Laser cutting path generation method, device, medium and product for board-level package
By scanning the board-level packaging substrate to obtain the die coordinates and dynamically generating the laser cutting path, the problem that traditional cutting methods cannot adapt to die offset is solved, and high-yield and high-precision board-level packaging cutting is achieved.
Patent Information
- Application Number
- CN202610176526.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-06
- Publication Date
- 2026-05-29
Smart Images
Figure CN122121568A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of chip technology, and in particular to a method, apparatus, medium and product for generating laser cutting paths for board-level packaging. Background Technology
[0002] In the semiconductor packaging field, Panel Level Package (PLP) has attracted much attention due to its ability to significantly improve production efficiency and reduce costs. The PLP process typically involves mounting diced chips / dies onto a substrate using a die-bonding process, followed by molding to form a protective layer. However, during the high-temperature, high-pressure molding process, factors such as mold flow impact and mismatched thermal expansion coefficients of the materials can cause unpredictable random die shifts on the substrate. These shifts include not only horizontal displacement but also frequent deviations in rotational angle.
[0003] Existing board-level packaging cutting processes mainly employ knife wheel cutting or diamond wheel cutting. These traditional mechanical cutting methods are limited by the rigid structure of the physical cutting tool; the cutting path must be a straight line running through the entire substrate, making non-linear adjustments for localized areas impossible. When the aforementioned random displacement of the die on the substrate occurs, the fixed straight cutting path can easily damage the die itself, resulting in irreversible product scrap and yield loss.
[0004] Furthermore, existing global compensation or linear cutting techniques cannot provide effective solutions for the complex nonlinear distortions in board-level packaging caused by the molding process, where the offset direction and degree of each die are different. Summary of the Invention
[0005] One objective of this application is to provide a laser cutting path generation method, device, medium, and product for board-level packaging, at least to solve the problem that in the prior art, in board-level packaging processes, due to the random positional shift and rotational distortion of the die caused by the molding process, traditional straight-line cutting methods or cutting methods based on global parameter compensation cannot adapt to this irregular array distribution, resulting in cutting damage to the die or causing the package's appearance dimensions to exceed the standard, ultimately leading to low product yield.
[0006] To achieve the above objectives, some embodiments of this application provide the following aspects:
[0007] In a first aspect, some embodiments of this application provide a method for generating laser cutting paths for board-level packaging, the method comprising:
[0008] Scan the board-level packaging substrate after molding to obtain the actual position coordinate information of each die on the substrate;
[0009] Based on the actual position coordinates of the grains, the node coordinates are calculated; wherein, the node coordinates are used to determine the laser cutting paths that divide adjacent packages.
[0010] A point-to-point laser cutting path file is generated based on the node coordinates to control the laser equipment to perform cutting according to the laser cutting path file, thereby dividing the board-level packaging substrate into multiple independent packages.
[0011] Secondly, some embodiments of this application also provide an electronic device, the electronic device comprising: one or more processors; and a memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method described above.
[0012] Thirdly, some embodiments of this application also provide a computer-readable medium having computer program instructions stored thereon, which can be executed by a processor to implement the method described above.
[0013] Fourthly, some embodiments of this application also provide a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the method described above.
[0014] Compared with related technologies, the solution provided in this application firstly involves performing a high-precision position scan on each die before cutting, and dynamically generating a programmable laser cutting path based on the actual coordinates of the die (including the center position and rotation angle). This allows the cutting line to accurately follow the actual offset of the die and adaptively adjust. This means that even if the die undergoes irregular displacement or rotation, the laser can avoid the effective area of the die, thereby significantly reducing product scrap caused by die shift.
[0015] Secondly, through the independent position compensation algorithm, an independent closed-loop cutting path can be generated for each package, ensuring that the single product has extremely high centering accuracy; through the adjacent node balancing algorithm, the nodes of the common cutting channel can be determined by calculating the average value of the corner points of four adjacent dies, effectively solving the path planning problem when the offset directions of adjacent dies conflict, and maximizing the utilization rate of the substrate without damaging the product.
[0016] Furthermore, by using the same set of measured grain coordinates to generate both the cutting path and the RDL pattern, strict alignment between the cutting edge and the internal circuit layout is ensured, avoiding electrical failures caused by inconsistencies in the position data of inner and outer layers. Simultaneously, leveraging the small laser spot size, a narrower cutting track width (e.g., less than 250 micrometers) can be achieved compared to traditional cutting wheels, further improving the yield rate. Attached Figure Description
[0017] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.
