A PCB multi-layer board stacking system

Abstract

This invention discloses a PCB multilayer board stacking system, including an intelligent inspection machine and an automatic riveting machine. The intelligent inspection machine is used to stack PCB multilayer board materials to form a stacked structure, and provides intelligent error prevention management for the stacking process. During the stacking process, the intelligent inspection machine identifies and compares the specifications, picking position, and real-time weight of the stacked structure based on the stacking information. The intelligent inspection machine integrates a laser alignment device for real-time monitoring of the relative offset between layers of the stacked structure. The laser alignment device includes a laser transcoding module for recognizing the etched QR code on the surface of the inner core board and transcoding it into a through-hole QR code for subsequent X-ray identification and traceability. This invention can achieve comprehensive closed-loop verification of material specifications, picking position, and stacked weight, which is suitable for subsequent traceability and can effectively ensure the interlayer alignment accuracy of the stacking process, improving the accuracy and efficiency of PCB multilayer board stacking.

Description

Technical Field

[0001] This invention relates to the field of PCB manufacturing technology, and more particularly to a PCB multilayer board stacking system. Background Technology

[0002] As electronic technology continues to develop towards high speed, multi-functionality, large capacity, and portability and low power consumption, printed circuit boards (PCBs), which serve as the support for electronic components and the electrical connection carrier for electronic components, have gradually evolved from single-layer boards to double-layer and multi-layer boards, and continue to develop towards multi-layering, high density, high precision, small aperture, fine wires, high reliability, and high transmission speed.

[0003] Multilayer boards are manufactured by alternately stacking and laminating multiple inner core boards with different circuit patterns and insulating dielectric layers. They have a certain load-bearing area and can meet the needs of higher precision and more complex circuit designs. The PCB multilayer board manufacturing process includes material preparation, stacking, lamination and fixing, and board removal. During the production of multilayer boards, especially those with 10 or more layers, interlayer slippage is the main cause of PCB multilayer board defects, and interlayer slippage leading to interlayer misalignment is closely related to the stacking process. In existing technologies, to control the interlayer offset of multilayer boards, the PIN-LAM process is generally used for laminating PCBs with 10 or more layers. In the PIN-LAM process, relying on a rigid platform with pins, each layer of material is manually inserted into the pins in sequence. The rigidity of the pins and the matching accuracy between the pins and the positioning holes ensure the position of each layer in the horizontal X / Y directions. After the multilayer board is locked with pins, subsequent heating and lamination are performed for fixation. After lamination, the pins need to be removed.

[0004] The above process has the following drawbacks: due to the fixed size of the steel plate, only one set of products can be stacked on each steel plate, resulting in poor flexibility. The stacking process relies on manual stacking, and due to the limitations of the stacking steel plate and the support plate, the overall weight after stacking is large, resulting in high labor intensity, low production efficiency, and difficulty in improving the efficiency of the PCB stacking process, thus becoming a bottleneck restricting the production of multilayer PCBs.

[0005] While some automated pre-stacking equipment exists in the prior art, such as the PCB lamination pre-stacking equipment and process disclosed in patent document CN108337824A, these devices mainly focus on the cutting, punching, automatic conveying, and riveting of prepreg. They merely replace manual stacking with automation, reducing the probability of human error, but do not improve the methods for fixing and eliminating errors in multilayer boards. Therefore, they cannot fundamentally solve the problem of interlayer alignment accuracy and cannot guarantee the interlayer alignment accuracy of multilayer PCBs. Furthermore, because the inner core boards of multiple layers are stacked together, their surface QR codes are covered by polypropylene (PP). During the lamination process, the PP melts and fills the QR codes, potentially causing the QR code function to fail and making it unsuitable for subsequent traceability.

[0006] Furthermore, for the automatic stacking of high-layer ultra-thin boards, in addition to error-proofing management and comprehensive closed-loop verification of material specifications, picking positions, and stacking weight to prevent incorrect layer / material placement, accurately controlling interlayer offset and eliminating both incoming material errors and cumulative errors is extremely challenging. Moreover, the control difficulty increases exponentially with the number of layers. Currently, the industry standard for controlling layer offset in 10-32 layer ultra-thin boards is generally only ±3mil. The main limitation is: 1. Traditional rivets will "bloom" (radial expansion) during riveting. For ultra-thin boards, this tiny radial force is enough to cause interlayer slippage. It is generally believed in the industry that riveting will inevitably cause radial bloom, so high-multilayer boards must use the PIN-LAM process. 2. In the PIN-LAM process, the clearance between the pin and the positioning hole will wear down after repeated insertion and removal, resulting in a decrease in accuracy; 3. During the stacking process, the expansion and contraction differences of each core board were not carefully managed, resulting in excessive cumulative errors. Summary of the Invention

[0007] The purpose of this invention is to provide a PCB multilayer board stacking system that can realize comprehensive closed-loop verification of material specifications, material picking position, and stacking weight, which is suitable for subsequent traceability and can effectively ensure the interlayer alignment accuracy of the stacking process, thereby improving the stacking accuracy and efficiency of PCB multilayer boards.

