Semiconductor probe card electrical parameter extraction method and system based on MEMS technology

Abstract

The invention discloses a semiconductor probe card electrical parameter extraction method and system based on an MEMS technology, and relates to the technical field of semiconductor testing, the method comprises the following components: initialization setting: aligning an MEMS probe card to a target test area on a wafer to be tested, firmly fixing the probe card and the wafer together by using a mechanical clamping device, and setting the MEMS probe card and the wafer to be tested; by integrating the decision tree model and the genetic algorithm, accurate mapping and optimization of semiconductor wafer surface features and optimal probe contact parameters are achieved, the intelligent level of testing is improved, the test time is shortened, the test efficiency is improved, and the test cost is reduced. The self-adaptability and stability of the testing process are remarkably enhanced, real-time adjustment of probe contact parameters is ensured through a dynamic monitoring and closed-loop control mechanism, the testing error is effectively reduced, the testing precision is improved, the optimal contact parameter combination is found through an optimization algorithm, the testing period is further shortened, and the testing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor testing technology, and in particular to a method and system for extracting electrical parameters of a semiconductor probe card based on MEMS technology. Background Art

[0002] With the rapid development of semiconductor technology, chip size continues to shrink, integration is increasing, and the requirements for semiconductor wafer testing technology are becoming increasingly stringent. Semiconductor wafer testing is a key link in the semiconductor manufacturing process, and its testing accuracy and efficiency directly affect the final quality and market competitiveness of the product.

[0003] In traditional semiconductor wafer testing technology, dynamic adjustment of probe contact parameters often relies on human experience and manual operation, which has many shortcomings. On the one hand, manual adjustment is difficult to ensure the optimal contact state between the probe and the wafer, and it is easy to introduce test errors. On the other hand, manual operation is inefficient and difficult to meet the needs of large-scale production. In addition, traditional testing technology lacks intelligent optimization algorithms and closed-loop control mechanisms, and cannot adjust probe contact parameters according to wafer surface characteristics in real time, resulting in a lack of adaptability and stability in the testing process.

[0004] Therefore, in view of the deficiencies and shortcomings of the prior art, the present invention proposes a semiconductor probe card electrical parameter extraction method and system based on MEMS technology, which realizes the precise mapping and optimization of semiconductor wafer surface features and optimal probe contact parameters, and significantly improves the intelligence level and adaptability of the test. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and provide a method and system for extracting electrical parameters of a semiconductor probe card based on MEMS technology. It realizes an intelligent leap in semiconductor wafer testing by integrating MEMS technology, decision tree model and genetic algorithm, accurately maps wafer surface features and optimal probe contact parameters, ensures high precision and stability in the testing process through closed-loop control, integrates high-precision probe cards, greatly improves testing efficiency and data reliability, provides an efficient and comprehensive quality monitoring solution for semiconductor manufacturing, and leads the new development of semiconductor testing technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: On the one hand, a method and system for extracting electrical parameters of a semiconductor probe card based on MEMS technology, the specific steps of the extraction method are:

[0007] S100, initialization setting: aligning the MEMS probe card with the target test area on the wafer to be tested, firmly fixing the probe card and the wafer together by using a mechanical clamping device, and starting the dynamic sensor and feedback control mechanism integrated in the probe card;

[0008] S200, real-time monitoring: During the test, sensors are used to capture the tiny morphological changes and motion status of the wafer surface in real time, and these data are transmitted to the control unit in real time through the data transmission channel;

[0009] S300, dynamic adjustment: The control unit receives data from the dynamic sensor, and uses the built-in advanced algorithm to quickly analyze and process the monitored data, calculate the optimal contact force, position and angle that the probe should adjust in the current state, and send the adjustment command to the probe card to drive the probe to make corresponding adjustments;

[0010] S400, electrical parameter extraction: After the probe is adjusted to the optimal contact state, the MEMS probe begins to extract the electrical parameters on the wafer. The probe establishes a stable electrical contact with the pad on the wafer and transmits the measured electrical signal to the signal processing circuit;

[0011] S500, data analysis and output: The signal processing circuit performs real-time analysis on the extracted electrical parameters, evaluates the performance of each test point on the wafer, and outputs the analysis results to the control unit. The control unit compiles the analysis results into a test report and outputs it to the user using the user interface.

[0012] Further, the S100 includes:

[0013] S101, placing the MEMS probe card at a preset position on the test platform, and accurately adjusting the position of the probe card through observation under a microscope;

[0014] S102, firmly fixing the probe card and the wafer to be tested together by using a vacuum suction cup;

[0015] S103, after the probe card and the wafer have been firmly fixed, starting a dynamic sensor and a feedback control mechanism integrated in the probe card through a software interface of the control unit;

[0016] S104, before the formal test, perform a pre-check operation to check the working status of each component;

[0017] S105, after completing all the above steps, confirm that the initialization settings are completed and enter the testing phase.