[0018] Figure 1 An exemplary flowchart of a laser cutting path generation method for board-level packaging provided in some embodiments of this application;
[0019] Figure 2 A comparison diagram of defective product distribution generated by conventional fixed-pattern cutting in some embodiments of this application and defective product distribution generated by the method of this embodiment;
[0020] Figure 3 A schematic diagram of the intersection of laser cutting paths of four adjacent grains in a laser cutting path generation method for board-level packaging provided in some embodiments of this application;
[0021] Figure 4 A schematic diagram of a corner positioning mark of a board-level packaging substrate provided for some embodiments of this application;
[0022] Figure 5 A schematic diagram illustrating the spatial relationship between the laser cutting path and the package and its internal redistribution layer, provided for some embodiments of this application;
[0023] Figure 6 A schematic diagram of the error of a laser-cut package after board-level packaging is provided for some embodiments of this application;
[0024] Figure 7 A comparison diagram of defective product distribution generated by another conventional fixed-mode cutting method provided for some embodiments of this application and defective product distribution generated by the method of this embodiment;
[0025] Figure 8 An exemplary structural diagram of the electronic device provided for some embodiments of this application. Detailed Implementation
[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0027] Figure 1 An exemplary flowchart of a laser cutting path generation method for board-level packaging provided in some embodiments of this application, the method comprising:
[0028] S101. Scan the board-level packaging substrate after molding to obtain the actual position coordinate information of each die on the substrate.
[0029] Specifically, the data acquisition step is performed first. The board-level packaging substrate, after the surface mount and molding processes are completed, is placed under an automated optical inspection device to identify the positioning marks located at the corners of the substrate to establish a global coordinate system. Subsequently, the automated optical inspection device scans each die on the substrate one by one. By identifying the outer contour of the die or the via mapping on the surface, the actual position coordinate information of each die in the global coordinate system is accurately measured and output. The actual position coordinate information may include the center coordinates (X, Y) and rotation angle (θ) of the die.
[0030] S102. Calculate the node coordinates based on the actual position coordinates of the grains; wherein the node coordinates are used to determine the laser cutting paths that divide adjacent packages.
[0031] Specifically, after acquiring the actual position coordinates of the aforementioned grains, the system transmits a dataset containing the precise pose of each grain (including the horizontal coordinate X, vertical coordinate Y, and rotation angle θ) to the backend computer-aided manufacturing (CAM) processing unit or a dedicated path planning algorithm module via a high-speed data interface. In this processing unit, the system executes a dynamic node generation algorithm based on measured data. This algorithm no longer uses the preset, absolutely orthogonal fixed grid in the design drawings as a reference, but instead uses the "actual physical position" of the grain as the sole benchmark. For each laser cutting kerf intersection node to be determined, the system indexes the adjacent grain groups within the geometric neighborhood of that node in real time (e.g., grains located in the upper left, lower left, upper right, and lower right directions of the intersection point), extracts the measured edge coordinates of these grains, and accurately locates the laser cutting kerf node coordinates at that location by calculating the geometric center point or weighted average point of the gap between adjacent grains.
[0032] S103. Generate a point-to-point laser cutting path file based on the node coordinates, so as to control the laser equipment to perform cutting according to the laser cutting path file and divide the board-level packaging substrate into multiple independent packages.
[0033] Finally, the laser cutting path file is generated and the physical cutting operation is performed. In the computer-aided manufacturing system, the algorithm logically sorts a series of discrete node coordinates calculated above to accurately avoid grain misalignment according to a preset topological order (e.g., along the extension direction of the cutting path). The system uses linear interpolation technology to concatenate these discrete coordinate points to generate a laser cutting path file (e.g., commonly used Excellon, DXF, or G-Code formats) containing point-to-point connection instructions. In this file, the cutting path, which originally appears as a straight line in the macroscopic design, is microscopically analyzed as a multi-segment broken line trajectory composed of several tiny line segments connected end to end. The start and end points of each micro-segment strictly correspond to the calibration nodes determined based on the measured grain positions, thereby ensuring that the path data logically perfectly adapts to the distortion characteristics of the substrate.
[0034] Next, the laser cutting equipment reads the path file and enters the execution phase. The control system drives the laser head (or a high-precision galvanometer scanning system) to strictly follow the coordinate instructions in the file, controlling the high-energy laser beam to scan the substrate surface. The laser beam moves precisely in a straight line between each node according to the instructions, using a high-energy-density spot to physically ablate or modify the molding compound and carrier material of the packaging substrate layer by layer. Because this cutting path is dynamically generated "point-to-point," the laser beam can flexibly travel through the gaps in the die array where nonlinear distortion or random shifts occur, always remaining on the effective safety boundary line between adjacent packages. As the path is completed, the entire board-level packaging substrate is finally precisely divided into multiple physically independent individual packages, all with dimensional yields meeting tolerance requirements, thus completing the manufacturing process from the entire board array to individual finished products.
[0035] In this embodiment, the "detect first, plan later" feedforward control mechanism significantly improves the dicing yield of packaged products. Utilizing AOI measured data to drive path generation allows the laser cutting track to accurately follow the random positional offset and rotation of the die, effectively solving the problem that traditional straight-line cutting cannot adapt to the nonlinear distortion of the die array. Furthermore, leveraging the advantages of laser cutting reduces the cutting track width, increasing the effective output per unit substrate while improving yield. Figure 2As shown, by comparing the defective product distribution map on the left generated by the traditional fixed-mode cutting with the defective product distribution map on the right generated by the method of this embodiment, it can be seen that on the same board-level packaging substrate containing 31,312 chips, the traditional process cannot avoid the severe chip displacement in local areas, resulting in as many as 1,381 defective products (NGs) caused by miscutting. However, after applying the adaptive path generation method described in this embodiment, thanks to the high passability of the cutting path to narrow gaps and the precise compensation for rotational posture, most of the chips that were originally at risk of being scrapped were successfully saved as qualified products, resulting in a significant reduction in the final number of defective products to 50. This measured data strongly confirms that the present invention has a decisive technical advantage in solving the cutting problem of high-density packages and maximizing the production first-pass yield.