[0008] To solve the above-mentioned technical problems, the embodiments of the present invention provide a technical solution as follows: A PCB multilayer board stacking system, comprising: an intelligent inspection machine, used for stacking the inner core board and insulating dielectric layer materials of the PCB multilayer board to form a stacked structure, and for real-time monitoring and error prevention management of the stacking process; the intelligent inspection machine integrates an industrial control computer, an image recognition device, a weighing device, and a laser alignment device; an automatic riveting machine, used for automatically feeding rivets into the riveting holes of the stacked structure for riveting and fixing; the intelligent inspection machine obtains data from the MES system. The production instruction sheet is retrieved and the stacking information is parsed. The stacking information includes material specifications, material picking position, and process weight information. During the stacking process, the intelligent inspection machine identifies and compares the current material specifications, material picking position, and real-time weight of the stacked structure based on the stacking information. The laser alignment device is used to monitor the relative offset between layers of the stacked structure in real time. The laser alignment device is equipped with a laser transcoding module, which is used to identify the etched QR code on the surface of the inner core board and transcode it into a through-hole QR code for subsequent X-ray identification and traceability.

[0009] Furthermore, the intelligent inspection machine also has a material preparation rack, which includes multiple layers of storage space. Each storage space is equipped with a signal indicator light and a motion detection sensor. The industrial control computer controls the signal indicator light of the corresponding storage space to light up according to the parsed material picking position data to indicate the material picking position. The weighing device is integrated into the inspection platform of the intelligent inspection machine. The weighing device is used to obtain the real-time superimposed weight of the material on the inspection platform after each layer of material is stacked. The industrial control computer is configured to: acquire the coded image of the current material through an image recognition device and compare it with the material specification data; after confirming that the material specification is correct, control the signal indicator lights of the material preparation rack to light up sequentially and verify the actual material picking position through the motion detection sensor; after each layer of material is stacked, obtain the real-time superimposed weight through the weighing device and compare it with the process weight information.

[0010] Furthermore, the automatic riveting machine includes a riveting station and a first loading station and a second loading station located on both sides of the riveting station; the industrial control computer is also configured to dynamically allocate stacking tasks according to the carrier occupancy status of the first loading station and the second loading station, so that the stacked structure output by the two intelligent inspection machines stacking in parallel alternately enters the riveting station to eliminate the waiting time of the riveting station; the riveting station is equipped with multiple riveting heads for synchronously riveting multiple riveting holes on the stacked structure.

[0011] Furthermore, the intelligent inspection machine includes a first inspection machine and a second inspection machine. The first inspection machine is set up corresponding to the first feeding station, and the second inspection machine is set up corresponding to the second feeding station. A robotic arm is provided between the first inspection machine and the first feeding station, and between the second inspection machine and the second feeding station. By stacking the two intelligent inspection machines in parallel, and by having the first feeding station and the second feeding station alternately supply materials to the same riveting station, a continuous production system is formed.

[0012] Furthermore, the intelligent inspection machine also includes a UVW alignment platform; the image recognition device, in conjunction with the visual guidance device, performs target recognition on each inner core board during the stacking process, and drives the UVW alignment platform to execute a three-axis movement compensation algorithm to automatically correct the interlayer alignment deviation, so that the alignment accuracy reaches ±1mil.

[0013] Furthermore, it also includes a expansion and contraction measurement module, used to measure the expansion and contraction deviation of preset target points on the inner core board, and classify and combine them according to the deviation value to reduce the interlayer alignment error caused by expansion and contraction; the expansion and contraction measurement module uses laser to measure the position deviation of at least three non-collinear expansion and contraction target points on the inner core board, and calculates the X / Y axis expansion and contraction coefficient and rotational deviation angle of the inner core board; the industrial control computer determines the inner core boards with expansion and contraction coefficient deviation within ±1mil as the first group, and the inner core boards with expansion and contraction coefficient deviation within ±1mil to ±2mil as the second group, and forces the inner core boards in the same group to be combined into a set for stacking; when the expansion and contraction deviation of a certain layer of material exceeds the preset deviation range, the industrial control computer controls the robot to alternately feed material from two opposite directions to offset the cumulative interlayer error.

[0014] Furthermore, it also includes a board bending and warping pressing device, which is set on the stacking path of the intelligent inspection machine and has a built-in pressure feedback sensor; the industrial control computer dynamically adjusts the output pressure and holding time of the board bending and warping pressing device according to the preset thickness and material parameters of the hard board to be pressed, and applies pressure to the already pressed hard board before or during each stacking to restore it to flatness.

[0015] Furthermore, it also includes an X-RAY re-inspection system, which is set after the automatic riveting machine. It is used to automatically take alignment mark images after riveting is completed, compare them with preset standards, upload the alignment deviation data to the MES system, and automatically mark and alarm for defective products.

