An inkjet type glue coating method based on topography compensation

Abstract

The present application provides a kind of based on topographic compensation's ink-jet type gluing method, the method obtains standard glue amount distribution data, and by measuring the actual three-dimensional topography of workpiece and comparing with standard topography, identify the surface recessed area, then only for the recessed area increase set glue amount, keep standard glue amount unchanged outside recessed area, generate compensated glue amount distribution data, then control print head jet glue drop array according to this, spread fusion forms glue film and solidifies.Existing fixed glue amount ink-jet mode assumes that workpiece is consistent with standard topography, when there is recess on the surface of workpiece, actual spray gap becomes larger and glue amount is unchanged, resulting in that the glue film after solidification is thin or even not fully covered.The present application compensates the glue amount loss of recessed area by targeting, while keeping the glue amount of other areas unchanged, avoids the risk of glue break caused by reducing glue, and uses the spreading and leveling property of glue itself to automatically equalize the glue amount difference, effectively ensures the uniformity of glue film thickness, and suppresses the defect of incomplete coverage.

Description

Technical Field

[0001] This invention relates to the field of automated fluid coating equipment technology, and in particular to an inkjet coating method based on morphology compensation. Background Technology

[0002] In precision manufacturing fields such as electronic packaging, optical bonding, and semiconductor mounting, array-type inkjet printheads based on fixed grayscale images are often used for parallel adhesive application to form a uniform adhesive layer on the workpiece surface. This method maps a standard CAD model into a grayscale image and quantitatively allocates the amount of adhesive sprayed by controlling the grayscale value of each pixel.

[0003] However, due to manufacturing tolerances and clamping errors, the surface of actual workpieces often has localized undulations. When localized depressions appear on the workpiece surface, the actual jetting gap between the printhead nozzle and the workpiece surface at that location will be larger than the standard working gap. In fixed grayscale jetting mode, the increased gap results in insufficient spreading kinetic energy when the adhesive droplets reach the depressed surface, leading to scattered droplet placement. Ultimately, this results in a significantly lower cured adhesive film thickness in that area than the target value, or even incomplete coverage, severely compromising the overall uniformity of the adhesive layer. Current technology lacks a solution that can accurately compensate for the adhesive loss caused by these localized depressions while maintaining the high efficiency of array inkjet printing. Summary of the Invention

[0004] The embodiments of the present invention provide an inkjet coating method based on topography compensation, which solves the technical problems of thin adhesive film and incomplete coverage caused by the inability to compensate for local depressions on the workpiece surface in existing array inkjet coating technology.

[0005] To address the aforementioned problems, embodiments of the present invention provide an inkjet coating method based on morphology compensation, comprising: acquiring standard adhesive distribution data corresponding to the morphology of a standard workpiece; measuring the actual three-dimensional morphology data of the workpiece to be coated, and comparing the actual three-dimensional morphology data with the morphology of the standard workpiece to determine a recessed area on the surface of the workpiece to be coated that causes an increase in the spray gap relative to the standard state; performing unidirectional compensation on the standard adhesive distribution data corresponding to the recessed area to increase the set adhesive amount in that area, while keeping the standard adhesive distribution data unchanged for areas outside the recessed area, generating compensated adhesive distribution data; controlling the inkjet printhead to spray an array of adhesive droplets onto the surface of the workpiece to be coated according to the compensated adhesive distribution data; and causing the array of adhesive droplets to spread and fuse to form a continuous adhesive film and then cure.

[0006] Furthermore, both the standard adhesive distribution data and the compensated adhesive distribution data are in grayscale image format. The unidirectional compensation is achieved by increasing the grayscale value of the corresponding pixel in the recessed area. The standard adhesive distribution data adopts the following pixel-to-nozzle mapping method: in the direction perpendicular to the scanning direction of the inkjet printhead, one pixel column corresponds to one nozzle; in the direction parallel to the scanning direction, one nozzle corresponds to multiple pixel rows during the scanning process.

[0007] Furthermore, the concave region where the spray gap increases is determined by calculating the height deviation field ΔZ. The height deviation field ΔZ is the height difference between the actual three-dimensional shape data and the standard workpiece shape at the same position along the vertical Z-axis of the equipment; ΔZ(x,y)=Z_act(x,y)-Z_ref(x,y), where (x,y) represents the pixel coordinate position, Z_act(x,y) is the Z-axis height value of the actual workpiece at position (x,y), Z_ref(x,y) is the Z-axis height value of the standard workpiece at the corresponding position, and ΔZ<0 indicates that the actual workpiece is too low and the spray gap is greater than the standard working gap d0.

[0008] Furthermore, the actual three-dimensional topography data is obtained by measuring with a 3D line laser profile scanner, and the measured actual three-dimensional point cloud is rigidly registered with the reference point cloud generated by the standard CAD model through an iterative nearest point algorithm to unify them into the same coordinate system.

[0009] Furthermore, before performing the measurement, a precision grid calibration plate is used to calibrate the coordinate system of the measuring equipment and the inkjet printhead to obtain the transformation matrix between the two coordinate systems.

[0010] Furthermore, the compensation amount of the unidirectional compensation is determined based on a pre-calibrated correlation model between the amount of adhesive, the spray gap, and the film thickness. The model is h=F(G,d), where d is the spray gap, G is the gray value, and h is the film thickness that can be achieved after spreading and fusion. For the region where ΔZ<0, the actual gap d_act=d0+|ΔZ|, and the compensation gray value is calculated to satisfy h_target=F(G_comp,d0+|ΔZ|).

[0011] Furthermore, the compensation grayscale is calculated within the linear interval of the model according to the following formula: G_comp(x,y)=G0(x,y)+k·|ΔZ(x,y)|, where G0 is the standard grayscale value in the standard glue quantity distribution data, and k is a positive compensation coefficient obtained through experimental calibration; the calibration method of the compensation coefficient k is as follows: a series of test blocks of different heights are made, the actual glue output at each height is measured, and the k value is obtained by linear regression fitting.

