A semi-automatic PIN calibration method for MEMS probe cards

Abstract

This application relates to the field of semiconductor testing technology and discloses a semi-automatic PIN calibration method for a MEMS probe card, comprising the following steps: securing the probe card on a fixed platform, adjusting the focal length and magnification of a large tool microscope through an operating interface so that the probe is displayed in an image acquisition system; activating an image acquisition and processing system to capture initial images of the probe and a target PIN point, and extracting the position and size parameters of the probe; the apparatus includes: a large tool microscope for performing high-resolution imaging of the probe and observing the relationship between the probe and the target PIN point; and a probe card fixing platform for fixing the probe card. The present invention utilizes a technical solution based on high-resolution image acquisition and real-time feedback adjustment to achieve precise alignment between the probe card and the target PIN point. Through real-time image feedback and precise path optimization, the probe card is ensured to be precisely aligned at every step during the calibration process.

Description

Technical Field

[0001] The present invention relates to the technical field of semiconductor testing, and in particular to a semi-automatic PIN correction method for a MEMS probe card. Background Art

[0002] In the field of semiconductor testing technology, the precision of the probe card directly determines the accuracy of the test. However, existing technologies face multiple problems during the calibration process, mainly reflected in manual reliance and inefficient path planning.

[0003] First, existing techniques typically rely on manual adjustments for probe card calibration, a time-consuming and prone to human error. While some devices utilize high-resolution microscopes for imaging, the lack of effective automated calibration mechanisms often requires operators to manually correct the deviation between the probe and the target PIN point. This not only increases operational complexity but also makes it difficult to ensure consistent accuracy across adjustments. This manual approach is particularly difficult to meet the high-precision and high-efficiency requirements of mass production.

[0004] Secondly, traditional path planning methods often lack comprehensive consideration of path optimization. In existing technologies, path planning is mostly based on simple geometric calculations or preset paths. This leads to lengthy and imprecise path adjustments, which often wastes significant time and energy, especially when frequent and precise adjustments are required. Furthermore, existing path planning methods often ignore physical constraints such as acceleration and velocity, making global optimization impossible. This results in low path efficiency and may even cause unnecessary vibration or interference, affecting calibration accuracy.

[0005] Furthermore, while some existing technologies have achieved some degree of optimization in path adjustment, analysis and verification of path stability are often lacking. Path instability can cause the probe to lose its stable alignment with the target PIN point during multiple adjustments, or even repeatedly deviate from the target. Existing technologies rarely have effective theoretical support to ensure stability during path adjustment. Path stability is crucial, yet often overlooked, especially in high-speed or high-precision applications.

[0006] Therefore, the shortcomings of existing technologies are mainly manifested in: heavy reliance on manual labor, low efficiency in path optimization, and insufficient verification of path stability.

[0007] Therefore, the present invention proposes a semi-automatic PIN calibration method for a MEMS probe card to address the deficiencies of the prior art. Summary of the Invention

[0008] In view of the shortcomings of the existing technology, the present invention provides a semi-automatic PIN correction method for a MEMS probe card, which solves the problems of heavy manual dependence, low path optimization efficiency, and insufficient path stability verification.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: A semi-automatic PIN calibration method for a MEMS probe card, comprising the following steps:

[0010] Fix the probe card on a fixed platform, and adjust the focus and magnification of the large tool microscope through the operation interface so that the probe is displayed on the image acquisition system;

[0011] Start the image acquisition and processing system, collect the initial images of the probe and the target PIN point, and extract the position and size parameters of the probe;

[0012] According to the preset calibration standard, the motion control system automatically adjusts the position of the probe card so that the probe is close to the standard position;

[0013] The user can make fine adjustments through the operation interface to adjust the position accuracy of the probe;

[0014] Repeat the above steps to calibrate multiple probes on the probe card in sequence;

[0015] After calibration is completed, the image acquisition and processing system is used to capture the final image of the probe and the target PIN point, and compared and analyzed with the image before calibration. At the same time, the probe card path is adjusted through real-time image feedback to ensure that the probe card is accurately aligned with the target PIN point and the probe card movement path is optimized;

[0016] The probe card movement path optimization is performed by combining optimal control theory with particle swarm optimization algorithm;

[0017] The stability of the probe card's moving path is analyzed using Lyapunov stability theory, and the path is further adjusted to verify the path optimization effect, ensuring that the probe card is always accurately aligned with the target PIN point.

[0018] The present invention also provides a semi-automatic PIN calibration device for a MEMS probe card, comprising:

[0019] A large tool microscope for high-resolution imaging of the probe and observing the relationship between the probe and the target PIN point;

[0020] A probe card fixing platform, used for fixing the probe card;

[0021] Motion control system for precise control of the probe card's position, including X, Y, and Z axis motion;

[0022] Image acquisition and processing system, used to collect microscope images in real time and calculate the deviation between the probe and the target PIN point;

[0023] Operation interface, used to display the image and deviation information of the probe and target PIN point, and receive the user's fine-tuning operation;

[0024] The control system is used to coordinate the work of the above components, adjust the probe card to the target PIN point position according to the image analysis results, and perform path planning and optimization through the path optimization module.

[0025] The present invention provides a semi-automatic PIN calibration method for a MEMS probe card. It has the following beneficial effects:

[0026] 1. This invention utilizes a technical solution based on high-resolution image acquisition and real-time feedback adjustment to achieve precise alignment between the probe card and the target PIN point. Real-time image feedback and precise path optimization ensure that the probe card is precisely positioned at every step during the calibration process. Compared to existing solutions that rely solely on manual fine-tuning, this invention significantly reduces human error, improves calibration accuracy, and ensures the high-precision requirements of semiconductor testing.