[0016] Furthermore, the rivet is a limited-expansion rivet, whose internal cavity is composed of coaxial stepped holes. During riveting, it only undergoes axial compression and does not produce radial expansion. When the number of laminate layers is 10-32 and the total thickness does not exceed 10mm, the wall thickness of the rivet is controlled to be 0.2-1.0mm, and the axial compression during riveting is 10%-20% of the original length of the rivet. After the rivet is deformed by riveting, its radial expansion rate is limited to ≤1%, and the fit clearance with the riveting hole is always maintained at ≤0.05mm.

[0017] Furthermore, the laser transcoding module includes a laser engraving device, which is linked with the image recognition device. After recognizing the etched QR code, the device uploads the QR code information to the MES system and engraves through-hole QR codes on the inner core board using the laser engraving device. The through-hole QR codes are filled with PP resin after lamination and can be read by an X-RAY device.

[0018] The PCB multilayer board stacking system provided by this invention, compared with existing technologies, integrates an industrial control computer, image recognition device, weighing device, and laser alignment device. During the stacking process, it identifies and compares the specifications, picking position, and real-time weight of the current material and the stacked structure. This organically integrates multiple error-proofing measures into a comprehensive closed-loop verification system encompassing material specifications, picking position, and stacked weight. This achieves intelligent error-proofing management of the stacking process, effectively improving product yield and operational standardization. Through automatic alignment, expansion and contraction compensation, board bending correction, laser transcoding, and X-ray... RAY re-inspection can form a more complete closed-loop system, which can not only effectively ensure the alignment accuracy between layers and improve the yield of PCB boards, but also reduce the dependence on manual labor in the stacking operation, further improving the stacking accuracy and efficiency of multilayer PCB boards and reducing production costs. In addition, the present invention uses the alternating feeding of the double-sided feeding station of the automatic riveting machine to enable the riveting station to work continuously without interruption, realizing continuous production; the rigidity of the rivets can be effectively enhanced by the non-split rivets and the rivet cavities formed by coaxial stepped holes, and the synchronous riveting of multiple riveting heads can reduce the amount of interlayer slippage during the riveting process. This allows the present invention to meet the stacking requirements of multilayer PCB boards with 10-32 layers and a total thickness of no more than 10mm, with an alignment accuracy of ±1mil between layers, solving the industry problem of riveting ultra-thin boards with high layer counts. Attached Figure Description

[0019] One or more embodiments are illustrated by way of example with reference to the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings represent similar elements. Unless otherwise stated, the figures in the drawings do not constitute a limitation on scale.

[0020] Figure 1 This is a perspective view of the PCB multilayer board stacking system in an embodiment of the present invention; Figure 2 This is a top view of the PCB multilayer board stacking system in an embodiment of the present invention; Figure 3 This is a partial structural diagram of the automatic riveting machine in an embodiment of the present invention; Figure 4 This is a schematic diagram of the intelligent inspection machine structure in an embodiment of the present invention; Figure 5 This is a schematic longitudinal cross-sectional view of the thick-walled non-flowering rivet in an embodiment of the present invention.

[0021] Explanation of reference numerals in the attached drawings: 10. Intelligent inspection machine; 101. Inspection table; 102. Material preparation rack; 1021. Motion detection sensor; 1022. Signal indicator light; 103. Industrial control computer; 104. Image recognition device; 105. Weighing device; 11. First inspection machine; 12. Second inspection machine; 20. Automatic riveting machine; 201. Feeding station; 21. First feeding station; 22. Second feeding station; 23. Riveting station; 231. Riveting head; 232. Vibratory feeder; 233. Material distribution channel; 24. Conveying device; 30. Rivet; 31. Body; 32. Base; 33. Internal cavity. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the various embodiments of this invention will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been provided in the various embodiments of this invention to facilitate a better understanding of this application. However, the technical solutions claimed in the claims of this application can be implemented even without these technical details and with various variations and modifications based on the following embodiments.

[0023] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a specific posture. If the specific posture changes, the directional indicators will also change accordingly.

[0024] like Figure 1-5 As shown, one embodiment of the present invention provides a PCB multilayer board stacking system, which includes an intelligent inspection machine 10 and an automatic riveting machine 20. The intelligent inspection machine 10 integrates an industrial control computer 103, an image recognition device 104, a weighing device 105, a motion detection sensor 1021, and a signal indicator 1022. The industrial control computer 103 communicates with the image recognition device 104, the weighing device 105, the laser alignment device, the motion detection sensor 1021, and the signal indicator 1022 to realize operation status display, material picking operation indication, alarm prompts, and data storage. The laser alignment device is used to monitor the relative offset between layers of the stacked structure in real time, which can realize all-round closed-loop verification of material specifications, picking position, and stacking weight, which is suitable for subsequent traceability and can effectively ensure the interlayer alignment accuracy of the stacking process.

[0025] To further improve the accuracy and efficiency of PCB multilayer board stacking, this invention also includes a expansion and contraction measurement module, a board bending and warping pressing device, and an X-ray machine. The RAY re-inspection system is used for real-time monitoring of the lamination process, encompassing material identification, automatic alignment, expansion and contraction compensation, plate bending correction, laser transcoding, and X-ray inspection. RAY's complete intelligent error prevention management closed-loop system for re-inspection.