[0036] In one embodiment, the step of calculating the node coordinates based on the actual position coordinate information of the grain specifically includes:
[0037] The actual center coordinates of each grain are compared with the preset ideal center coordinates to obtain the offset vector of each grain.
[0038] Based on the offset vector, the standard rectangular cutting frame corresponding to the grain is translated and transformed to generate an independent laser cutting path that follows the actual position of each grain, so as to determine the node coordinates.
[0039] Specifically, the board-level packaged substrate, after completing the molding process, is placed on the worktable of an automated optical inspection (AOI) machine. The machine uses a high-resolution camera to identify positioning marks at the corners of the substrate to construct a global coordinate system, and then scans each die in the substrate array one by one. The system reads the actual center coordinates of each die in the global coordinate system, and simultaneously retrieves the ideal center coordinates of that die from the database of the computer-aided manufacturing (CAM) system, which are preset in the design drawings (such as Gerber files). By comparing the actual center coordinates of each die with the ideal center coordinates, the system can accurately calculate the positional deviation values of each die in the horizontal direction (X-axis) and the vertical direction (Y-axis), thereby obtaining the unique positional offset vector of that die.
[0040] Subsequently, a dynamic reconstruction step of the cutting path is performed. Based on the calculated offset vectors, the CAM software performs targeted geometric transformations on the laser cutting path. The system first calls a preset standard rectangular cutting frame, whose dimensions strictly correspond to the standard length and width dimensions of the package design, and whose initial position is set at an ideal coordinate point. Next, the system applies the offset vector corresponding to each die to this standard rectangular cutting frame, performing a translation transformation operation. For example, if the actual position of a die is offset 15 micrometers to the right and 5 micrometers upward, its corresponding standard cutting frame will also be simultaneously translated 15 micrometers to the right and 5 micrometers upward. Through this one-to-one independent transformation, the system generates independent laser cutting paths in the virtual layer that correspond one-to-one with hundreds or thousands of dies on the substrate and are precisely concentric with their respective actual positions.
[0041] Finally, the generated independent cutting paths are encoded into a control file (e.g., Excellon2 format) that can be recognized by the laser equipment. This ensures that each package is centered, effectively avoiding defects such as uneven edges or chip damage caused by accumulated errors in traditional straight-line cutting, thus significantly improving the yield of individual packaged products.
[0042] Furthermore, in one embodiment, the actual position coordinate information further includes the actual rotation angle of the grain; the step of calculating the node coordinates based on the actual position coordinate information of the grain specifically includes:
[0043] Calculate the angular offset between the actual rotation angle and the preset ideal rotation angle;
[0044] Obtain the original corner coordinates of the standard rectangular cutting frame relative to the grain center;
[0045] Using a coordinate transformation algorithm, based on the offset vector and the angle offset, the original corner coordinates are converted into the corrected target corner coordinates;
[0046] The corrected corner coordinates are determined as the node coordinates.
[0047] Specifically, firstly, the system uses an automated optical inspection (AOI) device to acquire the actual center coordinates (Xreal, Yreal) and actual rotation angle θreal of each grain, and compares them with the ideal design values to obtain the center translation (ΔX, ΔY) and angle offset Δθ. The system pre-defines the standard rectangular cutting path corresponding to the grain, which contains four vertices (e.g., P1, P2, P3, P4), and these vertices have original relative coordinates with respect to the grain center. To achieve follow-up rotation, the system uses a two-dimensional coordinate rotation and translation algorithm to transform and calculate the coordinates (x, y) of each original vertex.
[0048] It can be calculated using the following formula:
[0049]
[0050]
[0051] The system sequentially calculates the coordinates of the four transformed new target corner points P1′, P2′, P3′, and P4′. This calculation process actually completes both the rotation transformation around the center and the translation transformation to the actual position at the algorithm level.
[0052] Subsequently, the path file is generated and the closed-loop cutting steps are performed. The calculated coordinates of the four new target points are written into the laser cutting path file (e.g., Excellon2 format), and a closed instruction sequence is constructed. The laser control system drives the galvanometer or motor according to the file, executing the following sequence of actions: First, the laser head is rapidly moved (Jump) to the first new target point P1′; after reaching the target point, the laser is turned on, and the laser head is controlled to perform linear interpolation motion, connecting P2′, P3′, and P4′ in sequence; finally, the laser is linearly interpolated back from P4′ to P1′ to form a closed loop, and then the laser is turned off. Although the laser device only moves linearly between coordinate points at the execution level, because these four target points are precisely calculated based on the actual tilt angle of the grain, the actual trajectory cut on the substrate presents as a rectangle that rotates with the grain's posture. This ensures the straightness of the cutting line while perfectly avoiding the grain body and the effective area, achieving all-round posture tracking and compensation.
