzure#2).

Furthermore, the print original image holder 10 shown in FIG7 holds rows of values indicating whether the print dot grid 2a is an object to be ejected (black dots) or an object not to be ejected (white circles) in the order of x1 to x31 relative to y2 in FIG6 , serving as the original image data (Din#2) for the second nozzle. For example, as shown in FIG9 , the original image data (Din#2) for the second nozzle is 0 (non-ejection target), 0, 0, 1 (ejection target), 1, 0, 0, 0, 0, ... 1, 1, 0, 0.

The one-dimensional offset pixel quantity data (1Dzure#2) for the second nozzle, 3.66, is synchronized with the ejection timing signal and output to the nozzle offset information adder 11 shown in Figure 7.

The two-dimensional offset pixel data for the second nozzle (2Dzure#2), namely 1.6, 0.24, 1.04, and 2.44, are synchronized with the split time point signal and sequentially output to each nozzle offset information adder 11 shown in Figure 7.

Each nozzle offset information adder 11 arithmetically adds the one-dimensional offset pixel data (1Dzure#2) for the second nozzle and the two-dimensional offset pixel data (2Dzure#2) for the second nozzle. As shown in Figure 8 , this added value (in other words, the delay amount) is output as the selection data SEL#2 signal to the nozzle 2 variable delay device of each nozzle variable delay device 12.

Meanwhile, the original image data from the second nozzle is output to the variable delay device of nozzle 2 of each nozzle variable delay device 12 in synchronization with the ejection timing signal in the aforementioned sequence. Specifically, as shown in Figure 8, the original image data from the second nozzle is output to the variable delay device of nozzle 2 as the selected data Din#2 signal.

Figure 9 shows the values of selection data SEL#2 and selection data Din#2 corresponding to each ejection time point t=1 to 40.

Here, using Figures 8 and 9, we illustrate the following operation: Selected data Din#2 is delayed according to selected data SEL#2 and output as Dout#2.

First, as shown in Figure 8, the *CLR signal is output from the ejection timing generator 6 to the variable delay device of the nozzle 2. This clears the Q outputs of pixel holders 30a to 30h to 0 (zero).

At this time, as shown in Figure 9, at time t=1, each nozzle offset information adder 11 calculates 3.66(1Dzure#2 (one-dimensional offset pixel data)) + 1.6(2Dzure#2 (two-dimensional offset pixel data for the first row)) = 5.26 pixels (SEL#2.float). Furthermore, each nozzle offset information adder 11 discards the five decimal points from the calculated 5.26 pixels as the selected data SEL#2 signal and outputs it to the decoder 34 of the variable delay unit of nozzle 2.

Next, the decoder 34 shown in FIG8 outputs 1 to S1 to S5 and 0 to S6 to S8, respectively, based on the input selection data SEL#2 signal.

Furthermore, the logic multipliers 31a-31e, partially negative logic multipliers 32a-32e, and logic adders 33a-33e connected to S1-S5 of the decoder 34 transmit the Q outputs of the pixel holders 30a-30e to Dout#2 or the D inputs of the pixel holders 30a-30d.

On the other hand, the logical multipliers 31f-31h, partially negative logical multipliers 32f-32h, and logical adders 33f-33h connected to S6-S8 of decoder 34 transmit the Din#2 input to the D inputs of pixel holders 30e-30g.

Pixel holder 30h has Din#2 input to its D input. Furthermore, by activating the CLK signal from output timing generator 6, the D input values of pixel holders 30a-30h are held and output to Q.

At this time, as shown in Figure 9, since Din#2 = 0 at times t = 1 to 3, Q of pixel holders 30a to 30h remains at 0. Furthermore, at times t = 4 and 5, after Din#2 = 1, Din#2 is transmitted to D5 of pixel holder 30e. Furthermore, at times t = 6 to 8, Din#2 is transmitted from D5 of pixel holder 30e to pixel holders 30a to 30d in sequence. This results in a delay of five pixels.

Then, as shown in Figure 9, at t=9, when the SEL#2 signal changes from 5 to 3, the output destination of Din#2 switches to D4-D3. As a result, the Q values previously held by pixel holders 30d-30e at t=7-8 are discarded, and starting at t=9, the delay increases from 5 pixels to 3 pixels.

Similarly, at t=17, when the SEL#2 signal changes from 3 to 1, the delay increases from 3 pixels to 1 pixel. Furthermore, at t=25, when the SEL#2 signal changes from 1 to 6, the delay increases from 1 pixel to 6 pixels. Finally, at t=40, the delay is complete.

As a result, as shown in Figure 9, Dout#2 outputs 1 at each of times t=9, 10, t=15, 16, t=24, 25, t=34, and 35.

As shown above, the value shown by Din#2 is delayed according to the increase or decrease in the delay amount shown by the SEL#2 signal, and is output as Dout#2.