[0026] Laser transcoding function: Previously, the QR codes on the inner core boards were etched onto the circuit board surface. After lamination, due to the stacking of multiple inner core boards and the surface being covered by PP, the PP melts and fills the QR codes during lamination, rendering the QR code function ineffective at this stage and unsuitable for subsequent traceability. Therefore, this invention adds a laser engraving function, using a laser alignment device to monitor the relative offset between layers of the stacked structure in real time. The laser alignment device includes a laser transcoding module that transcodes the identified etched QR codes into through-hole QR codes and uploads them to the MES (Manufacturing Execution System). After subsequent lamination and molding, only PP resin remains inside the QR code holes, with no metal parts. The content of the QR code can be identified via X-ray, and combined with the MES system, complete traceability can be achieved. Preferably, the engraving depth does not exceed 50% of the copper thickness of the inner core board to avoid damaging the underlying circuitry; after engraving, there is no copper residue on the inner wall of the through-hole, and after lamination, the PP resin completely fills the through-hole.

[0027] The new automatic alignment system for inner layers addresses the previous method of manual stacking and alignment using a PIN fixture for initial positioning, followed by precise positioning on the equipment. The new system employs a vision-guided system to identify targets for each inner layer. A UVW alignment platform (a three-axis parallel high-precision positioning module capable of X / Y translation and θ rotation with a single command) is added to the equipment platform. Utilizing vision guidance combined with the UVW three-axis movement algorithm, automatic alignment between inner core boards is achieved. The vision system identifies targets at the four corners of the inner core board, calculating the offsets ΔX, ΔY, and rotation angle Δθ between the current target and the standard position. The UVW platform automatically compensates for the deviation using a three-axis displacement calculation formula, ultimately achieving an interlayer alignment accuracy of ±1 mil.

[0028] A new expansion and contraction measurement module is added: After the inner core board is manufactured, four expansion and contraction target points are made on it. The equipment adds a laser measurement system to measure the expansion and contraction of these four target points and classify them. Preferably, the expansion and contraction measurement module uses laser to measure the positional deviation of multiple expansion and contraction target points on the inner core board and classifies them according to a preset deviation range (such as ±1mil, ±2mil). Inner core boards with the same deviation range are combined into a set for stacking. The specific control is as follows: The expansion and contraction measurement module uses laser to measure the positional deviation of at least three non-collinear expansion and contraction target points on the inner core board and calculates the X / Y axis expansion and contraction coefficient and rotational deviation angle of the inner core board; the industrial control computer 103 controls the expansion and contraction coefficient deviation within ±1mil. The inner core boards within the specified area are classified as the first group, and those with a shrinkage coefficient deviation within the range of ±1mil to ±2mil are classified as the second group. Inner core boards within the same group are then forcibly combined into a single stack. When the shrinkage deviation of a certain layer exceeds a preset deviation range, the industrial control computer 103 controls a robotic arm to alternately feed material from two opposing directions to offset accumulated interlayer errors. This effectively solves the problem of inaccurate interlayer alignment caused by shrinkage issues, overcoming incoming material errors at the source and improving the overall lamination yield. The shrinkage measurement module can be integrated into the intelligent inspection machine or be a separate peripheral device located at the feeding end of the intelligent inspection machine, measuring the shrinkage as the robotic arm picks up the inner core board and passes over it.

[0029] New Board Bending and Warping Pressing Device: With the rise of the AI ​​industry, the entire PCB manufacturing process has been improved. Many processes now use N+N processes, where boards that have already been riveted and pressed undergo secondary riveting and pressing, or even more pressings, to increase the number of riveting layers. A problem arises in this process: after the first pressing, the board becomes a rigid board, and due to the high temperature, unevenness and warping are inevitable. This leads to misalignment during the riveting and stacking process due to unevenness between layers. This invention introduces a new board bending and warping pressing device that can correct the board's flatness, ensuring that product yield is not reduced due to this problem. Preferably, the board bending / warping pressing device is installed on the stacking path of the intelligent inspection machine 10 and has a built-in pressure feedback sensor. The industrial control computer 103 dynamically adjusts the output pressure and holding time of the pressing device according to the preset thickness and material parameters of the rigid board to be pressed, so as to eliminate board bending / warping while avoiding damage to the internal circuitry. That is, pressure is applied to the already pressed rigid board before or during each stacking to restore its flatness, so as to ensure the accuracy of subsequent interlayer alignment. For FR-4 rigid boards with a thickness of 0.5-3mm, the preferred output pressure is 0.5-2MPa, the holding time is 5-15s, and the pressure feedback sensor sampling frequency is 100Hz, adjusting the pressure in real time to avoid overpressure damage to the circuitry.

[0030] X-RAY Inspection System: With the original equipment, after riveting, the machine was immediately moved to the next station for alignment inspection. The newly added X-RAY inspection system, after riveting, automatically captures alignment marks via X-RAY to check if customer requirements are met, and uploads the alignment deviation to the MES system. Preferably, the X... The X-RAY re-inspection system is installed after the automatic riveting machine 20. It automatically captures alignment mark images after riveting is completed, compares them with preset standards, uploads alignment deviation data to the MES system, and automatically marks and alarms for defective products. The X-RAY re-inspection system can be integrated into the intelligent inspection machine or used as a standalone peripheral device.

