PCB binding area anomaly detection device and method

Abstract

The invention provides a PCB binding area anomaly detection device and method, and the device comprises a detection platform which is used for bearing a bound product to be detected; the CCD camera is used for positioning the bound product to be detected; the adjusting mechanism is used for adjusting the position of the detection table according to the positioning information of the bound to-be-detected product; the spectrum confocal sensor is arranged above the detection table; the spectral confocal sensor is used for scanning the bound to-be-detected product through a spectral line to obtain data point information corresponding to the bound to-be-detected product; the processor is electrically connected with the spectrum confocal sensor and used for forming a curve graph according to the data point information and judging whether the binding area is abnormal or not according to the height difference Zn of the wave crest and the wave trough of the curve graph. According to the invention, online detection of the PCB binding area is realized, the working efficiency is improved, and unqualified products are prevented from flowing into back-end equipment to cause loss.

Description

Technical Field

[0001] The present invention relates to the technical field of binding equipment, and in particular to a device and method for detecting abnormality in a PCB binding area. Background Art

[0002] During the binding process between PCB (printed circuit board) and FPC (flexible printed circuit) or COF (chip on flexible substrate), due to foreign matter mixing, poor binding or ACF (anisotropic conductive film) attachment offset, bulge and other problems, the binding area often has appearance defects such as floating and bulging, which in turn causes abnormal conductivity and makes the screen unable to light up normally. The appearance of a normal binding area will be like this: Figure 1 As shown in A, defective products such as Figure 1 B. Currently, such problems are difficult to detect effectively during mass production, resulting in batches of defective products flowing into subsequent production links, causing significant losses.

[0003] Existing detection methods mainly rely on manual microscopic inspection, which is not only inefficient and easily affected by human factors, but can only conduct random inspections and cannot achieve comprehensive, online detection. Summary of the invention

[0004] In order to solve the above-mentioned defects, the present invention aims to provide a PCB binding area abnormality detection device and method that is efficient, accurate and capable of online detection, so as to replace the traditional manual detection method and improve production efficiency and product quality.

[0005] A PCB binding area abnormality detection device comprises: a detection platform, which is used to carry a bound product to be inspected; a CCD camera, which is used to position the bound product to be inspected; an adjustment mechanism, which is arranged at the lower side of the detection platform, and the adjustment mechanism is used to adjust the position of the detection platform according to the positioning information of the bound product to be inspected; a spectral confocal sensor, which is arranged above the detection platform; the spectral confocal sensor is used to scan the bound product to be inspected through a spectral line to obtain data point information corresponding to the bound product to be inspected; a processor, which is electrically connected to the spectral confocal sensor, and the processor is used to form a curve graph according to the data point information, and to obtain data point information corresponding to the peak and valley height difference Z of the curve graph according to the peak and valley height difference Z of the curve graph. n Determine whether there is any abnormality in the binding area.

[0006] In one embodiment of the present invention, the adjustment mechanism includes a base, a Y-axis linear module and an electric rotating platform. The detection platform is connected to the electric rotating platform through a mounting plate. The bottom of the electric rotating platform is installed on the Y-axis linear module, and the Y-axis linear module is installed on the base.

[0007] In one embodiment of the present invention, the PCB binding area abnormality detection device also includes a mounting frame, the mounting frame is vertically arranged with the base, an X-axis linear module is arranged on the mounting frame, and the spectral confocal sensor is installed on the X-axis linear module.

[0008] In one embodiment of the present invention, the PCB binding area abnormality detection device includes two detection platforms, and each of the detection platforms is respectively equipped with a corresponding adjustment mechanism.

[0009] In one embodiment of the present invention, a threshold parameter Z0 is set in the processor; when the height difference between the peak and the trough of the curve graph Zn≤Z0, it is judged to be qualified; when a convex area appears in the curve graph and the convex value Zn>Z0, it is judged to be unsatisfactory and an alarm is issued.