Alternatively, if a Din pixel is an output pixel at time t, when the two-dimensional offset is reduced, the SEL signal may not be switched until time t+n, when the pixel becomes a non-output pixel. This prevents the output pixel from being deleted.

Furthermore, when the pixel division interval is a fractional pixel value, accumulation can be performed using a fractional adder. Furthermore, the two-dimensional offset pixel data can be switched when the accumulated pixel division interval crosses an integer interval value. This eliminates the accumulation error that would otherwise occur if the decimal division position were converted to an integer, and reduces it to a fractional error. Consequently, accumulation error can be effectively suppressed.

As described above, the inkjet printing device 100 of embodiment 1 stores original image data in the print original image holder 10. If the ejection position changes due to a change in the characteristics of the nozzles 3a of the inkjet head 3, the one-dimensional offset pixel volume data representing the characteristics is stored in the normalized one-dimensional offset information holder 8. Furthermore, if the movement characteristics of the moving platform 4 change, the two-dimensional offset pixel volume data representing the characteristics is stored in the normalized two-dimensional offset information holder 9. The original image data is then printed based on the stored one-dimensional offset pixel volume data and two-dimensional offset pixel volume data.

Specifically, the inkjet printing device 100 independently stores the original image data and the two-dimensional offset pixel data. Furthermore, the original image data is printed based on the sum of the stored one-dimensional offset pixel data and the two-dimensional offset pixel data. Consequently, when performing one-dimensional offset correction, only the one-dimensional offset pixel data needs to be overwritten. Similarly, when performing two-dimensional offset correction, only the two-dimensional offset pixel data needs to be overwritten. Consequently, there is no need to reflect the two offset pixel data on the original image data or to write the original image data to memory.

Furthermore, there is no need to replace the original image data in the printing original image holder 10. Therefore, the processing time required for printing preparation can be significantly reduced.

As described above, the inkjet printing apparatus 100 of embodiment 1 can correct for deviations in the ink droplet landing position while reducing processing time. Consequently, a high-quality display or electronic device incorporating the same can be provided. The high-quality display can meet the demands of larger display panels, increased chamfers, and the shortened printing process required by increased production volume, while also minimizing color mixing and brightness unevenness.

(Implementation Form 2)

The following describes the inkjet printing device 100 according to Embodiment 2.

The inkjet printing device 100 of Embodiment 2 differs from Embodiment 1 in that, among the components shown in FIG4 , it does not include the normalized one-dimensional offset information holder 8 and the nozzle offset information adder 11.

Specifically, the inkjet printing device 100 of Embodiment 2 uses software processing to add the one-dimensional offset pixel data and the two-dimensional offset pixel data for each nozzle 3a. Furthermore, instead of storing the two-dimensional offset pixel data, the resulting addition value is stored in the normalized two-dimensional offset information holder 9. This enables the same operation as Embodiment 1.

In this case, the added value can also be stored in the normalized two-dimensional offset information holder 9 as integer data with the decimal point truncated. This reduces memory and circuit size compared to embodiment 1 in which the added value is decimal data.

(Implementation Form 3)

The following describes the inkjet printing device according to Embodiment 3 using FIG10 .

Figure 10 is a diagram showing an example of the tilted state of the inkjet head in embodiment 3.

As shown in Figure 10, the inkjet printing apparatus of Embodiment 3 differs from Embodiment 1 in that the two inkjet heads 3 shown in Figure 4 are arranged obliquely with respect to the scanning direction X, and the nozzle spacing Lny shown in Figure 4 is reduced. This results in a smaller nozzle spacing LnX in the scanning direction X (hereinafter referred to as "scanning direction nozzle spacing LnX").

Specifically, in Embodiment 3, a predetermined control computer (software) is used to stagger the nozzles 3a in the original image data by n times the nozzle pitch LnX in the scanning direction (hereinafter referred to as "staggering"). This allows for printing with offset correction.

The offset of each nozzle 3a is recalculated whenever the print image resolution (px) in the scanning direction changes. The original image data is then offset based on the recalculated offset.

Furthermore, n is determined by Lnx and the division value each time the original image data's scanning-direction print resolution (px) changes. Although a decimal point may appear in this case, it is added to the one-dimensional offset pixel data in the normalized one-dimensional offset information holder 8. This effectively reduces cumulative error.

Furthermore, the above-mentioned staggered processing refers to the processing of the original image data itself in accordance with the configuration of the nozzle 3a.

Generally, the placement of the nozzles 3a is determined during design. However, ink droplet placement can shift depending on the installation of the inkjet head 3 or the inkjet head's own ejection behavior (machine variability). Therefore, during offset processing, this shift is corrected based on one-dimensional offset pixel data. Meanwhile, the offset is processed based on the original image data itself, tailored to the nozzle 3a placement. In this case, the frequency of original image production is lower than the frequency of processing the offset correction. Therefore, even with large original image data, processing time does not need to be squeezed into production. This prevents a reduction in productivity.