[0027] 2. The present invention combines optimal control theory with the particle swarm optimization algorithm to intelligently optimize the path of the probe card, thereby greatly reducing the path time and energy consumption during the calibration process. Compared with traditional path planning methods, the particle swarm optimization algorithm can more effectively find the global optimal path, reduce unnecessary path adjustments, optimize the efficiency of the entire calibration process, reduce operational complexity, and improve work efficiency.

[0028] 3. The present invention introduces Lyapunov stability theory to analyze and verify the stability of the probe card's moving path. Compared with traditional methods that only rely on intuition and simple algorithms, the present invention can more accurately ensure the stability of the path optimization process, effectively avoiding the deviation problem caused by path instability during the probe card calibration process, and ensuring that the probe card is always stably and accurately aligned with the target PIN point after each path adjustment.

[0029] 4. The present invention provides a highly adaptable calibration device that can support probe cards of different specifications and models. By adjusting the settings of the fixed platform and the microscope, the system can flexibly adapt to different types of probe cards. Compared with the dedicated equipment in the prior art, the present invention solves the problem that different models of probe cards cannot be flexibly adapted, improves the versatility and application range of the device, and meets diverse calibration needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a flow chart of the method of the present invention;

[0031] Figure 2 Schematic diagram of the device structure of the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] See also Figure 1 , an embodiment of the present invention provides a semi-automatic PIN calibration method for a MEMS probe card, comprising the following steps:

[0034] S1. Fix the probe card on the fixed platform and adjust the focus and magnification of the large tool microscope through the operation interface so that the probe is displayed on the image acquisition system;

[0035] S2. Start the image acquisition and processing system, collect the initial images of the probe and the target PIN point, and extract the position and size parameters of the probe;

[0036] S3. According to the preset calibration standard, the motion control system automatically adjusts the position of the probe card so that the probe is close to the standard position;

[0037] S4, the user makes fine adjustments through the operation interface to adjust the position accuracy of the probe;

[0038] S5. Repeat the above steps to calibrate multiple probes on the probe card in sequence;

[0039] S6. After calibration is completed, the image acquisition and processing system is used to capture the final image of the probe and the target PIN point, and compared and analyzed with the image before calibration. At the same time, the probe card path is adjusted through real-time image feedback to ensure that the probe card is accurately aligned with the target PIN point and the probe card movement path is optimized;

[0040] S7, the path optimization of the probe card movement is performed by combining optimal control theory with particle swarm optimization algorithm;

[0041] S8. Analyze the stability of the probe card's moving path using Lyapunov stability theory, and further adjust the path to verify the path optimization effect, so that the probe card is always accurately aligned with the target PIN point.

[0042] In step S1, the initial step of the semi-automatic PIN calibration method for MEMS probe cards involves securing the probe card to be calibrated on a dedicated mounting platform. The focus and magnification of the large-scale tool microscope are adjusted through the user interface to display the probe on the image acquisition system. This step initiates the entire calibration process and ensures that subsequent calibration work can be performed accurately and efficiently.

[0043] In this embodiment, the probe card is secured to a dedicated fixed platform with multiple vacuum holes to prevent the probe card from moving during calibration. The porous base plate allows the probe card to be securely attached to the platform via a vacuum mechanism, ensuring stability and preventing calibration errors caused by probe card movement.

[0044] As an option, the mounting platform features adjustable positioning mechanisms to accommodate probe cards of varying sizes or shapes. The platform's inner diameter is matched to the diameter of the probe card substrate, and a stepped structure allows for precise positioning and fixation, ensuring that the probe card's edge aligns perfectly with the platform's positioning mechanisms. During the mounting process, the probe card substrate is suctioned and securely held to the platform by a vacuum mechanism, eliminating any potential probe position deviation.

[0045] During this step, the large tool microscope is adjusted to the appropriate focus and magnification to clearly display the probe card details. Through the user interface, the user can control the microscope's focus and magnification, ensuring that the image acquisition system can accurately capture the relative position between the probe and the target PIN point.

[0046] Specifically, the purpose of adjusting the microscope's focus and magnification is to ensure that the probe's details are clearly visible, facilitating subsequent image acquisition and deviation calculation. Generally, the adjusted image should have sufficient resolution to accurately capture the relative position of the probe and target PIN point, providing clear data support for subsequent calibration steps.

[0047] In one possible implementation, the image acquisition system analyzes the probe and target PIN point using image processing algorithms (such as edge detection or contour recognition). This image data serves as the basis for subsequent steps to accurately determine the deviation between the probe and the target PIN point through calculation, providing a basis for path optimization and adjustment.

[0048] During this step, the system captures images in real time, providing essential information for subsequent path optimization, deviation calculation, and precision adjustment. This process is crucial to the entire semi-automatic calibration process, as it determines the accuracy and precision of subsequent operations. The system makes adjustments based on real-time image feedback, ensuring that each stage of calibration meets the predetermined accuracy requirements.

[0049] To improve the system's adaptability and stability, the system can be adaptively adjusted according to different probe types and target PIN point requirements. Through the operation interface, users can flexibly adjust microscope parameters and set different magnifications to further improve accuracy and ensure image data quality.