[0012] Furthermore, the model h=F(G,d) is calibrated as follows: a standard test substrate is prepared, and the type of adhesive, ambient temperature and humidity, resting time and curing parameters are fixed; multiple different spray gaps are set as gap gradients, and multiple different gray values ​​are set as gray gradients under each gap; an array of adhesive dots is sprayed under each calibrated gradient combination, and after static diffusion and UV curing, the actual adhesive film thickness is measured; the measurement data is fitted into a curve function.

[0013] Furthermore, when the calculated compensation grayscale exceeds the maximum grayscale limit of the inkjet printhead, the compensation grayscale is clamped to the maximum grayscale limit value, and the relative movement speed between the inkjet printhead and the workpiece is kept constant; the film thickness is based on the final dry film thickness after UV curing and shaping, and the calculation of the height deviation field ΔZ and the calculation of the compensation grayscale are both based on the final dry film thickness as the target value.

[0014] Furthermore, the compensated adhesive distribution data is halftone processed and parsed to generate a waveform signal that drives the piezoelectric element in the inkjet printhead, so as to control the inkjet printhead to spray the adhesive droplet array at a constant height d0, forming an array of adhesive dots with spatially differentiated adhesive distribution.

[0015] Furthermore, the spacing P between adjacent adhesive droplets in the droplet array satisfies: P≤√2·R_spread, where R_spread is the diffusion radius of a single adhesive droplet after it has spread and stabilized on the surface of the workpiece to be coated; the diffusion radius R_spread is measured in the following way: isolated adhesive dots are sprayed on a standard test substrate, and after being left to stand under the same environmental conditions as the actual process until they stabilize, the diameter D of the adhesive dots is measured, and R_spread=D / 2 is calculated; the adhesive droplet array is left to stand for a preset time after spraying to form a continuous wet film, and then cured and shaped by UV irradiation.

[0016] Furthermore, the standard adhesive distribution data corresponds to a standard working gap d0; the unidirectional compensation is only activated when the depth of the recessed area exceeds a preset negative deviation threshold; and a buffer band is provided on the outer edge of the outline of the workpiece to be coated, the amount of adhesive corresponding to the buffer band being lower than the minimum amount of adhesive in the standard adhesive distribution data.

[0017] Furthermore, after the adhesive droplet array spreads, fuses, and solidifies to form a continuous adhesive film, the method further includes: measuring the thickness of the solidified continuous adhesive film to obtain actual adhesive film thickness distribution data; comparing the actual adhesive film thickness distribution data with the target adhesive film thickness to determine whether the coating quality is qualified; when qualified, conveying the workpiece to be coated to the unloading station for unloading; when unqualified, issuing an alarm signal or marking and rejecting the unqualified workpiece.

[0018] Furthermore, when the thickness of a local area in the actual film thickness distribution data is lower than the target film thickness, the position information of the local area is fed back to the inkjet printhead, and the compensation amount of the unidirectional compensation is increased for the corresponding position of the next workpiece in the same batch.

[0019] Furthermore, the thickness detection is performed using the same 3D line spectroscopy measuring instrument or 3D line laser profile scanner used to measure the actual three-dimensional topography data, and the thickness detection is completed online after the continuous adhesive film has been UV cured and shaped, and before the workpiece to be coated leaves the coating station.

[0020] Compared with the prior art, the inkjet coating method based on morphology compensation of the present invention has at least the following beneficial effects: The inkjet coating method based on morphology compensation provided by this invention first acquires standard adhesive distribution data as a reference for subsequent compensation; then, by measuring the actual three-dimensional morphology of the workpiece and comparing it with the standard morphology, surface depression areas are actively identified; subsequently, the set adhesive amount is increased only for the identified depression areas, while the standard adhesive amount remains unchanged for areas outside the depression areas, generating compensated adhesive distribution data; next, the printhead is controlled to spray an array of adhesive droplets based on the compensated data; finally, the adhesive droplets are fused to form a uniform adhesive film and cured by utilizing the spreading and leveling properties of the adhesive itself.

[0021] In existing technologies, fixed-volume inkjet printing assumes the workpiece has a completely identical shape to a standard workpiece. When there are depressions on the workpiece surface, the actual spray gap at those locations is larger than the standard gap, but the sprayed adhesive volume remains unchanged. This results in insufficient adhesive reaching the depressed surface, leading to a thin or even incomplete adhesive film after curing. This invention breaks this chain of problems: the adhesive loss caused by the increased gap in the depressed area is specifically compensated for, while the standard adhesive volume remains constant in areas outside the depressed area, avoiding the risk of adhesive breakage due to reduced adhesive. The adhesive's own spreading and leveling properties further automatically homogenize the adhesive volume differences. Through these methods, the uniformity of the cured adhesive film thickness is effectively guaranteed, and the defect of incomplete coverage is significantly suppressed.

[0022] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0024] Figure 1 A flowchart of an inkjet coating method based on morphology compensation provided in an embodiment of the present invention. Detailed Implementation

[0025] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the specific embodiments, structures, features, and effects according to the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "an embodiment" or "an embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0026] In the description of this invention, it should be clearly stated that the terms "first," "second," etc., in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence; the terms "vertical," "lateral," "longitudinal," "front," "rear," "left," "right," "up," "down," "horizontal," etc., indicate orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, and are merely for the convenience of describing this invention, and do not mean that the device or element referred to must have a specific orientation or position, and therefore should not be construed as a limitation of this invention.

[0027] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0028] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0029] This invention provides an inkjet coating method based on topography compensation, comprising the following steps, such as... Figure 1 As shown: S1: Obtain standard adhesive distribution data corresponding to the morphology of a standard workpiece; that is, firstly, determine the target adhesive film thickness based on the morphology of the standard workpiece of the standard product. Then, map the area to be coated on the surface of the standard workpiece into a two-dimensional distribution map, assigning a corresponding adhesive quantity value to each location in the two-dimensional distribution map. This adhesive quantity value represents the amount of adhesive sprayed at that location to achieve the target adhesive film thickness. The standard adhesive distribution data is the set of adhesive quantity values ​​for all locations. Once generated, this data can be reused in the production of the same model of product.