[0031] The end-to-end traceability system categorizes and packages the data of each product for both existing and new functions, uploading it to the MES system and retaining records locally for 3 years. Additionally, all personnel-operated parts of the equipment are recorded by video, allowing for real-time traceability of each PCS and each operator. Traceability data includes order information, material codes for each layer, operator ID, lamination time, expansion / contraction coefficient, alignment deviation, riveting parameters, and X-ray inspection results, retained locally for at least 3 years, meeting PCB industry traceability requirements.

[0032] The industrial computer 103 of this invention is configured to perform the following cross-validation: The image recognition device 104 acquires the coded image of the current material and compares it with the material specification data; After confirming that the material specifications are correct, the signal indicator lights 1022 of the control rack 102 light up in sequence, and the actual material picking position is verified by the motion detection sensor 1021. After each layer of material is stacked, the real-time superimposed weight is obtained by the weighing device 105 and compared with the process weight information. This triple-layer error prevention system, combining image recognition, weighing, and position verification, eliminates the possibility of misplacement of layers or materials, achieving comprehensive closed-loop verification of material specifications, pick-up position, and stack weight. In actual production, image recognition verification allows a maximum of three retries, weight verification deviations within ±2g can be manually confirmed before release, and pick-up position verification errors directly lock the equipment, requiring authorized personnel to unlock it, thus significantly improving the yield of Grade A boards. Of course, during debugging, image recognition errors can also immediately stop the machine and trigger an alarm. The specific number of retries and weight verification deviations can be set according to the actual debugging situation.

[0033] The PCB multilayer board stacking material of the present invention includes an inner core board and an insulating dielectric layer. The inner core board is a flexible or rigid substrate with circuitry, and the insulating dielectric layer is a flexible film, such as PP film. Multiple through-hole riveting holes are pre-processed around the perimeter of the inner core board and the insulating dielectric layer. An intelligent inspection machine 10 is used to stack the inner core board and the insulating dielectric layer to form a stacked structure, and to monitor and manage the stacking process in real time. The intelligent inspection machine 10 is configured to obtain production instructions from the MES system and parse the stacking information, including material specifications, picking position, and process weight information. During the stacking process, the intelligent inspection machine 10 identifies and compares the current material specifications, picking position, and real-time weight of the stacked structure based on the parsed stacking information to achieve a comprehensive closed-loop verification of material specifications, picking position, and stacking weight. When a mismatch occurs, the machine is stopped and an intelligent identification alarm is executed. For example, when the image recognition device 104 identifies a type A PP board as the material specification, but the preset material specification is type B PP board, the system immediately issues a red alarm via the signal indicator 1022. Simultaneously, the industrial control computer 103 displays an error message and suspends the stacking operation. The motion detection sensor 1021 monitors the material retrieval position in each storage space of the material preparation rack 102. An alarm is triggered when the retrieval position does not match the preset retrieval position. For example, if the preset retrieval position requires retrieving material from the first layer of the material preparation rack 102, but the motion detection sensor 1021 detects that the operator is retrieving material from the second layer of the material preparation rack 102, the system immediately triggers an alarm. The weighing device 105 monitors the weight of the stacked structure in real time. An alarm is triggered when the actual weight does not match the preset process weight information. For example, if the preset process weight is 100g, but the weighing device 105 detects an actual weight of 120g, the system determines that there may be excess or incorrect material and can immediately trigger an alarm to confirm whether there is leakage or excess material.

[0034] An automatic riveting machine 20 is used to automatically feed rivets 30 into the riveting holes of the laminated structure for riveting and fixing. The rivets 30 are thick-walled, non-split rivets 30, whose internal cavity 33 is composed of coaxial stepped holes. During riveting, only axial compression occurs without radial splitting deformation, and the fit clearance between the thick-walled, non-split rivet 30 and the riveting hole is ≤0.05mm. The automatic riveting machine 20 includes a riveting station 23 and a first feeding station 21 and a second feeding station 22 located on both sides of the riveting station 23. The first feeding station 21 and the second feeding station 22 are configured to alternately feed the laminated structure to the riveting station 23. The riveting station 23 is equipped with multiple riveting heads 231 for synchronously riveting multiple riveting holes on the laminated structure. Preferably, the intelligent inspection machine 10 includes a first inspection machine 11 and a second inspection machine 12. The first inspection machine 11 is correspondingly arranged with the first feeding station 21, and the second inspection machine 12 is correspondingly arranged with the second feeding station 22. Robotic arms are provided between the first inspection machine 11 and the first feeding station 21, and between the second inspection machine 12 and the second feeding station 22. By having two intelligent inspection machines 10 stacked in parallel and two feeding stations 201 alternately feeding materials to the same riveting station 23, a continuous production system is formed, effectively reducing the idle time of the riveting station 23 and increasing production capacity.