[0053] In this embodiment, a laser cutting path generation method based on independent position compensation is used to construct an independent cutting path for each die on the substrate, which follows its actual position change. This method completely breaks the limitation that adjacent products must share the cutting path in traditional linear cutting, fundamentally eliminating the impact of random die position offset on the package's appearance and dimensions. Regardless of the degree of nonlinear distortion of the die array on the substrate surface, this method can ensure that each cut package maintains extremely high alignment accuracy, effectively avoiding the risk of "large and small edge" defects or damage to the internal circuitry of the die caused by accumulated errors. It is particularly suitable for high-end board-level packaging products with extremely high requirements for unit yield and dimensional accuracy, significantly improving the production first-pass yield of high-value products.
[0054] In one embodiment, the step of calculating the node coordinates based on the actual position coordinate information of the grain specifically includes:
[0055] Identify four adjacent grains located around the intersection of laser cutting paths;
[0056] Based on the actual position coordinate information of each grain, the theoretical corner coordinate values of the intersection points of the laser cutting tracks near the four grains are obtained respectively.
[0057] The arithmetic mean of the theoretical corner coordinates is used to determine the node coordinates.
[0058] Specifically, in a computer-aided manufacturing (CAM) system, for each cross-point to be determined in the laser cutting grid, the system first uses a topology algorithm to identify the four adjacent grains around the cross-point (typically the grains at the top left, top right, bottom left, and bottom right positions). Next, based on the actual position coordinates of these four grains obtained from the AOI scan in the previous step (including the actual center coordinates X, Y and rotation angle θ), and combined with pre-stored grain physical size parameters (length and width), the system performs geometric calculations. For each grain, the system calculates the spatial coordinates of its corner point closest to the cross-point. For example, for the grain located at the top left, the actual position of its bottom right corner point in the global coordinate system is calculated; for the grain located at the bottom right, the actual position of its top left corner point is calculated. Due to the random stress introduced by the molding process, these four theoretically coincident corner points are often scattered in actual space, forming a misaligned polygonal region.
[0059] Subsequently, as Figure 3 As shown, the system extracts the coordinates of the four theoretical corner points obtained from the above calculations, and uses an arithmetic mean algorithm to calculate the average of their horizontal and vertical coordinates. This average coordinate value is then used as the final laser cutting node at the intersection. This calculation process is equivalent to finding a balance point among the four dispersed corner points, which can take into account the offset of the four chips to the greatest extent. The system traverses all intersection points on the substrate in this way, and connects these balanced nodes point-to-point in row and column order to generate a complete laser cutting path. The resulting cutting path appears microscopically as a series of connected broken lines, rather than perfectly straight lines. After reading the path file, the laser device controls the laser head to cut along this broken line path. The final geometry of the single package (PKG) may appear as a non-standard rectangular polygon with slight deformation.
[0060] The laser cutting path generation method based on adjacent node balancing provided in this embodiment has significant beneficial effects. By introducing a four-point averaging dynamic node calculation mechanism, it solves the problem of insufficient or overlapping cutting space caused by inconsistent offset directions of adjacent dies in board-level packaging (e.g., adjacent dies offset towards or away from each other). This adaptive polygonal path can automatically balance various errors, maximizing the preservation of the product's safety margin while ensuring that the cutting line does not intrude into the effective area of any die. It successfully saves products that would otherwise be easily scrapped due to local offset in traditional straight-line cutting into qualified products, thereby significantly improving the overall board yield and substrate utilization rate under shared cutting track processes.
[0061] It should be noted that the laser path generation strategy provided in the embodiments of this application can be applied as an independent solution to different production scenarios, or it can be combined and applied according to actual process requirements or substrate distortion characteristics.
[0062] In a preferred hybrid control mode, the system can preset an offset judgment threshold: during the path planning stage, the system first analyzes the degree of grain offset in a local area. When the detected grain offset is small and relatively uniformly distributed (e.g., in a linear deformation region), the adjacent node balancing method is preferentially used to generate continuous polygonal shared cutting paths, thereby maximizing cutting efficiency while ensuring yield. When the system detects severe random offset, large-scale rotation, or that the spacing between adjacent grains has been compressed to a critical value in a local area, the system switches to an independent position compensation mode to generate separate closed cutting paths for the affected grains in that area. This scheme of dynamically switching or mixing different strategies according to the actual condition of the substrate enables the present invention to achieve an optimal balance between processing efficiency and final yield, thereby adapting to a wide range of application needs from standard processes to extreme distortion processes.