[0030] S2: Measure the actual three-dimensional morphology data of the workpiece to be coated, and compare the actual three-dimensional morphology data with the morphology of the standard workpiece to determine the recessed areas on the surface of the workpiece that cause the spray gap to increase relative to the standard state. Specifically: Place the workpiece to be coated at the measurement station and obtain the actual three-dimensional morphology data of each position on the workpiece surface. Then, compare the measured actual three-dimensional morphology data with the morphology of the standard workpiece, and calculate the difference between the two for each position. When the actual three-dimensional morphology at a certain position is lower than that of the standard workpiece, it indicates that the position is a recessed area. Since the print head maintains a standard distance from the workpiece surface during operation, when a recess appears on the workpiece surface, the actual distance between the print head and the workpiece surface at that point is greater than the standard distance, i.e., the spray gap increases. By comparison, all recessed areas can be accurately located.

[0031] S3: Perform unidirectional compensation on the standard adhesive distribution data corresponding to the recessed area to increase the set adhesive amount in that area. For areas outside the recessed area, keep the standard adhesive distribution data unchanged, generating compensated adhesive distribution data. In other words, for the adhesive amount value in the standard adhesive distribution data corresponding to the recessed area, perform unidirectional compensation, increasing the adhesive amount in that area to obtain more adhesive than in the standard state. For areas outside the recessed area (i.e., flat or raised areas), keep the adhesive amount value in their standard adhesive distribution data unchanged, without any reduction. After all areas are processed, generate compensated adhesive distribution data specifically for the current workpiece. The core of this step is to only increase the adhesive amount in the recessed area, without reducing the adhesive amount in other areas. In subsequent steps, the adhesive's own spreading and leveling effect can automatically even out any small amount of excess adhesive that may exist in the raised areas.

[0032] S4: Based on the compensated adhesive distribution data, control the inkjet printhead to spray an array of adhesive droplets onto the surface of the workpiece to be coated. Specifically, this step converts the compensated adhesive distribution data generated in S3 into spraying data that the inkjet printhead can recognize, driving the inkjet printhead to spray adhesive droplets onto the workpiece surface at a fixed height. At locations with larger adhesive amounts in the compensated adhesive distribution data (i.e., recessed areas), the sprayed droplets are larger in volume; at locations with smaller adhesive amounts, the sprayed droplets are smaller in volume, thus forming an array of adhesive droplets with differentiated adhesive distribution on the workpiece surface.

[0033] S5: The adhesive droplet array is spread and fused to form a continuous adhesive film and then cured. In this step, the sprayed adhesive droplet array is left to stand on the workpiece surface, and the adhesive spreads slowly on the workpiece surface due to its own physical properties (such as surface tension and leveling). Adjacent adhesive droplets contact and fuse with each other during the spreading process, forming a wet film that continuously covers the entire area to be coated. During the spreading and fusing process, the recessed areas are fully filled due to the increased amount of adhesive, while the amount of adhesive in other areas is slightly homogenized to the periphery through leveling. After the wet film is formed, it is cured and shaped into the final solid adhesive film.

[0034] The method described in this embodiment solves the technical problem of thin adhesive film and incomplete coverage caused by the inability to compensate for local depressions on the workpiece surface in existing array inkjet coating technology. Specifically, the root cause of this technical problem is that the fixed adhesive amount inkjet method assumes that the workpiece and the standard workpiece have completely identical shapes. When there are depressions on the workpiece surface, the actual spray gap at that location is larger than the standard gap, but the sprayed adhesive amount remains unchanged, resulting in insufficient adhesive reaching the depression surface, and the cured adhesive film being thin or even incompletely covered. This embodiment breaks the above-mentioned problem chain through the complete steps of S1 to S5: First, standard adhesive amount distribution data is obtained to provide a benchmark for subsequent compensation; then, the actual three-dimensional shape of the workpiece is measured and compared with the standard shape to actively identify the surface depression area; then, unidirectional compensation is performed only on this area to increase its adhesive amount value, so that more adhesive droplets are obtained in the depression area, while the standard adhesive amount remains unchanged outside the depression area; then, the printhead sprays an array of adhesive droplets according to the compensated data; finally, the adhesive itself spreads and flows to form a uniform adhesive film, which is then cured. Through the above steps, the loss of adhesive due to the increased gap in the recessed area is specifically compensated, thereby ensuring that the adhesive film thickness is uniform after curing, and fundamentally eliminating the defects of thin adhesive film and incomplete coverage.

[0035] In a specific embodiment, both the standard adhesive distribution data and the compensated adhesive distribution data are in grayscale image format. The unidirectional compensation is achieved by increasing the grayscale value of the corresponding pixel in the recessed area. The standard adhesive distribution data adopts the following pixel-to-nozzle mapping method: in the direction perpendicular to the scanning direction of the inkjet printhead, one pixel column corresponds to one nozzle; in the direction parallel to the scanning direction, one nozzle corresponds to multiple pixel rows during the scanning process.

[0036] A grayscale image is a two-dimensional data matrix with pixels as the basic unit. Each pixel has a grayscale value, which represents the amount of adhesive at that pixel location. Using grayscale image format as the carrier of adhesive distribution data allows for continuously adjustable adhesive control. Unidirectional compensation in a grayscale image manifests as increasing the grayscale value of the pixel corresponding to the concave area from the standard value to the compensation value; the increment of the grayscale value is related to the degree of concavity.