[0035] The intelligent inspection machine 10 integrates an industrial control computer 103, an image recognition device 104, a weighing device 105, a motion detection sensor 1021, and a signal indicator 1022. During the stacking process, it identifies and compares the specifications of the current material, the picking position, and the real-time weight of the stacked structure. It organically integrates multiple error prevention measures into one, forming a comprehensive closed-loop verification of material specifications, picking position, and stacking weight. If any link is mismatched, the machine will automatically stop and alarm, fundamentally eliminating human errors such as using the wrong material, placing the wrong layer, or stacking in the wrong order, and significantly improving the first-pass yield of multilayer board stacking. Through the alternating feeding of the double-sided feeding stations 201 of the automatic riveting machine 20, the riveting station 23 can work continuously without interruption. Compared with the existing single-line serial operation method, it effectively eliminates the idle waiting time of the riveting station 23 and realizes continuous production; through the non-flowering rivet 30 and the internal cavity 33 of the rivet 30 composed of coaxial stepped holes, the rigidity of the rivet 30 can be effectively enhanced, and with the multi-riveting head 231 for synchronous riveting, the amount of interlayer slippage during the riveting process is reduced.

[0036] One embodiment relates to a PCB multilayer board stacking system, comprising an intelligent inspection machine 10 and an automatic riveting machine 20. The intelligent inspection machine 10 is used to stack the inner core boards and insulating dielectric layers of the PCB multilayer board to form a stacked structure, and to monitor and manage the stacking process in real time. The intelligent inspection machine 10 integrates an industrial control computer 103, a material preparation rack 102, motion detection sensors 1021 and signal indicator lights 1022 integrated on the material preparation rack 102, an inspection table 101, a weighing device 105 integrated on the inspection table 101, and an image recognition device 104 positioned above the inspection table 101. The material preparation rack 102 includes multiple storage spaces for storing inner core boards and insulating dielectric layers of different specifications. Each storage space is equipped with a signal indicator light 1022 and a motion detection sensor 1021. The industrial control computer 103 obtains the production instruction sheet from the MES system and parses the stacking information. Based on the parsed material picking position data, the industrial control computer 103 controls the corresponding storage space's signal indicator light 1022 to light up to indicate the material picking position. The operator or robot arm picks up the material from the lit position according to the indication. The motion detection sensor 1021 detects the actual material picking action and records the actual material picking position. The industrial control computer 103 compares the actual material picking position with the parsed material picking position data to perform material picking position verification. Once an incorrect material picking position is detected, the system can control the machine to stop and issue an alarm.

[0037] The inspection table 101 serves as the work area for material stacking. A weighing device 105 is integrated onto the inspection table 101. The weighing device 105 is used to obtain the real-time weight of the materials stacked on the inspection table 101 after each layer of material is stacked. The intelligent inspection machine 10 compares the obtained real-time weight with the process weight data parsed from the production instruction sheet to verify the weight of the currently stacked single piece of material and the total weight for the stage. When a weight mismatch is detected, the system stops the machine and issues an alarm to indicate potential material leakage, excess material, or incorrect material.

[0038] An image recognition device 104 is suspended above the inspection table 101. The image recognition device 104 is used to acquire the coded image of the material and identify the material code information. The intelligent inspection machine 10 compares the identified material code information with the material specification data parsed from the production instruction sheet to achieve material specification verification. If the identified material does not match the preset specifications, the system can immediately control the machine to stop and issue an alarm. In addition, a laser alignment device or calibration cursor (not shown in the attached diagram) is also provided on one side of the image recognition device 104. The laser alignment device or calibration cursor is used to provide visual positioning calibration when placing materials, ensuring that the materials are accurately placed at the predetermined position on the inspection table 101.

[0039] After the stacking is completed, the intelligent inspection machine 10 generates a traceable production log of the current stacked structure's product information and uploads it to the MES system. The product information includes the material code of each layer, the material picking position, the operation time, and the operator's information, to facilitate subsequent quality traceability.

[0040] One embodiment relates to a PCB multilayer board stacking system, which includes an intelligent inspection machine 10 and an automatic riveting machine 20. The automatic riveting machine 20 has a riveting station 23 equipped with multiple riveting heads 231, which, in conjunction with riveting molds (not shown in the figures), are used to simultaneously rivet multiple riveting holes on the stacked structure. The riveting heads 231 are connected to an automatic rivet feeding device 30, which includes a vibratory feeder 232 and a material distribution channel 233, enabling automatic feeding of thick-walled, non-split rivets 30 to the riveting heads 231. Each riveting head 231 is connected to a corresponding vibratory feeder 232 and a material distribution channel 233. Each riveting head 231 is individually fed through its corresponding vibratory feeder 232 and material distribution channel 233, achieving simultaneous riveting and continuous automatic stamping riveting of multiple riveting holes. This prevents interlayer misalignment caused by uneven stress in the stacked structure, further improving riveting efficiency and increasing the production efficiency of PCB multilayer board stacking.