[0010] In a second aspect, the present invention provides a method for detecting abnormalities in a PCB binding area, comprising the following steps: scanning the bound product to be inspected along the X direction by a spectral confocal sensor, obtaining data point information corresponding to the bound product to be inspected, and transmitting the data point information to a processor, wherein the processor forms a curve graph according to the data point information, and calculates the height difference Z between the peak and the valley of the curve graph according to the height difference Z between the peak and the valley of the curve graph. n Determine whether there is any abnormality in the binding area; after calculation and analysis by the processor, if it is judged to be qualified, it will be transported to the next process; if it is judged to be defective, it will enter the NG buffer and wait for manual inspection.

[0011] In one embodiment of the present invention, a threshold parameter Z0 is set in the processor; when the height difference between the peak and the trough of the curve graph Zn≤Z0, it is judged to be qualified; when a convex area appears in the curve graph and the convex value Zn>Z0, it is judged to be unsatisfactory and an alarm is issued.

[0012] In one embodiment of the present invention, the processor uses the following formula to perform data processing:

[0013] 1) Quadratic fitting:

[0014] Given a data point (X_i, Z_i), where i = 1, 2, ..., n;

[0015] 2) Fitting model selection:

[0016] Quadratic fitting model: Z = aX^2 + bX + c;

[0017] 3) Calculate the fitting parameters:

[0018] Least squares method goal: find a, b, c so that the error sum of squares S is minimized:

[0019] S=Σ(Z_i-(aX_i^2+bX_i+c))^2;

[0020] By taking partial derivatives of S with respect to a, b, and c respectively, and setting them equal to 0:

[0021]

[0022] Form a system of linear equations:

[0023] ΣX_i^4a+ΣX_i^3b+ΣX_i^2c=ΣX_i^2Z_i;

[0024] ΣX_i^3a+ΣX_i^2b+ΣX_i c=ΣX_i Z_i;

[0025] ΣX_i^2a+ΣX_i b+nc=ΣZ_i;

[0026] In matrix form:

[0027]

[0028] 4) Calculate the residual:

[0029] Residual

[0030] 5) Identify outliers:

[0031] Set the threshold T;

[0032] Mark the points exceeding the threshold as outliers: |r_i|>T;

[0033] 6) Output the results.

[0034] In summary, the present invention provides a PCB binding area abnormality detection device and method, and the beneficial effects of the present invention are:

[0035] (1) Improve detection efficiency: Realize online, fully automatic detection, replace manual microscope inspection, and realize double-station detection of PCB binding area abnormalities, significantly improve detection efficiency, and save detection time.

[0036] (2) Enhanced detection accuracy: Through high-precision scanning and data analysis by the spectrometer, human interference can be reduced and detection accuracy can be improved.

[0037] (3) High flexibility: The scanning speed and other parameters can be flexibly adjusted according to different binding processes and the diameter of conductive particles, with strong adaptability.

[0038] (4) Reduce production costs: Timely discover and handle abnormalities in the binding area to reduce the flow of defective products into subsequent production links and reduce production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Schematic diagram of the appearance of qualified products and defective products after binding; Figure 1 A is a qualified product, Figure 1 B is a defective product.

[0040] Figure 2 A schematic diagram of the three-dimensional structure of a binding abnormality detection device provided in one embodiment of the present invention.

[0041] Figure 3 A front view of a binding anomaly detection device provided in one embodiment of the present invention.

[0042] Figure 4 It is a curve chart of qualified products after bundling and defective products after bundling; Figure 4 A is a qualified product, Figure 4 B is a defective product.

[0043] Explanation of main element symbols: 1. Base; 2. Mounting frame; 3. X-axis linear module; 4. Spectral confocal sensor; 51. First detection platform; 52. Second detection platform; 61. First electric rotating platform; 62. Second electric rotating platform; 71. First Y-axis linear module; 72. Second Y-axis linear module. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0045] like Figures 2 to 3 As shown, a PCB binding area abnormality detection device includes: a detection platform, a CCD camera, an adjustment mechanism, a spectral confocal sensor 4 and a processor.

[0046] The testing platform is used to carry the products to be tested after binding. In this embodiment, a first testing platform 51 and a second testing platform 52 are provided to realize double-station testing of PCB binding area abnormalities, thereby improving the testing efficiency.

[0047] Next, a CCD camera is set above the workstation to locate the products to be inspected after binding.

[0048] Then, the adjustment mechanism is arranged at the lower side of the testing platform, and the adjustment mechanism adjusts the position of the testing platform according to the positioning information of the product to be tested after binding.