[0063] In one embodiment, the step of scanning the board-level packaging substrate after molding to obtain the actual position coordinate information of each die on the substrate specifically includes:
[0064] Identify the positioning marks located at the corners of the board-level packaging substrate and establish a global coordinate system for the substrate;
[0065] Scan the outer contour or via mapping of each grain;
[0066] Based on the outer contour or the via mapping obtained from the scan, calculate the center coordinates (X, Y, θ) of each die in the global coordinate system of the substrate, and use them as the actual position coordinate information.
[0067] Specifically, such as Figure 4As shown, the physical dimensions of the board-level packaging substrate are length (Carrier L) and width (Carrier W). The vision system of the AOI equipment first searches for and identifies preset positioning marks (positioning rings shown at the four corners in the figure) located at the corners of the substrate (usually diagonal positions or four corners) in the macroscopic field of view. The system constructs a unified global coordinate system for the substrate by calculating the geometric center of these positioning marks; for example, the center of the positioning mark located in the lower left corner is selected as the origin (0,0), and the X-axis and Y-axis are established along the Carrier L direction and Carrier W direction respectively, thereby incorporating each die arrayed on the substrate (the squares marked a, b, c... in the figure) into this unified coordinate system. This step ensures that the local coordinates of all subsequent dies can be unified to an absolute physical reference, eliminating mechanical positioning errors that may occur during the transfer of the substrate itself.
[0068] Subsequently, the AOI equipment switches to a high-magnification lens and scans each die on the substrate one by one according to a preset array sequence. During the scanning process, the equipment employs specific image recognition algorithms based on the different surface features of the dies: for dies with clear edges, an edge detection algorithm is used to extract the die outline; for dies with specific circuit patterns or copper pillars on the surface, the via map is identified. The system matches the acquired image features with a standard template to calculate the precise pose of each die in the global coordinate system. Finally, the system outputs an array (X, Y, θ) containing the horizontal coordinates, vertical coordinates, and rotation angle, as the final coordinate information representing the actual physical position of the die, and transmits it to the back-end data processing unit.
[0069] In this embodiment, by establishing a global coordinate system and combining a dual recognition mode of "outer contour" or "via mapping," this method can overcome interference factors such as molding compound overflow or low contrast, achieving sub-micron level precision in capturing grain positions. In particular, incorporating the rotation angle (θ) into the output standard ensures that subsequent cutting path generation can compensate for both translation and rotation errors, providing crucial underlying data support for achieving omnidirectional adaptive laser cutting. This effectively guarantees the alignment accuracy and yield of subsequent processes.
[0070] In one embodiment, the method further includes:
[0071] Based on the offset of each die, redistribution layer pattern data matching the actual position of the die is generated, ensuring that the laser cutting path and the redistribution layer pattern data are compatible. Specifically, after obtaining the actual center coordinates (X, Y) and rotation angle (θ) of each die, this high-precision position data is not only used in subsequent cutting processes but is also prioritized for transmission to the data processing system of the photolithography process. The system dynamically adjusts the preset standard redistribution layer (RDL) pattern data based on the offset of each die relative to the ideal position. This process typically employs adaptive lithography technology (such as laser direct-write LDI), where the system modifies the mask pattern or exposure path corresponding to each die in real time, causing it to translate and rotate accordingly to follow the actual displacement of the die.
[0072] Subsequently, since the RDL pattern has undergone position compensation following the die, the "effective functional area" of the package has actually undergone physical displacement on the substrate. To prevent laser cutting from accidentally damaging these moved circuits, the laser cutting path generated in this embodiment must reuse the same set of position offset data. When generating the laser cutting path file, the system calls the die-measured coordinate data that is completely consistent with the generated RDL pattern. This means that the laser cutting line is no longer defined by a fixed mechanical grid, but is linked to the RDL pattern. The system verifies the position of the cutting street to ensure that the scanning trajectory of the laser beam is always within a preset safety gap outside the RDL wiring area, so that the final physical shape and the internal circuit pattern maintain a perfect relative positional relationship. Figure 5 The diagram visually illustrates the spatial relationship between the laser cutting path and the package (PKG) and internal redistribution layer (RDL) in this embodiment. The left side of the diagram shows the internal structure including the die (blue box) and the redistribution layer (RDL layer). The green rectangular frame on the right side of the diagram identifies the laser output path generated by the system for performing physical cutting. The dimensions of this green frame define the final physical length and width of a single package. It is worth noting that the green cutting frame for each package is independently calculated and generated based on the actual measured coordinates (X, Y, θ) of that die.
[0073] In this embodiment, by integrating die position compensation technology throughout the entire manufacturing process of circuit fabrication and outline cutting, this method achieves alignment of the internal and external structures of the package. This not only eliminates the risk of via misalignment due to die displacement, but more importantly, it establishes a robust mistake-proofing mechanism to ensure that the laser cutting path automatically avoids redistribution layers that move with the die. This interconnected control strategy significantly reduces the probability of failure due to cutting deviations causing circuit breaks or interlayer short circuits, especially in high-density fan-out board-level packages (FOPLP), where it can significantly improve the electrical reliability and final yield of the product.