[0041] Preferably, the automatic riveting machine 20 can also automatically adjust the riveting pressure of the riveting head 231 according to the specifications of the rivets 30 used and the total thickness of the stacked structure to be riveted, so as to ensure the consistency of riveting quality.

[0042] like Figure 5 As shown, the rivet 30 used in the system is a thick-walled, non-split rivet, whose internal cavity 33 is formed by coaxial stepped holes. During the riveting process, the thick-walled, non-split rivet only undergoes axial compression and does not experience radial splitting deformation. The clearance between the thick-walled, non-split rivet and the riveting hole is no greater than 0.05mm to ensure positioning accuracy. Preferably, when the number of stacked PCB layers is 10-32 and the total thickness does not exceed 10mm, the wall thickness of the rivet 30 is controlled to be 0.2-1.0mm, and the axial compression during riveting is 10%-20% of the original length of the rivet 30. After riveting deformation, the radial expansion rate of the rivet 30 is limited to ≤1%, and the clearance between it and the riveting hole is always maintained at ≤0.05mm to suppress interlayer slippage. This allows the present invention to meet the requirements of multilayer PCB stacking with 10-32 layers and a total thickness not exceeding 10mm. Through actual testing, by adopting triple cross-verification of image, weighing, and position, and realizing a comprehensive closed-loop verification of material specifications, material picking position, and stacking weight, the yield rate of the Class A board of this invention has been increased from 95% to 99.9%. By combining expansion and contraction classification, alternating feeding, and expansion-limiting rivets, the layer offset of 10-32 layer ultra-thin boards has been reduced from the industry-standard ±3mil to ±1mil, solving the industry problem of stacking and riveting of high-layer ultra-thin boards.

[0043] In one example, the rivet 30 is made of high-purity copper. It includes a cylindrical body 31 with an internal cavity 33, and a base 32 extending radially outward from one end of the body 31. The internal cavity 33 is formed by two coaxially arranged stepped holes, with the diameter of the stepped hole near the base 32 being larger than the diameter of the end away from the base 32. This internal step structure provides better support, enhancing the support strength of the rivet 30 and further ensuring controllable axial deformation during the riveting process.

[0044] The PCB multilayer board stacking system provided by this invention can achieve comprehensive closed-loop verification of material specifications, material picking position, and stacking weight; and can further integrate material identification, automatic alignment, expansion and contraction compensation, board bending correction, laser transcoding, and X-ray machine processing. RAY's complete intelligent error-proofing management closed-loop system effectively improves the stacking accuracy, product yield, and standardization of PCB multilayer boards, reduces the reliance on manual labor in stacking operations, increases efficiency, and reduces production costs. The specific effective results are shown in Table 1. Table 1 1000 pieces each of 10-layer, 18-layer, and 32-layer ultra-thin PCBs were selected for testing. The inner core board used in the test was made of conventional FR-4 material, the insulating dielectric layer was made of 1080 type PP sheet, and the rivet specifications were 0.5mm wall thickness and 15% axial compression. The test environment temperature was 25±5℃ and the humidity was 50±10%RH. After using the system of this invention, the interlayer alignment deviations were ±0.8mil, ±0.9mil, and ±1mil, respectively, and the yield of Grade A boards was ≥99.9%. In contrast, the same batch of samples using the traditional PIN-LAM process had interlayer alignment deviations of ±2.8mil, ±3.1mil, and ±3.5mil, respectively, and the yield of Grade A boards was only 94.2%, 92.7%, and 90.5%. This shows that the present invention can control the interlayer alignment accuracy of 10-32 layer ultra-thin PCBs to ±1mil, and improve the yield of Grade A boards to 99.9%, solving the industry problem of riveting and stacking of high-layer ultra-thin boards.

[0045] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be defined by the claims.

Claims

1. A PCB multilayer board stacking system, characterized in that, include: The intelligent inspection machine (10) is used to perform stacking operations on the inner core board and insulating dielectric layer materials of PCB multilayer board to form a stacked structure, and to perform real-time monitoring and error prevention management of the stacking process; the intelligent inspection machine (10) integrates an industrial control computer (103), an image recognition device (104), a weighing device (105) and a laser alignment device; An automatic riveting machine (20) is used to automatically feed rivets (30) into the riveting holes of the laminated structure for riveting and fixing; The intelligent inspection machine (10) obtains the production instruction table from the MES system and parses the stacking information. The stacking information includes material specifications, picking position and process weight information. During the stacking process, the intelligent inspection machine (10) identifies and compares the current material specifications, picking position and real-time weight of the stacked structure based on the stacking information. The laser alignment device is used to monitor the relative offset between layers of the stacked structure in real time. The laser alignment device is equipped with a laser transcoding module, which is used to identify the etched QR code on the surface of the inner core board and transcode it into a through-hole QR code for subsequent X-ray identification and traceability.