[0050] At this stage, the system provides an interactive interface where users can adjust the probe's focus and magnification to optimize microscope observation of the probe. Real-time image data displayed within the interface helps users quickly assess image clarity and magnification, enabling timely adjustments to ensure smooth progress in subsequent steps.

[0051] In summary, step S1 ensures the stability and accuracy of the entire calibration process. By fixing the probe card and precisely adjusting the focus and magnification of the microscope, it lays a solid foundation for subsequent image acquisition, deviation calculation, and path optimization.

[0052] During this process, the accuracy of the probe card's fixation and microscope adjustment are crucial to the effectiveness of the calibration process. The design of the fixation platform, particularly its vacuum suction function and positioning mechanism, ensures the precise positioning of the probe card. In practice, the platform can be designed with multiple adjustable holes or fixtures to accommodate different probe card models.

[0053] Adjusting the microscope's focus and magnification requires real-time adjustments to parameters such as image resolution and frame rate. The image acquisition system must have high resolution and fast response capabilities to ensure that the relative position of the probe and target PIN point is clearly displayed and accurately recorded during fine-tuning.

[0054] By displaying real-time images on the operating interface, the system can provide feedback on the deviation between the probe and the target, assisting the operator in adjusting the position of the microscope or platform, and ensuring that the system can operate accurately during dynamic processes.

[0055] During actual operation, the system will automatically adjust the magnification and focus settings according to different calibration requirements to adapt to the calibration requirements of different probes and targets.

[0056] Step S2 is crucial to the entire calibration process because it provides data for accurately measuring the deviation between the probe and the target PIN point through high-resolution image acquisition, ensuring the accuracy of the subsequent calibration process.

[0057] Specifically, in this example, the image acquisition and processing system is first activated. Using a large-scale tool microscope, the system captures real-time images of the area between the probe and the target PIN point. The primary purpose of this image capture is to extract the relative position, size, and other relevant parameters of the two points. Key parameters of the microscope imaging system, such as resolution, frame rate, and exposure time, are carefully tuned to ensure the clarity and accuracy of the image data.

[0058] Typically, image acquisition systems process images using edge detection or contour recognition algorithms to identify the boundary between the probe and the target PIN point. These identified boundary features are then used to calculate the deviation between the probe and the target PIN point. Alternatively, image acquisition and processing systems can be equipped with a variety of image processing algorithms to enhance image quality through filtering, denoising, and other methods.

[0059] Specifically, the image acquisition and processing system's workflow involves first capturing an initial image of the probe and the target PIN point. Next, algorithms are used to perform edge recognition and grayscale processing on the image. Finally, the deviation between the probe's position and the target PIN point is calculated. These deviations include positional deviation (e.g., deviation in the X and Y coordinate directions) and angular deviation (i.e., the angular error between the probe and the target).

[0060] In one possible implementation, the formula for calculating the deviation is:

[0061] ;

[0062] ;

[0063] ;

[0064] in, and Indicates the position deviation between the probe and the target PIN point in the X and Y directions; Indicates the angular deviation between the two; 、 are the coordinates of the target PIN point; 、 are the coordinates of the probes on the probe card, and and are the angles of the target PIN point and the probe, respectively.

[0065] In some embodiments, the system may also compare the deviation value with a tolerance standard to further verify that the relative position of the probe and the target PIN point meets the required accuracy. In this case, the system outputs the calibration results and provides a basis for the next step (such as path optimization or fine-tuning). The accuracy of the deviation calculation directly affects the accuracy of subsequent steps, so this step requires sufficient computational accuracy and efficient image processing capabilities.

[0066] During image acquisition, the large-scale tool microscope used in this example provides high-resolution imaging, enabling each image to be captured with micron-level clarity. This is crucial for detecting minute deviations, ensuring that the deviation between the probe and the target PIN point can be accurately calculated and reported. The collaborative work between the microscope and the image acquisition system is key to the efficient calibration of this method.

[0067] During image processing, the system automatically optimizes images based on pre-defined algorithms, including but not limited to image enhancement, contrast adjustment, and noise removal, to improve image quality and computational accuracy. Supported by the image acquisition and processing system, image deviation calculations can achieve high accuracy, meeting the stringent precision requirements of the semiconductor industry.

[0068] This step provides critical data for subsequent motion control and path optimization. By accurately tracking the deviation between the feedback probe and the target PIN point, the system can accurately calculate the required adjustments for path optimization, ensuring that subsequent steps are executed accurately.

[0069] The core of this step is to provide the foundational data for the subsequent calibration process through high-resolution image acquisition and precise deviation calculation. The application of image processing algorithms improves image clarity and accuracy, enabling more precise deviation calculations between the probe and the target PIN point. This accuracy is crucial for subsequent path optimization and probe positioning, ensuring the efficiency and accuracy of the calibration process.

[0070] In step S3, the image acquisition and deviation calculation steps in step S2 calculate the deviation between the probe and the target PIN point, ensuring that the probe position and angle errors are clearly defined. Based on this deviation information, step S3 controls the motion system to automatically adjust the probe card position, gradually guiding the probe to the predetermined standard position.

[0071] Specifically, in this embodiment, the motion control system initiates precise motion adjustments based on the aforementioned deviation calculation results. The system first calculates the deviation between the probe and the target PIN point and uses this data to generate a target position. Generally, the deviation between the probe and the target PIN point may include errors in the X, Y, and Z coordinate axes and angles. Alternatively, the system automatically adjusts the probe card's position using a multi-axis motion platform (including the X, Y, and Z axes) to ensure the probe remains close to the standard position.