[0074] In one embodiment, the step of generating a point-to-point connected laser cutting path file based on the node coordinates, and controlling the laser device to perform cutting according to the laser cutting path file to divide the board-level packaging substrate into multiple independent packages specifically includes:
[0075] Generate the laser cutting path file containing a segmented sequence of switching light commands.
[0076] Control the laser head to position it at the first node on the edge of the first package;
[0077] Turn on the laser and control the laser head to move to the second node while keeping the laser on;
[0078] The laser is turned off after reaching the second node;
[0079] The laser head is controlled to jump and move to the starting node of the next cutting path; wherein, the geometry of the single package formed by cutting based on the instruction sequence is a non-standard rectangular polygon.
[0080] Specifically, the computer-aided manufacturing (CAM) system generates a specific laser cutting path file (e.g., in Excellon2 format) based on the calculated node coordinates. This file is not a simple set of continuous lines, but contains a series of segmented on / off light command sequences. Because the aforementioned node calculations (especially the adjacent node balancing method) result in a broken line generated to avoid the grains in the cutting path, the final geometry of the single-packet (PKG) formed based on this command sequence is no longer a strict mathematical rectangle, but rather a non-standard rectangular polygon with slight deformation to adapt to grain offset.
[0081] When the laser equipment executes this command, the control system strictly follows the following instruction logic: First, the "positioning" command is executed, controlling the laser head (or galvanometer system) to move in a high-speed jump mode to the first node position on the edge of the first package. During this process, the laser remains off to avoid accidentally damaging the substrate. Second, the "on and process" command is executed. Once the laser head accurately reaches the first node, the laser beam is immediately turned on, and the laser head is controlled to move along the path to the second node at a preset processing speed. During this period, the laser continues to emit light, thereby cutting a physical slit connecting the two nodes on the substrate. Third, the "off" command is executed. Once the system detects that the laser head has reached the coordinates of the second node, the laser is immediately turned off. Finally, the "jump" command is executed, controlling the laser head to re-enter the high-speed idling state and jump to the starting node of the next cutting path (such as the edge of an adjacent package or the next edge of the same package). The system repeats this cycle until the entire board is cut.
[0082] In this embodiment, by employing a segmented control logic of "jump-cut-jump," this method perfectly adapts to the complex cutting paths caused by die offset in board-level packaging, which are either discontinuous or non-collinear. Unlike traditional cutters that must cut continuously across the entire board, this flexible on / off optical control allows the cutting path to freely fold and break in space, thus enabling the cutting of "non-standard rectangular polygonal" packages. This seemingly irregular shape is actually a deliberate process result, representing that the cutting line has successfully found the optimal path within the complex offset gaps between adjacent dies. This ensures both physical separation of the package and minimizes damage to the effective die area, thereby achieving a significant improvement in process yield.
[0083] Furthermore, in one embodiment, the method further includes:
[0084] Measure the length and width of the package after laser cutting;
[0085] Determine whether the length or the width exceeds a preset tolerance range;
[0086] If the tolerance is exceeded, the package will be rejected.
[0087] Specifically, after the laser equipment completes the segmentation of the board-level packaging substrate according to the generated path file, an integrated online vision inspection system or a separate automated optical inspection (AOI) device is used to perform high-precision geometric parameter scanning on each separated package on the substrate. The system uses image processing algorithms to extract the contour features of the package edges and accurately measure the actual physical dimensions of each package in the long and short side directions.
[0088] Subsequently, as Figure 6As shown, the system compares the measured actual length and width data with preset product specification standards in real time. This preset standard includes an allowable tolerance range (e.g., based on the process standards in the documentation, this range can be set to ±50μm of the standard size). The system performs logical judgments on each package: if the measured size of a package is within this tolerance range, the system marks it as a qualified product and records it in the final wafer map or strip map; if the length or width of a package exceeds the preset tolerance range (e.g., due to excessive die offset causing excessive compression or stretching of the package edge to avoid the die), the system immediately marks the package as a defective product. In subsequent sorting or packaging stages, the picking device reads the updated map, automatically skipping these NG-marked packages, or grabbing and discarding them to the waste area, thus performing a rejection operation.
[0089] In this embodiment, by adding a screening step after the adaptive laser cutting process, it can be ensured that the deformation caused by the previous cutting is strictly controlled within acceptable tolerances (e.g., ±50μm). This not only prevents irregularly shaped products with severely out-of-size dimensions from flowing into subsequent processes or to the customer, ensuring the consistency and reliability of shipped products, but also provides valuable data feedback for process engineers to continuously optimize upstream molding processes or laser compensation parameters.
[0090] In one embodiment, the width of the laser cutting track is smaller than the standard cutting track width using a blade wheel cutting process.
[0091] Furthermore, in one embodiment, the width of the laser cutting path corresponding to the laser cutting path is less than 250 μm.