2. The PCB multilayer board stacking system according to claim 1, characterized in that, The intelligent inspection machine (10) also has a material preparation rack (102), which includes multiple layers of storage space. Each storage space is equipped with a signal indicator (1022) and a motion detection sensor (1021). The industrial control computer (103) controls the signal indicator (1022) of the corresponding storage space to light up according to the parsed material picking position data to indicate the material picking position. The weighing device (105) is integrated on the inspection platform (101) of the intelligent inspection machine (10). The weighing device (105) is used to obtain the real-time weight of the material on the inspection platform (101) after each layer of material is stacked. The industrial control computer (103) is configured as follows: The image recognition device (104) acquires the coded image of the current material and compares it with the material specification data; After confirming that the material specifications are correct, the signal indicator (1022) of the control rack (102) lights up in sequence, and the actual material picking position is verified by the motion detection sensor (1021). After each layer of material is stacked, the real-time superimposed weight is obtained by a weighing device (105) and compared with the process weight information.

3. The PCB multilayer board stacking system according to claim 1, characterized in that, The automatic riveting machine (20) includes a riveting station (23) and a first feeding station (21) and a second feeding station (22) located on both sides of the riveting station (23); The industrial control computer (103) is also configured to dynamically allocate stacking tasks according to the carrier occupancy status of the first loading station (21) and the second loading station (22), so that the stacked structure output by the two intelligent inspection machines (10) stacking in parallel alternately enters the riveting station (23) to eliminate the waiting time of the riveting station. The riveting station (23) is equipped with multiple riveting heads (231) for synchronously riveting multiple riveting holes on the laminated structure.

4. The PCB multilayer board stacking system according to claim 3, characterized in that, The intelligent inspection machine (10) includes a first inspection machine (11) and a second inspection machine (12). The first inspection machine (11) is set up in correspondence with the first loading station (21), and the second inspection machine (12) is set up in correspondence with the second loading station (22). A robotic arm is provided between the first inspection machine (11) and the first loading station (21), and between the second inspection machine (12) and the second loading station (22). A continuous production system is formed by stacking the two intelligent inspection machines (10) in parallel and feeding the same riveting station (23) alternately from the first loading station (21) and the second loading station (22).

5. The PCB multilayer board stacking system according to claim 1, characterized in that, The intelligent inspection machine (10) also includes a UVW alignment platform; The image recognition device (104), in conjunction with the visual guidance device, performs target recognition on each inner core board during the stacking process and drives the UVW alignment platform to execute a three-axis movement compensation algorithm to automatically correct the interlayer alignment deviation, so that the alignment accuracy reaches ±1mil.

6. The PCB multilayer board stacking system according to claim 1, characterized in that, It also includes a expansion and contraction measurement module, which is used to measure the expansion and contraction deviation of preset target points on the inner core board and classify and combine them according to the deviation value to reduce the interlayer alignment error caused by expansion and contraction. The expansion and contraction measurement module uses laser to measure the positional deviation of at least three non-collinear expansion and contraction target points on the inner core plate, and calculates the X / Y axis expansion and contraction coefficient and rotational deviation angle of the inner core plate. The industrial control computer (103) determines the inner core boards with expansion and contraction coefficient deviation within ±1mil as the first group, and the inner core boards with expansion and contraction coefficient deviation within ±1mil to ±2mil as the second group, and forces the inner core boards in the same group to be combined into a set for stacking. When the expansion and contraction deviation of a certain layer of material exceeds the preset deviation range, the industrial control computer (103) controls the robot to alternately feed material from two opposite directions to offset the cumulative error between layers.

7. The PCB multilayer board stacking system according to claim 1, characterized in that, It also includes a plate bending and warping pressing device, which is set on the stacking path of the intelligent inspection machine (10) and has a built-in pressure feedback sensor; The industrial control computer (103) dynamically adjusts the output pressure and holding time of the plate bending and warping pressing device according to the preset thickness and material parameters of the hard plate to be pressed. Before or during each stacking, pressure is applied to the hard plate that has been pressed to restore it to flatness.

8. The PCB multilayer board stacking system according to claim 1, characterized in that, It also includes an X-RAY re-inspection system, which is set after the automatic riveting machine (20) to automatically take alignment mark images after riveting is completed and compare them with preset standards. The alignment deviation data is uploaded to the MES system, and non-conforming products are automatically marked and alarms are triggered.

9. The PCB multilayer board stacking system according to claim 1, characterized in that, The rivet (30) is a limited expansion type rivet, and its internal cavity (33) is composed of a coaxial stepped hole. During riveting, it is only axially compressed and does not cause radial splitting. When the number of laminated plates is 10-32 layers and the total thickness does not exceed 10mm, the wall thickness of the rivet (30) is controlled to be 0.2-1.0mm, and the axial compression during riveting is 10%-20% of the original length of the rivet (30); After the rivet (30) is deformed by riveting, its radial expansion rate is limited to ≤1%, and the fitting clearance with the riveting hole is always maintained at ≤0.05mm.

10. The PCB multilayer board stacking system according to claim 1, characterized in that, The laser transcoding module includes a laser engraving device, which is linked with the image recognition device (104). After recognizing the etched QR code, the QR code information is uploaded to the MES system, and the laser engraving device engraves through-hole QR codes on the inner core board. The through-hole QR codes are filled with PP resin after pressing and can be read by an X-RAY device.

Images