[0018] Furthermore, the S200 includes:

[0019] S201, capturing the minute morphological changes and motion states of the wafer surface through sensors;

[0020] S202, the sensor continuously scans and monitors the wafer surface, capturing the minute morphological changes and motion state data of the wafer surface in real time;

[0021] S203, preprocessing the collected raw data;

[0022] S204, using the data transmission channel, transmitting the preprocessed data to the control unit in real time.

[0023] Furthermore, the S300 includes:

[0024] S301, the control unit receives in real time the morphological changes, vibration frequency and amplitude of the wafer surface collected by the dynamic sensor;

[0025] S302, performing data verification and preprocessing on the received data;

[0026] S303, based on the pre-processed data, using a decision tree model to perform data analysis to understand the complex mapping relationship between wafer surface features and optimal contact parameters;

[0027] S304, using a genetic algorithm to determine an optimal probe contact parameter combination under a current wafer surface state under given test conditions and constraints;

[0028] S305, based on the optimal contact parameters, the control unit generates an accurate adjustment instruction, sends the adjustment instruction to the MEMS probe card in real time, and adjusts the probe accordingly;

[0029] S306, after the adjustment is completed, the control unit continues to receive real-time monitoring data from the dynamic sensor to form a closed-loop control.

[0030] Furthermore, the S303 performs data analysis based on the preprocessed data using a decision tree model to grasp the complex mapping relationship between the wafer surface features and the optimal contact parameters, extract key features from the data of the wafer surface morphology changes, vibration frequency and amplitude, and calculate the information gain after splitting for each feature. The formula of information gain is: in: Here, K is the number of categories, p k is the proportion of samples belonging to category k in the parent node, p ik is the proportion of samples belonging to category k in child node i. The feature with the largest information gain is selected as the split feature. This process is iterated continuously until the maximum depth is reached to build a complete decision tree. The tree structure is simplified by removing subtrees. The performance of the decision tree is evaluated using validation data. The decision tree model is used to identify the mapping relationship between surface features and optimal contact parameters.

[0031] Furthermore, the S205 automatically controls the fermentation conditions according to PID control, measures the temperature, pH value, dissolved oxygen concentration and carbon dioxide concentration through sensors, converts the analog signals collected by the sensors into digital signals, and transmits them to the controller, integrating the PID control algorithm. The expression of PID control is: Where: u(t) is the output of the controller, e(t) is the deviation between the set value and the measured value (e(t) = r(t) - y(t), r(t) is the set value, y(t) is the measured value), K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient, and the best parameter K of the PID controller is determined by trial and error. p , K i and K d In the actual fermentation environment, the temperature, pH value, dissolved oxygen concentration and carbon dioxide concentration d parameters are collected in real time, the deviation is calculated, the control output is calculated according to the PID algorithm, and the heating / cooling device, acid and alkali addition device, ventilation equipment and carbon dioxide emission device are automatically adjusted.

[0032] Furthermore, the S304 uses a genetic algorithm to find the optimal contact parameter combination under the constraints of the probe movement range and the contact force, and determines that the contact parameters to be optimized are the horizontal position (x, y) of the probe, the vertical position z, the angle θ and the contact force F, and encodes these parameters into a chromosome using real number coding: X = [x, y, z, θ, F], and randomly generates an initial population P, which includes N chromosomes: P = {X1, X2, ..., X N}, define the fitness function f(X) to evaluate the quality of each chromosome, use the roulette wheel selection method, generate a new population through the roulette wheel selection method, crossover and mutation operations, and repeat the iterative process until the predetermined number of iterations is reached. When the termination condition is met, the current best individual is output as the optimal contact parameter combination to guide the adjustment operation of the probe card.

[0033] Furthermore, the S400 includes:

[0034] S401, using a contact force sensor to monitor and adjust the contact force between the probe and the pad until an optimal contact state is reached;

[0035] S402, after confirming that the contact is stable, applying a predetermined test signal to the MEMS probe through a signal generator;

[0036] S403, after receiving the electrical signals from the semiconductor device, the MEMS probe, as a sensor, converts the electrical signals into a processable electrical signal form;

[0037] S404, transmitting the measured electrical signal to the signal processing circuit via a low-noise, high-precision transmission line.

[0038] Furthermore, the S500 includes:

[0039] S501, the signal processing circuit captures the electrical parameter values ​​from the wafer testing equipment in real time, and performs preprocessing operations on the received original electrical parameter values;

[0040] S502, applying statistical analysis to verify the electrical parameter values, evaluating the performance of each test point on the wafer, and automatically identifying and marking test points with abnormal performance and potential failures;

[0041] S503, the control unit receives the output result of the signal processing circuit, integrates the analysis data of multiple test points, and forms a comprehensive wafer performance evaluation report;

[0042] S504, the control unit automatically generates a formatted test report according to the integrated evaluation results, and displays the generated test report to the user through a user-friendly interface.