[0037] In this embodiment, a fixed mapping rule is used between pixels in the grayscale image and the physical nozzles of the inkjet printhead. Perpendicular to the scanning direction, which is the arrangement direction of the inkjet printhead nozzles, a one-to-one mapping method is used, with each nozzle responsible for spraying the amount of adhesive corresponding to the pixel column directly below it. Parallel to the scanning direction, which is the movement direction of the workpiece-carrying motion platform relative to the inkjet printhead, a one-to-many mapping method is used, with each nozzle corresponding to multiple pixel rows during scanning. The printhead triggers spraying row by row as the motion platform moves, and the same nozzle corresponds to different pixel rows at different times. Through this mapping method, all pixel positions in the standard adhesive distribution data can be uniquely mapped to the physical position of the printhead spraying. This mapping method ensures that the position of each pixel in the grayscale image precisely corresponds to the physical position on the workpiece surface, thereby enabling the compensation grayscale to be accurately applied to the actual recessed position.

[0038] In a specific embodiment, the concave area where the spray gap increases is determined by calculating the height deviation field ΔZ. The height deviation field ΔZ is the height difference between the actual three-dimensional shape data and the standard workpiece shape at the same position along the vertical Z-axis of the device; ΔZ(x,y)=Z_act(x,y)-Z_ref(x,y), where (x,y) represents the pixel coordinate position, Z_act(x,y) is the Z-axis height value of the actual workpiece at position (x,y), Z_ref(x,y) is the Z-axis height value of the standard workpiece at the corresponding position, and ΔZ<0 indicates that the actual workpiece is too low and the spray gap is greater than the standard working gap d0.

[0039] This embodiment further defines the quantitative determination method for recessed areas. The height deviation field ΔZ uses the vertical Z-axis of the device (i.e., the printhead jetting direction) as the calculation benchmark. This choice allows ΔZ to directly reflect the actual jetting gap change between the printhead and the workpiece surface. Here, Z_act(x,y) is the Z-axis height value of the actual workpiece at position (x,y), and Z_ref(x,y) is the Z-axis height value of the standard workpiece at the corresponding position. ΔZ(x,y) = Z_act(x,y) - Z_ref(x,y) represents the height difference between the actual workpiece and the standard workpiece at the same position in the Z-axis direction. When ΔZ < 0, it indicates that the surface of the actual workpiece is lower than the standard position; this is a recessed area, and the actual jetting gap between the printhead and this area, d_act = d0 + |ΔZ|, is greater than the standard working gap d0. When ΔZ ≥ 0, it indicates that the area is flat or raised, and the actual jetting gap is not greater than the standard working gap. By calculating the ΔZ distribution across the entire field, all recessed areas requiring compensation can be located.

[0040] More specifically, the actual three-dimensional topography data is obtained by measuring with a 3D line laser profile scanner, and the measured actual three-dimensional point cloud is rigidly registered with the reference point cloud generated by the standard CAD model through an iterative nearest point algorithm to unify them into the same coordinate system.

[0041] A 3D line laser profiler projects a line laser onto the workpiece surface. A camera captures images of the reflected laser lines, and triangulation is used to calculate the workpiece surface height at the location of the laser line. The scanner moves along the workpiece surface, acquiring dense profile lines one by one. These lines are then stitched together to form an actual 3D point cloud covering the entire workpiece surface. The Iterative Closest Point (ICP) algorithm is a point cloud registration algorithm that iteratively optimizes the rigid transformations (including rotation and translation) between the actual and reference point clouds to achieve optimal spatial alignment. After registration, the two point clouds are unified to the same coordinate system, allowing for point-by-point calculation of height deviations under the same coordinate reference. Rigid registration assumes that the workpiece itself does not deform, only undergoes rigid displacement and rotation, which aligns with workpiece posture changes caused by manufacturing tolerances and clamping errors.

[0042] In addition, before performing the measurement, a precision mesh calibration plate is used to calibrate the coordinate systems of the measuring equipment and the inkjet printhead, obtaining the transformation matrix between the two coordinate systems. The precision mesh calibration plate is an optical calibration plate with known precise dimensions and feature point positions, on which regularly arranged markers are distributed. During calibration, the calibration plate is first scanned with a 3D line laser profile scanner to obtain the position coordinates of the markers in the measuring equipment's coordinate system. Then, the inkjet printhead sprays a test pattern at the same position on the calibration plate to obtain the position coordinates of the markers in the printing coordinate system. Through the correspondence between the two sets of coordinates, the transformation matrix (including rotation and translation components) between the measuring coordinate system and the printing coordinate system is calculated. After calibration, the position of any recessed area obtained by the measuring equipment can be accurately mapped to the spraying position of the printhead through this transformation matrix, ensuring that the compensation adhesive is applied to the correct position.

[0043] The compensation amount of the unidirectional compensation is determined based on a pre-calibrated correlation model between the amount of adhesive, the spray gap, and the film thickness. The model is h=F(G,d), where d is the spray gap, G is the gray value, and h is the film thickness that can be achieved after spreading and fusion. For regions where ΔZ<0, the actual gap d_act=d0+|ΔZ| is calculated, and the compensation gray value G_comp is calculated to satisfy h_target=F(G_comp,d0+|ΔZ|).

[0044] The correlation model h=F(G,d) is the core basis for compensation calculation. This model reflects the relationship between three key variables: the larger the spray gap d, the smaller the spreading diameter of the adhesive droplet when it reaches the workpiece surface under the same gray level G, and the thinner the film thickness; the larger the gray level G, the more adhesive is sprayed, and the thicker the film. Given the target film thickness h_target and the actual spray gap d_act=d0+|ΔZ|, a compensation gray level G_comp needs to be solved so that the amount of adhesive sprayed at this gap and gray level, after spreading and fusion, exactly reaches the target film thickness. For recessed areas where ΔZ<0, since the actual gap is larger than the standard gap, the solved G_comp will definitely be greater than the standard gray level G0, thus achieving positive compensation of the adhesive amount. For areas where ΔZ≥0, this calculation is not performed. This model allows the compensation amount to be precisely adjusted according to the recess depth, avoiding excessive thickness or waste caused by blindly increasing the adhesive amount.