[0049] In this embodiment, the first inspection table 51 and the second inspection table 52 are respectively provided with a first adjustment mechanism and a second adjustment mechanism, and the first adjustment mechanism and the second adjustment mechanism are installed on the base 1. The first adjustment mechanism includes a first Y-axis linear module 71 and a first electric rotating platform 61, and the first inspection table 51 is connected to the first electric rotating platform 61 through a mounting plate, and the bottom of the first electric rotating platform 61 is installed on the first Y-axis linear module 71, and the first Y-axis linear module 71 is installed on the base 1. The second adjustment mechanism includes a second Y-axis linear module 72 and a second electric rotating platform 62, and the second inspection table 52 is connected to the second electric rotating platform 62 through a mounting plate, and the bottom of the second electric rotating platform 62 is installed on the second Y-axis linear module 72, and the second Y-axis linear module 72 is installed on the base 1.

[0050] Next, the spectral confocal sensor 4 is set above the detection platform; the spectral confocal sensor 4 is used to scan the bound products to be inspected through the spectral line to obtain the data point information corresponding to the bound products to be inspected. The spectral confocal detection 4 uses a special lens to extend the focus halo range of different color lights to form a special amplification color difference, so that according to the different distances from the object to be tested to the lens, a precise wavelength of light will be focused on the object to be tested. By measuring the wavelength of the reflected light, the precise distance from the object to be tested to the lens can be obtained.

[0051] The processor is electrically connected to the spectral confocal sensor 4, and the processor is used to form a curve graph according to the data point information, and to generate a curve graph according to the height difference Z between the peak and the valley of the curve graph. n Determine whether there is any abnormality in the binding area.

[0052] Furthermore, the PCB binding area abnormality detection device also includes a mounting frame 2, which is vertically arranged with the base 1, and an X-axis linear module 3 is arranged on the mounting frame 2. The spectral confocal sensor 4 is installed on the X-axis linear module 3, and the spectral confocal sensor 4 moves along the X direction.

[0053] The processor is provided with a threshold parameter Z0. In this embodiment, Figure 4 A. When the height difference between the peak and valley of the curve is Z1≤Z0 (threshold parameter), it is judged as qualified. Figure 4 B. When a convex area appears on the curve graph and the convex value Z2>Z0 (threshold parameter), it is judged as bad and an alarm is issued.

[0054] In other embodiments, the processor processes the data obtained by the spectral confocal sensor 4 in the following manner:

[0055] (1) Quadratic fitting:

[0056] Given a data point (X_i, Z_i), where i = 1, 2, ..., n;

[0057] (2) Fitting model selection:

[0058] Quadratic fitting model: Z = aX^2 + bX + c;

[0059] (3) Calculate the fitting parameters:

[0060] Least squares method goal: find a, b, c so that the error sum of squares S is minimized:

[0061] S=Σ(Z_i-(aX_i^2+bX_i+c))^2;

[0062] By taking partial derivatives of S with respect to a, b, and c respectively, and setting them equal to 0:

[0063]

[0064] Form a system of linear equations:

[0065] ΣX_i^4a+ΣX_i^3b+ΣX_i^2c=ΣX_i^2Z_i;

[0066] ΣX_i^3a+ΣX_i^2b+ΣX_i c=ΣX_i Z_i;

[0067] ΣX_i^2a+ΣX_i b+nc=ΣZ_i;

[0068] In matrix form:

[0069]

[0070] (4) Calculate the residual:

[0071] Residual

[0072] (5) Identify outliers:

[0073] Set the threshold T;

[0074] Mark the points exceeding the threshold as outliers: |r_i|>T;

[0075] (6) Output the results.

[0076] Working process: The bound products to be inspected on the first inspection platform 51 and the second inspection platform 52 are inspected respectively by the spectral confocal sensor 4. When the spectral confocal sensor 4 scans the bound products to be inspected on the upper side of the first inspection platform 51, the CCD camera positions the bound products to be inspected on the upper side of the second inspection platform 52, and the second adjustment mechanism adjusts the Y direction and angle direction of the bound products to be inspected on its upper side. After the bound products to be inspected on the upper side of the first inspection platform 51 are scanned and processed, the spectral confocal sensor 4 moves along the X direction to the upper side of the second inspection platform 52 to scan the bound products to be inspected on its upper side. The bound products to be inspected on the upper side of the first inspection platform 51 are calculated and analyzed by the processor. If they are judged to be qualified, they are transported to the next process. If they are judged to be defective, they enter the NG buffer and wait for manual inspection.