[0072] In one possible implementation, it is assumed that the initial deviations between the probe and the target PIN point are , and , the system will generate a preliminary path adjustment instruction based on these calculated deviations. The goal of this path adjustment is to minimize the deviation between the probe and the target PIN point. The adjustment strategy can be described using the following formula:

[0073] ;

[0074] ;

[0075] ;

[0076] in, , , Respectively represent the actual deviation between the current probe and the target PIN point in the X, Y, and angle directions. , , is the standard deviation. The motion control system calculates these deviations in real time and gradually guides the probe card to the target PIN point for precise alignment.

[0077] In actual operation, the motion control system executes control commands based on these calculated deviations. The control system activates the precise motion stages on the X, Y, and Z axes, gradually moving the probe card toward the target position. Depending on the control algorithm, the system can automatically execute complex motions such as translation and rotation to bring the probe card closer to the target position, ensuring target accuracy.

[0078] Additionally, as a possible option, the motion control system can be corrected using different calibration standards during path adjustment, such as adjusting the target position based on a standard angle range or substrate alignment deviation tolerance to accommodate different types of probe cards.

[0079] During implementation, the motion control system continuously optimizes the probe card's precise positioning through continuous feedback adjustments. Real-time image feedback combined with position sensors ensures the probe card reaches the desired target position after each adjustment, with positional and angular errors within acceptable tolerances.

[0080] In its specific implementation, the system not only controls translation along the X, Y, and Z axes but also enables fine angular adjustment of the probe card. Motion control systems typically incorporate multiple sensors to ensure precise positioning of the probe card in three-dimensional space. This multi-dimensional control system allows the probe card to adaptively adjust based on real-time feedback, optimizing its motion path and accuracy.

[0081] For example, a motion control system can combine accelerometers and gyroscopes to provide precise angular adjustment, ensuring that the probe card's angular adjustments also meet precision requirements. During the adjustment process, the motion system analyzes real-time data to promptly correct the probe card's movement path to avoid path deviation.

[0082] The key to this step is ensuring that the probe card can quickly and accurately move to the target position based on the image acquisition and deviation calculation results through a precise motion control system. A real-time feedback system continuously adjusts the probe card's position to ensure precise alignment with the target PIN point. The precision of motion control and adjustment directly impacts the accuracy of subsequent calibration steps. Therefore, the precise control mechanism provided in this embodiment ensures the efficiency and accuracy of the entire calibration process.

[0083] Furthermore, precise motion control enables efficient probe card calibration and adapts to the requirements of different probes. This precise control minimizes errors throughout the calibration process, further improving the stability and reliability of the entire system and ensuring high-precision calibration results.

[0084] The core goal of step S4 is to further refine the probe position to ensure that the probe accuracy meets the predetermined standard. Since the probe card has been moved close to the target position in step S3, step S4 uses the user interface to make fine adjustments to correct the slight error between the probe and the target PIN point.

[0085] Specifically, in this embodiment, the user interface allows users to fine-tune the probe position. The image acquisition and processing system displays the error between the probe and the target PIN point in real time, allowing the user to manually adjust the probe position based on this information. The image acquisition system continuously acquires image data between the probe and the target PIN point and calculates the deviation. This deviation information is fed back to the user interface in the form of images and data, enabling users to perform precise operations.

[0086] Typically, fine-tuning involves finely adjusting the probe's position along the X and Y axes, as well as the angles. During this stage, the system provides real-time feedback to ensure the effectiveness of each fine-tuning step, preventing accuracy issues caused by over-adjustment or misoperation. Optionally, the system automatically limits the amount of fine-tuning to avoid unnecessary errors caused by excessive adjustments.

[0087] Specifically, during the fine-tuning process, the system sets a maximum allowable deviation value and gradually reduces the deviation between the probe and the target PIN point based on real-time image data. This fine-tuning process utilizes a sophisticated motion control system and user interface to ensure precise control of each adjustment step, achieving ideal alignment accuracy. If the deviation value exceeds the set tolerance during fine-tuning, the system automatically issues a warning signal, prompting the user to make timely adjustments.

[0088] For example, suppose the initial calculated deviation between the probe and the target PIN point in the X direction is , the deviation in the Y direction is , the angular deviation is During the fine-tuning phase, the system will provide adjustment suggestions based on these deviation information. Assume that after fine-tuning, the position of the probe is changed to , and , the system will update these data in real time and continue to adjust until the deviation between the probe and the target PIN point meets the accuracy requirements.

[0089] In a practical way, the system can set a final accuracy standard, such as the deviation value in the X and Y directions is , the allowable range in the angular direction is If the deviation value exceeds this range after adjustment, the system will continue to make fine adjustments until the deviation between the probe and the target PIN point meets the standard.

[0090] The key to the fine-tuning process lies in precisely controlling the probe card's motion. To achieve this, the system typically incorporates high-precision X- and Y-axis motion stages and angle control devices, using real-time image data to guide the fine-tuning process. Using these control devices, the system can precisely adjust the probe card's position in every dimension and make corresponding corrections based on deviation data.

[0091] For example, the system might use stepper motors or servo motors to control the tiny movements of the X, Y, and Z axes, ensuring that each adjustment is made with minimal effort, minimizing the impact of each fine-tuning. This process, through a combination of feedback control and real-time data analysis, ensures a gradual improvement in probe accuracy.