[0045] In a specific embodiment, the compensation grayscale is calculated within the linear interval of the model according to the following formula: G_comp(x,y)=G0(x,y)+k·|ΔZ(x,y)|, where G0 is the standard grayscale value in the standard glue quantity distribution data, and k is a positive compensation coefficient obtained through experimental calibration; the calibration method of the compensation coefficient k is as follows: a series of test blocks of different heights are made, the actual glue output at each height is measured, and the k value is obtained by linear regression fitting.

[0046] When the indentation depth is within a certain range, the film thickness, spray gap, and grayscale value can be approximately linearly related. In this case, the difference between the compensated grayscale value G_comp and the standard grayscale value G0 is proportional to the indentation depth |ΔZ|, with a proportionality coefficient of k. The compensation coefficient k is positive, and its physical meaning is the grayscale increment required per unit indentation depth. The calibration method for k is as follows: A series of test blocks with different heights are fabricated, and the height difference of these test blocks simulates different indentation depths. At a standard working gap, an array of adhesive dots of the same standard grayscale is sprayed onto each test block. After curing, the actual film thickness at each height is measured. Through linear regression analysis, the relationship between the actual film thickness (or adhesive output) and the height of the test block is fitted to a straight line; the slope of this line is the k value. This calibration method is simple and reliable, and can be reused after a single calibration.

[0047] More specifically, the model h=F(G,d) is calibrated as follows: a standard test substrate is prepared, and the type of adhesive, ambient temperature and humidity, resting time and curing parameters are fixed; multiple different spray gaps are set as gap gradients, and multiple different gray values ​​are set as gray gradients under each gap; an array of adhesive dots is sprayed under each calibrated gradient combination, and after static diffusion and UV curing, the actual adhesive film thickness is measured; the measurement data is fitted into a curve function.

[0048] Calibration experiments should strictly control fixed conditions, including adhesive type, ambient temperature and humidity, settling time, and curing parameters, to ensure that the calibration results are consistent with actual production process conditions. During calibration, multiple spray gap values ​​are first set as gap gradients, and then multiple grayscale values ​​are set under each gap as grayscale gradients. For each gap-grayscale combination, an array of adhesive dots is sprayed. After applying the same settling time and UV curing parameters to all adhesive dot arrays, the thickness of the cured adhesive film is measured. All measurement data are then summarized, and a surface function h=F(G,d) is fitted with spray gap and grayscale values ​​as independent variables and adhesive film thickness as the dependent variable. This fitted function can be a linear function, a polynomial function, or a more complex curve function, depending on the adhesive properties and the distribution of the actual measurement data.

[0049] When the calculated compensation grayscale exceeds the maximum grayscale limit of the inkjet printhead, the compensation grayscale is clamped to the maximum grayscale limit value, and the relative movement speed between the inkjet printhead and the workpiece is kept constant; the film thickness is based on the final dry film thickness after UV curing and shaping, and the calculation of the height deviation field ΔZ and the calculation of the compensation grayscale are both based on the final dry film thickness as the target value.

[0050] The piezoelectric elements and drive circuitry of the inkjet printhead have a maximum ejection capacity limit, which corresponds to a maximum grayscale upper limit. When the recess depth is too large, the compensated grayscale calculated according to the model may exceed this upper limit. In this case, the compensated grayscale of that pixel is clamped to the maximum grayscale upper limit, i.e., output according to the maximum physical ejection capacity of the printhead. At the same time, the relative movement speed between the printhead and the workpiece is kept constant. The reason for not reducing the movement speed is that once the speed is reduced, the residence time of the printhead at all positions increases synchronously, the physical amount of adhesive corresponding to the standard grayscale will change, and the benchmark of the entire compensation algorithm will be destroyed. Therefore, for extremely deep recesses, even if it is not possible to fully compensate to the target thickness, the maximum capacity should be output. In addition, the adhesive film thickness is based on the final dry film thickness after UV curing and shaping, i.e., the calculation of the height deviation field ΔZ and the calculation of the compensated grayscale are both based on the final dry film thickness as the target value, so that the target physical quantities of the entire process control are unified.

[0051] In a specific embodiment, the compensated adhesive distribution data is halftone processed and parsed to generate a waveform signal that drives the piezoelectric element in the inkjet printhead, so as to control the inkjet printhead to spray the adhesive droplet array at a constant height d0, forming an array of adhesive dots with spatially differentiated adhesive distribution.

[0052] After compensation, the grayscale image is processed by the main control computer, first undergoing halftone processing. Halftone is a technique that converts a multi-level grayscale image into binary or finite-level drive signals that the printhead can recognize, while maintaining grayscale gradation through spatial frequency modulation. Subsequently, the processed data is parsed into voltage waveform signals that drive the piezoelectric elements of the printhead. Different grayscale values ​​correspond to different waveform parameters (such as voltage amplitude, pulse width, or number of pulses), thereby precisely controlling the volume of the adhesive droplets ejected each time. During ejection, the printhead operates within a defined standard working gap d0, ejecting larger droplets from concave areas and standard-volume droplets from flat areas, ultimately forming an array of adhesive dots with spatially differentiated adhesive volume distribution.

[0053] In a specific embodiment, the spacing P between adjacent adhesive droplets in the droplet array satisfies: P ≤ √2·R_spread, where R_spread is the diffusion radius of a single adhesive droplet after it has spread and stabilized on the surface of the workpiece to be coated. The diffusion radius R_spread is measured by spraying isolated adhesive dots onto a standard test substrate, allowing them to stand under the same environmental conditions as the actual process until they stabilize, measuring the diameter D of the adhesive dots, and calculating R_spread = D / 2. After the adhesive droplet array is sprayed, it is allowed to stand for a preset time to form a continuous wet film, which is then cured and shaped by UV irradiation. The preset time is 15 seconds.