[0092] Specifically, in computer-aided manufacturing (CAM) systems, engineers or automated algorithms no longer adhere to the wide-spacing standards required for traditional diamond wheel cutting when planning the cutting path. Traditional cutting wheels, due to their physical thickness (typically tens to hundreds of micrometers) and the mechanical vibration and chipping risks associated with high-speed rotation, often require a wide safety kerf (e.g., over 300 μm) to prevent damage to the grains. In this embodiment, however, the system utilizes the ability of a laser beam to be focused to a micrometer-level spot size, setting the physical kerf width corresponding to the generated laser cutting path to less than 250 μm. This value is calculated based on laser energy density, material ablation threshold, and beam mode, ensuring that the cutting path is sufficiently fine while still effectively cutting through the molding compound and substrate material, achieving physical separation of the package.
[0093] Based on the aforementioned narrow kerf parameters, the system reconstructs the overall layout of the board-level package for compactness. When generating the kerf path file, the system explicitly compresses the spacing between adjacent packages to a width smaller than the standard width set for wheel cutting. For example, if the same packaged product requires a 350μm kerf when using standard wheel cutting, the laser cutting path generation method in this embodiment will reduce it to 200μm or even narrower. The system controls the scanning trajectory of the laser beam, strictly confining it to this reduced narrow channel for material removal. This avoids mechanical stress transfer, thus eliminating the need for a large stress buffer zone as required by wheel cutting.
[0094] In this embodiment, by utilizing the characteristics of laser processing—small spot size and non-contact ablation—a high-precision cutting path with a physical kerf width of 250μm or even narrower can be generated. This micro-cutting path possesses extremely high spatial throughput, allowing it to precisely traverse the critical gaps that become extremely narrow due to grain offset, completing physical separation without touching the grain itself. This effectively avoids the risk of miscutting caused by insufficient cutting space, successfully transforming products that would be deemed unprocessable or scrapped due to insufficient spacing in traditional wide-blade processes into good dies. Thus, under the same grain offset conditions, it maximizes the preservation of effective grains on the substrate, significantly improving the final yield.
[0095] like Figure 7 As shown, this diagram visually demonstrates the significant effect of the synergistic effect of high-precision position compensation and narrow dicing technology described in this embodiment in high-density packaging scenarios. The left side of the diagram shows a wafer mapping diagram using a standard cutter wheel width. Under severe die misalignment, despite attempts at alignment, the inherent physical width of the cutter prevents it from navigating the narrowed gap caused by irregular die misalignment. This results in up to 1381 packages (NG: 1381 / 31312) being marked as defective due to miscutting. On the right side of the diagram, thanks to the rotation compensation path calculated by the system based on the actual die coordinates (X, Y, θ), combined with a laser spot with a physical width significantly less than 250μm (as shown in the enlarged view in the lower right corner, where purple represents the laser path), the dicing line can safely pass through the extremely narrow space adjacent to the die edge. This successfully solves the spatial interference problem caused by die misalignment, reducing the number of defective products under the same conditions to zero, achieving zero defects in the dicing process.
[0096] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this application. Adding insignificant modifications or introducing insignificant designs to the algorithm or process, but without changing the core design of the algorithm and process, are also within the scope of protection of this application.
[0097] Furthermore, some embodiments of this application also provide an electronic device. The electronic device can be various forms of digital computer, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, etc. The electronic device can also be various forms of mobile devices, such as cellular phones, smartphones, wearable devices, and other similar computing devices.
[0098] The electronic device includes: one or more processors; and a memory storing computer program instructions that, when executed, cause the processor to perform the steps of the methods provided in any one or more of the above embodiments. Figure 8 An exemplary structural diagram of the electronic device is disclosed. The electronic device includes one or more processors 1101, a memory 1102, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components are interconnected via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple electronic devices can be connected, each providing some of the necessary operations. The components, their connections and relationships, and their functions shown herein are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.
[0099] The electronic device may further include an input device 1103 and an output device 1104. The processor 1101, memory 1102, input device 1103 and output device 1104 may be connected by a bus or other means, as shown in the figure, which is connected by a bus.
[0100] Input device 1103 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the electronic device, such as a touch screen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 1104 may include a display device, auxiliary lighting device (e.g., LED), and haptic feedback device (e.g., vibration motor). The display device may include, but is not limited to, a liquid crystal display, a light-emitting diode display, and a plasma display. In some embodiments, the display device may be a touch screen.
[0101] To provide interaction with the user, the electronic device can be a computer. The computer has: a display device (e.g., a cathode ray tube or LCD monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback); and input from the user can be received in any form (e.g., voice input or tactile input).
[0102] In this embodiment, a computer-readable medium stores a computer program / instructions that, when executed by a processor, implement the steps of the methods provided in any one or more of the above embodiments. This computer-readable medium may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into that device. The aforementioned computer-readable medium carries one or more computer-readable instructions.
[0103] The memory 1102 can serve as a non-transitory computer-readable storage medium, used to store non-transitory software programs, non-transitory computer-executable programs, and modules. The processor 1101 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 1102, thereby implementing the program instructions / modules corresponding to the methods provided in any one or more of the embodiments described above in this application.