[0092] During the implementation process, the user interface typically provides real-time feedback, displaying the precise distance and angle between the probe and the target, helping the operator better judge the extent of the adjustment. The continuous operation of the image acquisition and processing system ensures that the actual effect of each fine-tuning is immediately verified, avoiding any mistakes during the fine-tuning process.

[0093] The core goal of this step is to gradually reduce the deviation between the probe and the target PIN point through careful user manipulation, ultimately achieving precise alignment. Real-time image feedback and precise motion control enable the user to efficiently fine-tune the probe to ensure accurate positioning. This fine-tuning step extends the calibration process beyond coarse adjustments to meticulously optimize the probe's positioning accuracy, thereby improving the accuracy and stability of the entire semi-automatic calibration process.

[0094] By performing fine-tuning under strict precision requirements, the present invention can significantly improve the correction accuracy, especially in multi-probe cards and complex test environments. The application of this step effectively ensures the accurate alignment of each probe, thereby improving the reliability of the test data.

[0095] In addition, the fine-tuning process enables the system to be adaptively adjusted according to different probe types and target requirements, thereby further enhancing the flexibility and applicability of the present invention and ensuring a wide range of engineering application possibilities.

[0096] Step S5 involves repeating the calibration process for multiple probes on the probe card. This step follows step S4, in which the user fine-tunes a probe through the user interface to ensure that its accuracy meets the predetermined standard. Therefore, the key to step S5 is to effectively extend the calibration process to every probe on the probe card, ensuring that all probes meet the same accuracy requirements.

[0097] When calibrating multiple probes on a probe card, the system automatically performs the same calibration steps for each probe, following the same calibration methods as those performed in the previous steps (e.g., steps S3 and S4). Specifically, the image acquisition and processing system first captures an initial image of each probe relative to the target PIN point. The system then calculates the deviation between each probe and the target PIN point. Using this deviation data, the system automatically adjusts the probe card's position, gradually aligning each probe with the predetermined standard position.

[0098] Alternatively, the system may perform a separate calibration step for each probe to ensure precise adjustment even when multiple probes on the probe card have different initial position deviations. In this case, the motion control system generates path adjustment instructions for each probe, adjusting the position of the probe card in turn based on the deviation data. As each probe is adjusted, the image acquisition and processing system continuously calculates the deviation between each probe and the target PIN point and dynamically adjusts the path based on the calculation results.

[0099] Specifically, for each probe, the system combines optimal control theory with real-time image feedback to ensure that the precision error between the probe and the target PIN point is gradually reduced until the predetermined accuracy requirement is achieved. Image feedback from the image acquisition system plays a crucial role in this process, providing real-time data support for each path adjustment and promptly correcting the probe position based on the feedback results.

[0100] In some embodiments, if probes on multiple probe cards have different sizes or structural features, the system can automatically adjust the calibration standards based on the characteristics of each probe. These standards include, but are not limited to, a standard distance between the probe and the target PIN point, a standard angle range, and alignment tolerances for the probe substrate. In this way, the system can flexibly adapt to different probe types and ensure that each probe is accurately calibrated.

[0101] Through this series of steps, multiple probes are calibrated according to the same high-precision standard, ensuring that after the entire probe card is calibrated, all probes can operate within the same accuracy range. This not only improves calibration efficiency, but also ensures the accuracy and consistency of the entire calibration process.

[0102] In summary, the implementation of step S5 ensures the calibration accuracy of each probe through automated path adjustment and precise fine-tuning process, thereby improving the stability and reliability of the entire system.

[0103] Step S6, following step S5, further ensures that the probe card is accurately aligned with the target PIN point. This step begins with the image acquisition and processing system capturing the final image of the probe and target PIN point. This image is then compared and analyzed with the pre-calibration image to assess the calibration results. Image comparison and analysis is a key step in verifying calibration accuracy, clearly displaying the deviation between the probe and target PIN point, providing intuitive evidence of whether the calibration has met the predetermined standards.

[0104] Specifically, the image acquisition and processing system first captures the final image of the probe and the target PIN point. During image acquisition, the system ensures that the image resolution and exposure time are optimally set to ensure accurate image data. During this process, microscope parameters such as magnification and focal length are continuously adjusted to maintain high image quality.

[0105] The system then compares and analyzes the captured image with a previously calibrated image. This comparison allows the system to calculate the difference between the probe and the target PIN point, typically including positional and angular deviations. This deviation data is displayed in real time within the system and further refined using image processing algorithms (such as image registration and edge detection).

[0106] Typically, image comparison analysis is used to determine the success of calibration. If the image comparison reveals significant deviations, the system uses a feedback mechanism to optimize the appropriate path. During this process, image data not only provides the system with a basis for verifying calibration accuracy but also provides a critical reference for subsequent path adjustments and optimizations.

[0107] As an option, the image acquisition system can apply advanced image processing algorithms, such as filtering, noise reduction, and grayscale enhancement, to further improve image clarity and data accuracy. The processed images accurately display the relative position of the probe and target PIN point, providing higher-quality data support for subsequent path adjustments.

[0108] In the second half of step S6, the system adjusts the probe card's path using real-time image feedback. This process aims to ensure precise alignment of the probe card with the target PIN point by dynamically adjusting the path. Specifically, the system calculates the deviation between the probe card's current position and the target PIN point based on the real-time image feedback and adjusts the probe card's position using a precise motion control system. This feedback system continuously optimizes the probe card's position until precise alignment is achieved.