[0054] The relationship between the droplet spacing P and the diffusion radius R_spread determines whether the adhesive droplet array can fuse into a continuous film. When the center-to-center distance P between adjacent droplets is too large, even if the droplets are fully spread, gaps may still exist between adjacent droplets, resulting in a discontinuous film. When P ≤ √2·R_spread, the spreading fronts of adjacent droplets will inevitably contact and fuse, ensuring the formation of a continuous film. The diffusion radius R_spread is measured as follows: a single isolated adhesive dot is sprayed onto a flat standard test substrate and left to stand under environmental conditions identical to the actual process, such as temperature and humidity. After the adhesive dot diffusion stabilizes, the diameter D of the dot is measured using an optical microscope, and D / 2 is taken as the diffusion radius R_spread. After the adhesive droplet array is sprayed, it is left to stand for a preset time to form a continuous wet film, which is then cured and shaped by UV irradiation. The preset time can be set according to the adhesive viscosity, surface tension, and environmental conditions, usually based on the complete spreading of the adhesive dots and the contact of the leading edges of adjacent dots. It should be noted that R_spread is only related to the properties of the adhesive itself and the process environment. Once the type of adhesive and process conditions are determined, R_spread becomes a fixed value.

[0055] The standard adhesive distribution data corresponds to a standard working gap d0, and the unidirectional compensation is only activated when the depth of the depression in the depression area exceeds a preset negative deviation threshold; and a buffer band is set on the outer edge of the contour of the workpiece to be coated, and the amount of adhesive corresponding to the buffer band is lower than the minimum amount of adhesive in the standard adhesive distribution data.

[0056] The standard adhesive distribution data is obtained under standard working gap d0, meaning the standard adhesive value corresponds to the spraying conditions under the standard gap. The purpose of setting a negative deviation threshold is to avoid unnecessary compensation for minor surface roughness fluctuations. Actual workpiece surfaces may have minute surface roughness fluctuations. While these fluctuations may produce ΔZ<0 values, their depth is typically on the order of micrometers, insufficient to have a noticeable impact on the adhesive film thickness. Compensating for all these minute fluctuations would significantly increase computational load and could lead to uneven adhesive distribution due to overcompensation. Therefore, a negative deviation threshold is set, and compensation is only initiated when the indentation depth (i.e., |ΔZ|) exceeds this threshold, ensuring targeted compensation while improving computational efficiency. On the other hand, a buffer zone is set along the outer edge of the product contour. This is a transitional region with controllable width added outside the actual product boundary in the compensated adhesive distribution data. The adhesive value in this region is lower than the minimum adhesive value in the standard adhesive distribution data. The buffer zone's function is to gradually reduce the adhesive value at the boundary to a lower level, preventing adhesive overflow into non-coated areas of the workpiece or contamination of the carrier.

[0057] In a specific embodiment, after the adhesive droplet array spreads, fuses, and solidifies to form a continuous adhesive film, the method further includes: measuring the thickness of the solidified continuous adhesive film to obtain actual adhesive film thickness distribution data; comparing the actual adhesive film thickness distribution data with the target adhesive film thickness to determine whether the coating quality is qualified; when it is qualified, conveying the workpiece to be coated to the unloading station for unloading; when it is unqualified, issuing an alarm signal or marking and rejecting the unqualified workpiece.

[0058] After curing, the entire cured adhesive film is scanned and inspected to obtain the actual adhesive film thickness distribution data. The measured thickness distribution data is compared position-by-position with the target adhesive film thickness to calculate quality indicators such as thickness deviation and coefficient of variation. Based on preset acceptance criteria (e.g., thickness deviation not exceeding a certain percentage of the target value), the adhesive coating quality of the workpiece is determined to be acceptable. If acceptable, the workpiece is conveyed to the unloading station for unloading, and can then proceed to subsequent processes. If unacceptable, the system issues an audible and visual alarm to alert the operator, or physically marks the unacceptable workpiece, or automatically removes it from the production line. This re-inspection process constitutes a complete closed-loop quality control, ensuring the controllability and traceability of product quality.

[0059] Furthermore, when the thickness of a local area in the actual adhesive film thickness distribution data is lower than the target adhesive film thickness, the location information of the local area is fed back to the inkjet printhead, increasing the unidirectional compensation amount for the corresponding position of the next workpiece in the same batch. When a re-inspection reveals that the thickness of a specific local area of ​​a workpiece is lower than the target value, it indicates that the current compensation amount set for that area is still insufficient to compensate for the adhesive loss caused by the indentation. The system feeds back the location information and thickness deviation value of the local area to the compensation calculation module. When processing the next workpiece in the same batch, the compensation amount is further increased based on the existing compensation amount at that position (e.g., increasing the grayscale increase value at that position). Since workpieces in the same batch usually have similar manufacturing tolerance distributions, this feedback learning mechanism can iteratively optimize the compensation parameters piece by piece, continuously improving the compensation accuracy as production progresses. This mechanism fully utilizes the re-inspection data from each adhesive application, achieving adaptive evolution of process parameters.

[0060] In a specific embodiment, the thickness detection is performed using the same 3D line spectral analyzer or 3D line laser profile scanner used to measure the actual three-dimensional morphology data, and the thickness detection is completed online after the continuous adhesive film has been UV cured and shaped, and before the workpiece to be coated leaves the coating station.

[0061] Using the same equipment as the one used to measure the actual 3D morphology data for post-curing thickness measurement offers several advantages: First, the same equipment can be used for both pre-coating workpiece morphology measurement and post-coating film thickness measurement, reducing equipment costs. Second, the coordinate systems of the two sets of data are naturally consistent, allowing for direct comparison of thickness distribution data with the height deviation field ΔZ, facilitating analysis of compensation effects and process optimization. Regarding the timing of inspection, thickness measurement is completed online after UV curing and before the workpiece leaves the coating station, meaning it's performed on the same production line and at the same workstation, eliminating the need to transfer the workpiece to offline inspection equipment. This online full inspection method avoids the risk of missed inspections that may occur with offline sampling, while not affecting production cycle time, achieving a balance between high efficiency and high quality.