[0104] The memory 1102 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device. Furthermore, the memory 1102 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 1102 may optionally include memory remotely located relative to the processor 1101, and these remote memories can be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0105] It should be noted that the computer-readable medium described in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.
[0106] Computer-readable media include permanent and non-permanent, removable and non-removable media, which can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory, static random access memory, dynamic random access memory, other types of random access memory, read-only memory, electrically erasable programmable read-only memory, flash memory or other memory technologies, read-only optical discs, digital versatile optical discs or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.
[0107] Computer program code for performing the operations of this application can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, and C++, and conventional procedural programming languages such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including local area networks (LANs) or wide area networks (WANs), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0108] In the above embodiments, all or part of the implementation can be achieved through software, hardware, firmware, or any combination thereof. For example, it can be implemented using an application-specific integrated circuit (ASIC), a general-purpose computer, or any other similar hardware device. In some embodiments, the software program of this application can be executed by a processor to implement the above steps or functions. Similarly, the software program of this application (including related data structures) can be stored in a computer-readable recording medium, such as RAM memory, magnetic or optical drives, floppy disks, and similar devices. In addition, some steps or functions of this application can be implemented in hardware, for example, as circuitry that cooperates with a processor to perform the various steps or functions.
[0109] The computer program product provided in this application includes one or more computer programs / instructions. When executed by a processor, these computer programs / instructions generate, in whole or in part, the processes or functions described in this application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.
[0110] The flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-specific system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0111] The scope of this application is defined by the appended claims rather than the foregoing description, and is therefore intended to encompass all variations falling within the meaning and scope of equivalents of the claims. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a device claim may also be implemented by a single unit or device in software or hardware. Terms such as "first," "second," etc., are used only for distinguishing descriptions and do not indicate any particular order, nor should they be construed as indicating or implying relative importance.
[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily made by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims, and the above embodiments should be regarded as exemplary and non-limiting.
Claims
1. A method for generating laser cutting paths for board-level packaging, characterized in that, The method includes: Scan the board-level packaging substrate after molding to obtain the actual position coordinate information of each die on the substrate; Based on the actual position coordinates of the grains, the node coordinates are calculated; wherein, the node coordinates are used to determine the laser cutting paths that divide adjacent packages. A point-to-point laser cutting path file is generated based on the node coordinates to control the laser equipment to perform cutting according to the laser cutting path file, thereby dividing the board-level packaging substrate into multiple independent packages.
2. The laser cutting path generation method according to claim 1, characterized in that, The step of calculating the node coordinates based on the actual position coordinate information of the grain specifically includes: The actual center coordinates of each grain are compared with the preset ideal center coordinates to obtain the offset vector of each grain. Based on the offset vector, the standard rectangular cutting frame corresponding to the grain is translated and transformed to generate an independent laser cutting path that follows the actual position of each grain, so as to determine the node coordinates.
3. The laser cutting path generation method according to claim 1, characterized in that, The step of calculating the node coordinates based on the actual position coordinate information of the grains specifically includes: Identify four adjacent grains located around the intersection of laser cutting paths; Based on the actual position coordinate information of each grain, the theoretical corner coordinate values of the intersection points of the laser cutting tracks near the four grains are obtained respectively. The arithmetic mean of the theoretical corner coordinates is used to determine the node coordinates.
4. The laser cutting path generation method according to claim 1, characterized in that, The step of scanning the board-level packaging substrate after molding to obtain the actual position coordinate information of each die on the substrate specifically includes: Identify the positioning marks located at the corners of the board-level packaging substrate and establish a global coordinate system for the substrate; Scan the outer contour or via mapping of each grain; Based on the outer contour or the via mapping obtained from the scan, calculate the center coordinates (X, Y, θ) of each die in the global coordinate system of the substrate, and use them as the actual position coordinate information.
5. The laser cutting path generation method according to claim 1, characterized in that, The method further includes: Based on the offset of each die, redistribution layer pattern data that matches the actual position of the die is generated, so that the laser cutting path and the redistribution layer pattern data are mutually adapted.
6. The laser cutting path generation method for board-level packaging according to claim 1, characterized in that, The step of generating a point-to-point connected laser cutting path file based on the node coordinates, and controlling the laser equipment to perform cutting according to the laser cutting path file to divide the board-level packaging substrate into multiple independent packages specifically includes: Generate the laser cutting path file containing a segmented sequence of switching light commands, so that the laser device performs the following operations: Control the laser head to position it at the first node on the edge of the first package; Turn on the laser and control the laser head to move to the second node while keeping the laser on; The laser is turned off after reaching the second node; The laser head is controlled to jump and move to the starting node of the next cutting path; wherein, the geometry of the single package formed by cutting based on the instruction sequence is a non-standard rectangular polygon.
7. The laser cutting path generation method for board-level packaging according to claim 1, characterized in that, The width of the laser cutting track is smaller than the width of the standard cutting track using a blade wheel cutting process.
8. An electronic device, characterized in that, The electronic device includes: One or more processors; and A memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method as described in any one of claims 1 to 7.
9. A computer-readable medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.
10. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.