[0109] Furthermore, path optimization isn't limited to position adjustment; it also optimizes the entire path based on feedback. Based on real-time data, the system dynamically adjusts the probe card's path to ensure it reaches the target location in the shortest possible time while minimizing energy consumption. Using path optimization algorithms (such as optimal control theory combined with particle swarm optimization), the system effectively adjusts the probe card's path, reduces unnecessary path deviations, and ensures that the probe card consistently reaches the target pin point.

[0110] In some embodiments, during path adjustment, the system may consider physical constraints such as acceleration and velocity to prevent alignment instability caused by drastic probe card motion. These constraints are automatically adjusted through optimization algorithms to ensure smooth and accurate path adjustment.

[0111] In summary, step S6 ensures precise alignment of the probes with the target PIN points. Image comparison analysis and real-time path optimization further enhance calibration accuracy and system stability. This series of precise operations effectively improves the performance of the probe card throughout the calibration process, ensuring that each probe performs accurately and stably.

[0112] Step S7 focuses on optimizing and adjusting the probe card path. Specifically, this process combines optimal control theory with a particle swarm optimization algorithm. The goal of this process is to precisely control the probe card's path from its current position to the target PIN point, ensuring the path is as short and efficient as possible while avoiding excessive errors.

[0113] Generally, the goal of path optimization is to minimize the time or energy required for a route while maintaining sufficient accuracy. Optimal control theory provides a mathematical framework for describing path optimization problems. This theory enables the system to calculate the most appropriate control strategy based on its current and target states. Particle swarm optimization (PSO) builds on this foundation, combining global search capabilities to further optimize path planning.

[0114] Specifically, the particle swarm optimization algorithm searches the path space by simulating the behavior of a swarm of particles. These particles are iteratively updated in the path space according to a fitness function to find the optimal path. The fitness function typically measures the quality of a path based on objectives such as path length, energy consumption, or time requirement. In some embodiments, the system uses multiple particles, each representing a possible path, and selects the optimal path by updating the particle's speed and position. During the iterative process, the particles continuously optimize their paths to find the global optimal solution.

[0115] During the optimization process, the control algorithm and particle swarm optimization complement each other. The control algorithm ensures that the probe card's path matches the motion requirements, while the particle swarm optimization algorithm further optimizes the path through a global search, avoiding local optimal solutions. For example, the particle swarm optimization algorithm can find a path that not only minimizes path length but also complies with the system's physical constraints, such as acceleration and speed limits.

[0116] In one possible implementation, the particle swarm optimization algorithm uses the following steps:

[0117] Initialize the particle swarm and set the particle position and velocity. The particle position represents the possible path.

[0118] Calculate the fitness value of each particle. The fitness function may include the length of the path, the total time or the total energy consumption;

[0119] Update the velocity and position of the particles, and update the position of the particles based on the historical optimal position of each particle and the optimal position in the group;

[0120] This process is repeated until convergence or the maximum number of iterations is reached, and the global optimal path is finally selected.

[0121] As an option, the particle swarm optimization algorithm can be combined with other local optimization algorithms (such as gradient descent method, genetic algorithm, etc.) to enhance the efficiency and accuracy of path optimization.

[0122] Specifically, during path optimization, the system must not only consider the shortest path but also the physical constraints along the path, such as acceleration and speed. For a probe card, its motion path must not only reach the target but also meet the device's motion constraints to avoid vibration or unstable position of the probe due to excessive speed or acceleration.

[0123] In some embodiments, the system defines an optimization objective function for the path based on optimal control theory. This objective function usually includes the time, energy, and accuracy requirements of the path. Assume that the objective function of the path is , which can be expressed as:

[0124] ;

[0125] in, To optimize the objective function, it is usually used to minimize the path time, energy consumption or other performance indicators; The time period for path optimization, which represents the total time from the start to the end of the path; is the objective function, which describes the performance at each moment on the path. It is the time Position at the moment and control inputs Function, control input It can be physical quantities such as acceleration and velocity; is the system state, which in this case represents the probe card at time Position at the moment; The control input is usually an external control variable that affects the system state. In this application, it may be a physical quantity such as velocity or acceleration.

[0126] During the optimization process, the particle swarm optimization algorithm optimizes the path by adjusting the control inputs (such as speed, acceleration) while satisfying physical constraints such as maximum speed. and maximum acceleration , ensuring the stability and feasibility of the path.

[0127] In summary, this step effectively optimizes the probe card path by combining optimal control theory and particle swarm optimization algorithm, thereby ensuring that it moves to the target PIN point accurately and efficiently. The path optimization process takes into account physical constraints and the global optimal solution, providing a more accurate and stable control strategy for probe card calibration.

[0128] In step S8, the stability of the probe card's path is analyzed using Lyapunov stability theory, and the path is further adjusted to verify the effectiveness of the path optimization. This process is crucial to ensuring that the probe card is always accurately aligned with the target PIN point, as the path stability directly affects the probe card's positioning accuracy, which in turn determines the reliability of the entire calibration process.

[0129] In the previous steps, the probe card path has been initially optimized using optimal control theory and the particle swarm optimization algorithm (step S7). However, path stability remains a key factor requiring further verification. To ensure stability during the path optimization process, Lyapunov stability theory is employed. The core of Lyapunov stability theory is to construct a Lyapunov function to determine whether the path will become unstable during optimization, allowing for the necessary adjustments.