(Implementation Form 4)

The following describes the inkjet printing device according to Embodiment 4, with reference to Figure 4 and Figure 10.

Implementation 4 differs from Implementation 3 in that it performs offset correction by n times the nozzle pitch LnX in the scanning direction, as described in Implementation 3, without processing the original image data.

In addition to the components shown in FIG4 , the inkjet printing device 100 of embodiment 4 is further equipped with a nozzle arrangement offset information holder (not shown) that stores the offset amount of each nozzle 3a.

The nozzle configuration offset information holder stores the recalculated offset of each nozzle 3a each time the print image resolution px in the scanning direction changes.

Furthermore, each nozzle offset information adder 11 expands the variable delay range of each nozzle variable delay device 12 to a range capable of performing offset correction by an offset of n times the nozzle pitch LnX in the scanning direction, based on the offset amount of each nozzle 3a stored in the nozzle configuration offset information holder.

With the above, in Embodiment 4, it is possible to print with offset correction without processing the original image data.

In addition, the present disclosure is not limited to the above-described embodiments, and various modifications are possible without departing from the spirit and scope of the present disclosure.

1: Printed substrate

1c: Pixel area

3: Inkjet Head

3a: Nozzle

4: Mobile Platform

5: Position detector

6: Output time point generator

7: Drive signal generator

8: Normalized one-dimensional offset information preserver

9: Normalized 2D offset information preserver

10: Printing original image holder

11: Adder for offset information of each nozzle

12: Variable delay device for each nozzle

13: Drive signal selector

100: Inkjet printing device

Lny: interval

X: Scan direction

Y: Nozzle direction

#1, #2, #20: Identification numbers

Claims (6)

An inkjet printing device comprises a movable platform and an inkjet head. The movable platform moves a printing substrate, and the inkjet head ejects ink droplets from a plurality of nozzles relative to the printing substrate. The inkjet printing device further comprises: A printing original image holder, which holds original image data indicating the locations of the ink droplets landing on the printing substrate; A normalized two-dimensional offset information holder, which holds either a two-dimensional offset correction value or an added value. The two-dimensional offset correction value is used to correct the two-dimensional offset of the ink droplet landing position for each nozzle. The added value is the sum of the two-dimensional offset correction value and a one-dimensional offset correction value at the time of ejection of each ink droplet. The one-dimensional offset correction value is used to correct the one-dimensional offset of the ink droplet landing position. Each nozzle variable delay device generates ejection on/off instruction information for varying the ejection timing based on the original image data and the added value. A drive signal selector turns ejection of the ink droplets on/off for each nozzle based on the ejection on/off instruction information. The one-dimensional offset is the sum of the offset in the nozzle configuration and the angular offset of the ejected ink droplets relative to the vertical direction. The two-dimensional offset is the sum of the yaw and vertical yaw generated during scanning of the printed circuit board. The inkjet printing device of claim 1 further comprises: a normalized one-dimensional offset information holder for holding the one-dimensional offset correction; and each nozzle offset information adder for calculating the added value based on the one-dimensional offset correction held by the one-dimensional offset information holder and the two-dimensional offset correction held by the two-dimensional offset information holder. An inkjet printing device as claimed in claim 1 or 2, wherein the aforementioned added value maintained by the aforementioned two-dimensional offset information holder is an integer. In the inkjet printing device of claim 1 or 2, the inkjet head is tilted relative to a scanning direction, and when a nozzle spacing has been generated in the scanning direction, the nozzle spacing is corrected according to the misalignment of each nozzle. As in claim 3, the inkjet printing device, wherein the inkjet head is tilted relative to a scanning direction, and when a nozzle spacing has been generated in the scanning direction, the nozzle spacing is corrected according to the misalignment of each nozzle. An inkjet printing method is performed using a moving stage and an inkjet head. The moving stage moves a printed substrate, and the inkjet head ejects ink droplets from a plurality of nozzles relative to the printed substrate. The inkjet printing method first stores original image data and either a two-dimensional offset correction value or an added value in separate memories. The original image data indicates the droplet locations of the ink droplets on the printed substrate. The two-dimensional offset correction value is used to correct the two-dimensional offset of the droplet locations of each ink droplet from the nozzle. The added value is obtained by adding a one-dimensional offset correction value to the two-dimensional offset correction value at the time of ejection of each ink droplet. The one-dimensional offset correction value is used to correct the one-dimensional offset of the droplet locations. Based on the original image data and the added value, ejection on/off instruction information is generated to change the ejection timing. Based on the ejection on/off instruction information, the ejection of the ink droplets is turned on/off for each nozzle. The one-dimensional offset is the sum of the offset in the nozzle configuration and the angular offset of the ejected ink droplets relative to the vertical direction. The two-dimensional offset is the sum of the yaw and yaw generated during scanning of the printed circuit board.