[0062] To verify the effectiveness of this embodiment, the following comparative experiment was conducted: The experiment used a test workpiece that was 50 mm long and 30 mm wide. The surface of the workpiece was designed with systematic height deviations, including sinusoidal waveform fluctuations with amplitudes ranging from -200 μm (dimples) to +150 μm (protrusions), to simulate surface morphology deviations caused by actual manufacturing tolerances and clamping errors.

[0063] The experiment used the following fixed process parameters: standard printhead working gap d0 = 1.0 mm, target curable film thickness h_target = 50 μm, adhesive settings as a typical UV curable adhesive, diffusion and fusion time uniformly set to 15 seconds, print resolution 600 DPI, grayscale value range 0–255, standard grayscale G0 = 180 (corresponding to a target thickness of 50 μm). The coefficient k in the compensation model was determined through pre-calibration.

[0064] To verify the technical effectiveness of this invention, which "only compensates for concave areas and does not reduce the amount of adhesive for convex areas," three comparative schemes were set up for comparison. Scheme A is a fixed grayscale scheme without compensation, where all positions are sprayed with standard grayscale G0 without any height deviation compensation. Scheme B is a symmetrical bidirectional compensation scheme, where grayscale G=G0-k·ΔZ is reduced for convex areas, and grayscale G=G0+k·|ΔZ| is increased for concave areas, achieving symmetrical compensation for all deviations. Scheme C is the unidirectional compensation scheme of this invention, where grayscale G=G0+k·|ΔZ| is increased only for concave areas, while the grayscale G0 remains unchanged for convex areas.

[0065] The experiment used four evaluation indicators: the coefficient of variation of adhesive thickness (CV, standard deviation / mean), which measures the overall uniformity of the adhesive film; minimum adhesive thickness, which reflects the most serious risk of adhesive shortage; defect rate, which is defined as the percentage of pixel area with adhesive thickness less than 70% of the target value (i.e., less than 35μm), which is used to characterize the severity of defects such as incomplete coverage and adhesive breakage; and robustness to printhead fluctuations, which observes the extent of the deterioration of the defect rate when ±8% random error is introduced into the printhead response.

[0066] The experimental results of the three schemes are shown in Table 1. Table 1 From the perspective of uniformity and defect control, Scheme C has a glue thickness variation coefficient of only 2.8%, which is far better than the 18.7% of the uncompensated scheme and also better than the 4.2% of the symmetrical bidirectional compensation scheme. More importantly, although Scheme B has a reasonable overall variation coefficient, its minimum glue thickness is only 23.1μm, and its defect rate is as high as 8.9%. Analysis shows that these serious defects are concentrated in the raised areas of the workpiece. This is because the symmetrical bidirectional compensation excessively reduces the grayscale in the raised areas, resulting in insufficient glue volume and discontinuous glue dots. In contrast, Scheme C in this embodiment does not reduce the glue volume in the raised areas, maintaining a minimum glue thickness of 44.6μm and a defect rate of only 0.7%, with almost no risk of incomplete coverage.

[0067] From the perspective of process robustness, when random errors in nozzle response are introduced, the defect rate of the symmetrical bidirectional compensation scheme B surges from 8.9% to 14.6%, indicating its extreme sensitivity to nozzle parameter fluctuations. In contrast, the defect rate of the present invention's scheme C only increases from 0.7% to 1.5%, demonstrating extremely strong process robustness. This is because the unidirectional compensation strategy retains the standard glue allowance in the raised areas, allowing the glue to naturally absorb disturbances during spreading and leveling, unlike the symmetrical bidirectional compensation which places every glue amount in a precise but fragile equilibrium.

[0068] The experimental data above verifies the practical effect of the asymmetric compensation concept in this embodiment. It not only achieves better uniformity but also completely eliminates irreversible defects caused by excessive glue reduction. This technical effect of achieving high robustness and high quality by simplifying the algorithm breaks through the conventional understanding of those skilled in the art that "height deviation must be compensated by symmetrical bidirectional compensation".

[0069] The inkjet coating method of this embodiment can be widely applied in precision manufacturing fields such as electronic packaging, optical bonding, and semiconductor mounting. In electronic packaging, it can be used for processes such as chip underfill and damming; in optical bonding, it can be used for coating optically transparent adhesive layers on touch screens and display panels; in semiconductor mounting, it can be used in scenarios with extremely high requirements for film thickness uniformity, such as wafer-level packaging and MEMS device packaging. As electronic devices develop towards miniaturization and high density, the impact of minute morphological deviations on the workpiece surface on film uniformity is becoming increasingly significant. This invention provides an efficient and precise solution with good application prospects.

[0070] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An inkjet coating method based on morphology compensation, characterized in that, include: Obtain standard adhesive distribution data corresponding to the morphology of a standard workpiece; Measure the actual three-dimensional morphology data of the workpiece to be coated, and compare the actual three-dimensional morphology data with the morphology of the standard workpiece to determine the concave area on the surface of the workpiece to be coated that causes the spray gap to increase relative to the standard state. The standard adhesive distribution data corresponding to the recessed area is unidirectionally compensated to increase the set adhesive amount in that area, while the standard adhesive distribution data for areas outside the recessed area remains unchanged, thus generating compensated adhesive distribution data. Based on the compensated adhesive distribution data, the inkjet printhead is controlled to spray an array of adhesive droplets onto the surface of the workpiece to be coated. The adhesive droplet array is spread and fused to form a continuous adhesive film and then cured.

2. The inkjet coating method based on morphology compensation according to claim 1, characterized in that, Both the standard adhesive distribution data and the compensated adhesive distribution data are in grayscale image format. The unidirectional compensation is achieved by increasing the grayscale value of the corresponding pixel in the recessed area. The standard adhesive distribution data adopts the following pixel-to-nozzle mapping method: in the direction perpendicular to the scanning direction of the inkjet printhead, one pixel column corresponds to one nozzle; in the direction parallel to the scanning direction, one nozzle corresponds to multiple pixel rows during the scanning process.