[0130] Specifically, Lyapunov stability theory analysis is carried out through the following steps:

[0131] First, based on the optimized path, construct the Lyapunov function. The Lyapunov function is usually a positive definite function used to describe the energy or error measurement of the system. Set the Lyapunov function As system status function, where Represents the deviation between the probe card and the target PIN point. Specifically, the Lyapunov function can be constructed as follows:

[0132] ;

[0133] in, To represent the Lyapunov function, which is usually used to measure the state of the system Energy or error measure of ; The state vector representing the system, usually the deviation between the probe card and the target PIN point, contains the position information of the system; Represents the state vector The transpose of , i.e. the row vector form; Is a symmetric positive definite matrix, usually used to measure the error of the system state; To ensure that the Lyapunov function satisfies the conditions, the coefficients are usually introduced. To simplify the derivative calculation and make the derivative result more concise.

[0134] Second, calculate the derivative of the Lyapunov function , and use the derivative to analyze whether the path is stable. It can be expressed as:

[0135] ;

[0136] in, represents the Lyapunov function About time The derivative of is often used to analyze the rate of change of the energy or error measure of the system; represents the Lyapunov function, usually the system state Function used to describe the energy or error measure of the system; is the Lyapunov function for the system state The gradient (partial derivative) of Represents the state vector About time The derivative of (i.e., the rate of change of the state); in general, if , the system is stable; if , then the system is unstable and path adjustment is required;

[0137] Based on this, the system analyzes the derivative of the Lyapunov function to determine whether the probe card path is unstable. If the path is unstable, the system automatically adjusts the path by changing control inputs or optimizing path parameters to ensure path stability, thereby preventing the accumulation of path deviations.

[0138] In one possible implementation, the system combines path optimization with stability analysis to adjust path planning parameters in real time. Specifically, the system continuously performs stability analysis during the optimization process and adjusts the probe card's trajectory based on feedback from Lyapunov stability theory. This adjustment not only eliminates instabilities in the path but also ensures that the path remains precisely aligned with the target PIN point.

[0139] For example, if there is local instability in the optimized path, the derivative of the Lyapunov function may be positive, which means that there is an unstable part on the path. In this case, the system will adjust the path according to the calculation result of the derivative so that Path optimization and adjustment not only rely on traditional optimal control theory and particle swarm optimization algorithms, but also incorporate feedback provided by Lyapunov stability theory to further enhance the path optimization effect and ensure that the probe card is always accurately aligned with the target PIN point.

[0140] In summary, this embodiment incorporates Lyapunov stability theory to perform real-time stability analysis of the probe card path and make necessary path adjustments to ensure path stability and optimization. This approach effectively addresses probe card instability during path optimization, ensuring that the probe card remains precisely aligned with the target PIN point throughout the calibration process, thereby improving calibration accuracy and reliability.

[0141] See also Figure 2 , also provides a MEMS probe card semi-automatic PIN correction device, including:

[0142] A large tool microscope for high-resolution imaging of the probe and observing the relationship between the probe and the target PIN point;

[0143] A probe card fixing platform, used for fixing the probe card;

[0144] Motion control system for precise control of the probe card's position, including X, Y, and Z axis motion;

[0145] Image acquisition and processing system, used to collect microscope images in real time and calculate the deviation between the probe and the target PIN point;

[0146] Operation interface, used to display the image and deviation information of the probe and target PIN point, and receive the user's fine-tuning operation;

[0147] The control system is used to coordinate the work of the above components, adjust the probe card to the target PIN point position according to the image analysis results, and perform path planning and optimization through the path optimization module.

[0148] Specifically, a large tool microscope is used to provide high-resolution images, precisely observing the relative position of the probe and the target PIN point. The microscope has adjustable focus and magnification. Through the user interface, the image acquisition accuracy can be adjusted in real time to accommodate the calibration requirements of different probe card types and target PIN points.

[0149] The probe card mounting platform features a porous base plate with multiple vacuum holes to securely hold the probe card in place, preventing it from shifting during calibration. A retaining ring precisely mates with the probe card base, ensuring the probe card remains stable under the motion control system.

[0150] Vacuum adsorption device: By adjusting the vacuum suction force, the fixed platform can adapt to probe cards of different sizes and ensure that they will not move during the entire calibration process.

[0151] Fixed ring structure: The inner diameter of the ring structure is slightly larger than the diameter of the probe card substrate, ensuring the stability of the probe card and avoiding deviation caused by inaccurate positioning.

[0152] The motion control system precisely controls the position of the probe card, including precise motion control along the X, Y, and Z axes. This system utilizes a high-precision platform to ensure the probe card can make minute adjustments along the predetermined path. By adjusting the control system, the probe card is precisely aligned with the target pin point in three dimensions.

[0153] Three-dimensional control system: Controls translational motion along the X, Y, and Z axes, and provides angle control, enabling the probe card to align and maintain precise alignment with the target PIN point.

[0154] The image acquisition and processing system captures images of the probe and target PIN point in real time. It analyzes the image data using image processing algorithms (such as edge detection and contour recognition) to calculate the deviation between the probe and target PIN point. This system provides deviation data to the motion control system, ensuring that the probe card can accurately adjust its position as needed.

[0155] The user interface provides an intuitive user interface that allows users to control the microscope's focus and magnification, and view real-time images and information about the deviation between the probe and the target PIN point. Furthermore, users can make fine adjustments on the interface to ensure the probe card is precisely aligned with the target.