[0077] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A PCB binding area abnormality detection device, characterized in that: It includes: A testing table, which is used to carry the products to be tested after binding; CCD camera, which is used to locate the products to be inspected after binding; An adjustment mechanism, which is arranged at the lower side of the testing platform, and is used to adjust the position of the testing platform according to the positioning information of the product to be tested after binding; A spectral confocal sensor is arranged above the detection platform; the spectral confocal sensor is used to scan the bound products to be inspected through spectral lines to obtain data point information corresponding to the bound products to be inspected; A processor is electrically connected to the spectral confocal sensor, and the processor is used to form a curve graph according to the data point information, and to generate a curve graph according to the height difference Z between the peak and the valley of the curve graph. n Determine whether there is any abnormality in the binding area.

2. The PCB binding area abnormality detection device according to claim 1, characterized in that: The adjustment mechanism includes a base, a Y-axis linear module and an electric rotating platform. The detection platform is connected to the electric rotating platform through a mounting plate. The bottom of the electric rotating platform is mounted on the Y-axis linear module, and the Y-axis linear module is mounted on the base.

3. The PCB binding area abnormality detection device according to claim 2, characterized in that: It also includes a mounting frame, which is vertically arranged with the base, an X-axis linear module is arranged on the mounting frame, and the spectral confocal sensor is installed on the X-axis linear module.

4. The PCB binding area abnormality detection device according to claim 1, characterized in that: It comprises two testing platforms, each of which is respectively equipped with a corresponding adjustment mechanism.

5. The PCB binding area abnormality detection device according to claim 1, characterized in that: The processor is provided with a threshold parameter Z0; when the height difference between the peak and the trough of the curve graph Zn≤Z0, it is judged as qualified; when a convex area appears in the curve graph and the convex value Zn>Z0, it is judged as unqualified and an alarm is issued.

6. A method for detecting abnormalities in a PCB binding area, characterized in that: The following steps are involved: The bound products to be inspected are scanned along the X direction by the spectral confocal sensor to obtain data point information corresponding to the bound products to be inspected, and the data point information is transmitted to the processor, and the processor forms a curve graph according to the data point information, and the height difference Z between the peak and the valley of the curve graph is calculated. n Determine whether there is an abnormality in the binding area; After calculation and analysis by the processor, if it is judged to be qualified, it will be transported to the next process. If it is judged to be defective, it will enter the NG buffer and wait for manual inspection.

7. The method for detecting abnormality in a PCB binding area according to claim 6, characterized in that: The processor is provided with a threshold parameter Z0; when the height difference between the peak and the trough of the curve graph Zn≤Z0, it is judged as qualified; when a convex area appears in the curve graph and the convex value Zn>Z0, it is judged as unqualified and an alarm is issued.

8. The method for detecting abnormality in a PCB binding area according to claim 6, characterized in that: The processor uses the following formula to process data: 1) Quadratic fitting: Given a data point (X_i, Z_i), where i = 1, 2, ..., n; 2) Fitting model selection: Quadratic fitting model: Z = aX^2 + bX + c; 3) Calculate the fitting parameters: Least squares method goal: find a, b, c so that the sum of squared errors S is minimized: S = Σ(Z_i-(aX_i^2+bX_i+c))^2; By taking partial derivatives of S with respect to a, b, and c respectively, and setting them equal to 0: Form a system of linear equations: ΣX_i^4a+ΣX_i^3b+ΣX_i^2c=ΣX_i^2Z_i; ΣX_i^3a+ΣX_i^2b+ΣX_i c=ΣX_i Z_i; ΣX_i^2a+ΣX_i b+nc=ΣZ_i; In matrix form: 4) Calculate the residual: Residual 5) Identify outliers: Set the threshold T; Mark the points exceeding the threshold as outliers: |r_i|>T; 6) Output the results.

Images