3. The inkjet coating method based on morphology compensation according to claim 1, characterized in that, The concave area where the spray gap increases is determined by calculating the height deviation field ΔZ. The height deviation field ΔZ is the height difference between the actual three-dimensional shape data and the standard workpiece shape at the same position along the vertical Z-axis of the equipment. ΔZ(x,y)=Z_act(x,y)-Z_ref(x,y), where (x,y) represents the pixel coordinate position, Z_act(x,y) is the Z-axis height value of the actual workpiece at position (x,y), and Z_ref(x,y) is the Z-axis height value of the standard workpiece at the corresponding position. ΔZ<0 indicates that the actual workpiece is too low and the spray gap is greater than the standard working gap d0.

4. The inkjet coating method based on morphology compensation according to claim 1, characterized in that, The actual three-dimensional topographic data is obtained by measuring with a 3D line laser profile scanner, and the measured actual three-dimensional point cloud is rigidly registered with the reference point cloud generated by the standard CAD model through an iterative nearest point algorithm to unify them into the same coordinate system.

5. The inkjet coating method based on morphology compensation according to claim 1, characterized in that, Before performing the measurement, a precision grid calibration plate is used to calibrate the coordinate system of the measuring equipment and the inkjet printhead to obtain the transformation matrix between the two coordinate systems.

6. The inkjet coating method based on morphology compensation according to claim 3, characterized in that, The compensation amount of the unidirectional compensation is determined based on a pre-calibrated correlation model between the amount of adhesive, the spray gap, and the film thickness. The model is h=F(G,d), where d is the spray gap, G is the gray value, and h is the film thickness that can be achieved after spreading and fusion. For regions where ΔZ<0, the actual gap d_act=d0+|ΔZ| is calculated, and the compensation gray value G_comp is calculated to satisfy h_target=F(G_comp,d0+|ΔZ|).

7. The inkjet coating method based on morphology compensation according to claim 6, characterized in that, The compensation gray value G_comp is calculated within the linear interval of the model according to the following formula: G_comp(x,y)=G0(x,y)+k·|ΔZ(x,y)|, where G0 is the standard gray value in the standard glue quantity distribution data, and k is a positive compensation coefficient obtained through experimental calibration. The calibration method of the compensation coefficient k is as follows: a series of test blocks of different heights are made, the actual glue output at each height is measured, and the k value is obtained by linear regression fitting.

8. The inkjet coating method based on morphology compensation according to claim 6, characterized in that, The model h=F(G,d) was calibrated as follows: a standard test substrate was prepared, and the type of adhesive, ambient temperature and humidity, resting time and curing parameters were fixed; multiple different spray gaps were set as gap gradients, and multiple different gray values ​​were set as gray gradients under each gap; an array of adhesive dots was sprayed under each calibrated gradient combination, and after resting, diffusion and UV curing, the actual adhesive film thickness was measured; the measurement data were fitted into a curve function.

9. The inkjet coating method based on morphology compensation according to claim 6, characterized in that, When the calculated compensation grayscale exceeds the maximum grayscale limit of the inkjet printhead, the compensation grayscale is clamped to the maximum grayscale limit value, and the relative motion speed between the inkjet printhead and the workpiece is kept constant. The film thickness is based on the final dry film thickness after UV curing and shaping. The calculation of the height deviation field ΔZ and the calculation of the compensation grayscale are both based on the final dry film thickness as the target value.

10. The inkjet coating method based on morphology compensation according to claim 3, characterized in that, The compensated adhesive distribution data is halftone processed and parsed to generate a waveform signal that drives the piezoelectric element in the inkjet printhead, thereby controlling the inkjet printhead to eject the adhesive droplet array at a constant height d0, forming an array of adhesive dots with spatially differentiated adhesive distribution.

11. The inkjet coating method based on morphology compensation according to claim 1, characterized in that, The spacing P between adjacent adhesive droplets in the droplet array satisfies: P ≤ √2·R_spread, where R_spread is the diffusion radius of a single adhesive droplet after it has spread and stabilized on the surface of the workpiece to be coated; the diffusion radius R_spread is measured by spraying isolated adhesive dots onto a standard test substrate, allowing them to stand under the same environmental conditions as the actual process until they stabilize, measuring the diameter D of the adhesive dots, and calculating R_spread = D / 2; the adhesive droplet array is allowed to stand for a preset time after spraying to form a continuous wet film, and then cured and shaped by UV irradiation.

12. The inkjet coating method based on morphology compensation according to claim 1, characterized in that, The standard adhesive distribution data corresponds to a standard working gap d0; the unidirectional compensation is only activated when the depth of the depression in the depression area exceeds a preset negative deviation threshold; and a buffer band is provided on the outer edge of the outline of the workpiece to be coated, the amount of adhesive corresponding to the buffer band being lower than the minimum amount of adhesive in the standard adhesive distribution data.

13. The inkjet coating method based on morphology compensation according to claim 1, characterized in that, After the adhesive droplet array spreads, fuses, and solidifies to form a continuous adhesive film, the process further includes: measuring the thickness of the solidified continuous adhesive film to obtain actual adhesive film thickness distribution data; comparing the actual adhesive film thickness distribution data with the target adhesive film thickness to determine whether the coating quality is qualified; when qualified, conveying the workpiece to be coated to the unloading station for unloading; when unqualified, issuing an alarm signal or marking and rejecting the unqualified workpiece.

14. The inkjet coating method based on morphology compensation according to claim 13, characterized in that, When the thickness of a local area in the actual film thickness distribution data is lower than the target film thickness, the location information of the local area is fed back to the inkjet printhead, and the compensation amount of the unidirectional compensation is increased for the corresponding position of the next workpiece in the same batch.

15. The inkjet coating method based on topography compensation according to claim 13, characterized in that, The thickness detection is performed using the same 3D line spectrum measuring instrument or 3D line laser profile scanner used to measure the actual three-dimensional topography data, and the thickness detection is completed online after the continuous adhesive film has been UV cured and shaped, and before the workpiece to be coated leaves the coating station.

Images