[0156] The control system coordinates all components, ensuring the probe cards accurately adjust their positions based on image analysis results. The path optimization module also performs path planning and optimization. The control system coordinates the work of multiple probe cards and optimizes their paths based on real-time feedback, ensuring precise alignment of the probe cards to the target pins in the shortest possible time.

[0157] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A semi-automatic PIN calibration method for a MEMS probe card, characterized in that: The following steps are involved: Fix the probe card on a fixed platform, and adjust the focus and magnification of the large tool microscope through the operation interface so that the probe is displayed on the image acquisition system; Start the image acquisition and processing system, collect the initial images of the probe and the target PIN point, and extract the position and size parameters of the probe; According to the preset calibration standard, the motion control system automatically adjusts the position of the probe card so that the probe is close to the standard position; The user can make fine adjustments through the operation interface to adjust the position accuracy of the probe; Repeat the above steps to calibrate multiple probes on the probe card in sequence; After calibration is completed, the image acquisition and processing system is used to capture the final image of the probe and the target PIN point, and compared and analyzed with the image before calibration. At the same time, the probe card path is adjusted through real-time image feedback to ensure that the probe card is accurately aligned with the target PIN point and the probe card movement path is optimized; The probe card movement path optimization is performed by combining optimal control theory with particle swarm optimization algorithm; The stability of the probe card's moving path is analyzed using Lyapunov stability theory, and the path is further adjusted to verify the path optimization effect, ensuring that the probe card is always accurately aligned with the target PIN point. The preset calibration standards include: The standard distance between the probe and the target PIN point; Standard angle range of the probe; Alignment deviation tolerance of probe substrate; The path optimization and adjustment steps include: Generate a preliminary path based on the calculated deviation between the probe and the target PIN point; Use optimal control theory to adjust the path so that it meets the predetermined time or energy consumption requirements; Adjust the motion parameters in the path through acceleration and velocity physical constraints; Real-time feedback correction path for precise alignment of the probe card and the target PIN point; The optimal control theory path adjustment step includes: According to the deviation between the probe and the target PIN point, the path optimization objective function is set; Use optimal control theory to solve the time minimization or energy minimization problem of the path; During the solution process, the motion trajectory of the path is adjusted through the acceleration and velocity physical constraints; Generate the optimal path and update it in real time to ensure that the path meets the control requirements.

2. A semi-automatic PIN calibration method for a MEMS probe card according to claim 1, characterized in that: The particle swarm optimization algorithm steps include: Initialize the particle swarm, set the particle position, velocity and fitness function, and calculate the fitness function based on the deviation between the probe and the target PIN point; According to the calculated fitness value, the speed and position of each particle are updated, and the particle moves along the optimal path in the search space; Select the optimal particle according to the fitness value, use it as the current optimal solution, and save the position of the particle; Repeatedly update the particle's velocity and position until the stopping condition is met or the set accuracy requirement is reached; Finally, the globally optimal particle is selected to generate the optimal path, which is used to guide the probe card to accurately align with the target PIN point.

3. A semi-automatic PIN calibration method for a MEMS probe card according to claim 1, characterized in that: The step of adjusting the probe card path by real-time image feedback includes: Collect image data between the probe and the target PIN point in real time and analyze the deviation in the image; Calculate the deviation value based on the image processing results and update the path planning parameters; The motion trajectory of the probe card is adjusted according to the updated path parameters to ensure that it is accurately aligned with the target PIN point.

4. A semi-automatic PIN calibration method for a MEMS probe card according to claim 1, characterized in that: The stability steps of the Lyapunov stability theory analysis path include: Based on the optimized path, construct the Lyapunov function; By calculating the derivative of the Lyapunov function, it is determined that there is an unstable state in the path planning process; The path is adjusted according to Lyapunov stability theory to ensure that the path remains stable during the optimization process.

5. A semi-automatic PIN calibration method for a MEMS probe card according to claim 4, characterized in that: The Lyapunov function formula is: ； in, Is a symmetric positive definite matrix, usually used to measure the error of the system state; is the state vector of the system, representing the deviation between the probe card and the target PIN point; is the state vector The transpose of represents the dot product operation of the row vector and the column vector.

6. A semi-automatic PIN calibration device for a MEMS probe card, applied to a semi-automatic PIN calibration method for a MEMS probe card according to any one of claims 1 to 5, characterized in that: include: A large tool microscope for high-resolution imaging of the probe and observing the relationship between the probe and the target PIN point; A probe card fixing platform, used for fixing the probe card; Motion control system for precise control of the probe card's position, including X, Y, and Z axis motion; Image acquisition and processing system, used to collect microscope images in real time and calculate the deviation between the probe and the target PIN point; Operation interface, used to display the image and deviation information of the probe and target PIN point, and receive the user's fine-tuning operation; The control system is used to coordinate the work of the above components, adjust the probe card to the target PIN point position according to the image analysis results, and perform path planning and optimization through the path optimization module.

7. A semi-automatic PIN calibration device for a MEMS probe card according to claim 6, characterized in that: The probe card fixing platform includes: A porous bottom plate, wherein a plurality of vacuum holes are provided on the bottom plate; A fixing ring, wherein the fixing ring is a structure with a central through hole, the inner diameter of which is slightly larger than the diameter of the probe card substrate and fits with the edge of the probe card substrate; The adsorption device fixes the probe card substrate on the fixed platform by vacuum adsorption.

Images