[0043] On the other hand, a multi-energy complementary heating system for a near-zero energy building comprises the following components:

[0044] MEMS probe array: composed of multiple MEMS probes, used to contact the pads and bumps on the wafer to extract electrical parameters;

[0045] Dynamic sensor: composed of a micro gyroscope and an accelerometer, used to monitor the motion state and surface morphology changes of the wafer during the test in real time;

[0046] Feedback control mechanism: Based on the data collected by the dynamic sensor, the algorithm calculates the optimal contact force, position and angle that the probe should adjust, and controls the probe to make corresponding adjustments;

[0047] Signal processing circuit: used to receive and process the signals transmitted from the MEMS probe and dynamic sensor, and extract the required electrical parameters and motion state data;

[0048] Control unit: As the core control component of the system, it is responsible for coordinating the work of various parts, executing feedback control algorithms, and outputting test results.

[0049] Compared with the prior art, the semiconductor probe card electrical parameter extraction method and system based on MEMS technology has the following beneficial effects:

[0050] 1. The present invention realizes accurate mapping and optimization of semiconductor wafer surface features and optimal probe contact parameters by integrating decision tree model and genetic algorithm, which not only improves the intelligence level of test, but also significantly enhances the adaptability and stability of the test process. The dynamic monitoring and closed-loop control mechanism ensures the real-time adjustment of probe contact parameters, effectively reduces test errors, improves test accuracy, and finds the optimal contact parameter combination through optimization algorithm, further shortens the test cycle and improves test efficiency.

[0051] 2. The present invention integrates high-precision MEMS probe cards and intelligent control systems to achieve accurate extraction of electrical parameters and performance evaluation of semiconductor wafers, which not only improves the automation and intelligence level of testing, but also significantly enhances the accuracy and reliability of testing. Through real-time monitoring and closed-loop control, it ensures the optimal contact state between the probe and the wafer, reduces test errors, and has comprehensive performance evaluation and automatic report generation functions, providing an efficient and convenient quality monitoring method for semiconductor manufacturing, and promoting continuous improvement and enhancement of product quality.

[0052] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0054] Figure 1 A flowchart of a method for extracting electrical parameters of a semiconductor probe card based on MEMS technology;

[0055] Figure 2 The flowchart of the electrical parameter extraction system of the semiconductor probe card based on MEMS technology. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Embodiment 1:

[0058] A method for extracting electrical parameters of a semiconductor probe card based on MEMS technology includes the following steps: initialization setting, real-time monitoring, dynamic adjustment, electrical parameter extraction, data analysis and output.

[0059] First, enter the initialization setting stage (S100), carefully place the MEMS probe card at the preset position of the test platform, and use the vacuum suction cup device on the test platform to firmly fix the adjusted probe card and the wafer to be tested together. After confirming that the probe card and the wafer are firmly fixed, the operator starts the dynamic sensor and feedback control mechanism integrated in the probe card through the software interface of the control unit. The operator checks the working status of each component one by one through the software interface of the control unit, including the sensor, transmission mechanism, and power supply, confirms that the initialization setting has been completed, and enters the test stage.

[0060] Next, the real-time monitoring stage (S200) is entered, and a high-precision sensor array is deployed on the test platform to ensure that the sensor can accurately cover the key areas of the wafer surface. The sensor is started to continuously scan and monitor the wafer surface, and the minute morphological changes and motion state data of the wafer surface are captured in real time. Since the data directly obtained by the sensor often contains noise and redundant information, the quality and availability of the data are improved through pre-processing methods such as filtering, denoising, and compression. The pre-processed data is transmitted to the control unit in real time using an efficient data transmission channel.

[0061] Subsequently, the dynamic adjustment stage (S300) is entered. The control unit starts in real time and monitors the data stream from the dynamic sensor. After receiving the data, the control unit first performs data verification and preprocessing. Based on the preprocessed data, the control unit uses a pre-trained decision tree model to perform data analysis, grasps the complex mapping relationship between the wafer surface features and the optimal contact parameters, and combines the genetic algorithm to determine the optimal probe contact parameter combination under the current wafer surface state under given test conditions and constraints. According to the analysis results of the decision tree model and the optimization algorithm, the control unit calculates the optimal contact parameters that the probe should adjust under the current wafer surface state. Based on the calculated optimal contact parameters, the control unit generates precise adjustment instructions and sends them to the MEMS probe card in real time through the high-speed communication interface. After the adjustment is completed, the control unit does not stop working, but continues to receive real-time monitoring data from the dynamic sensor.

[0062] Then, the electrical parameter extraction stage (S400) is entered, the MEMS probe card is installed on the test platform, and its position is adjusted so that the probes on it can be accurately aligned with the pads of the semiconductor device to be tested. After confirming that the contact is stable, a predetermined test signal is applied to the MEMS probe through a signal generator. After receiving the electrical signal from the semiconductor device, the MEMS probe, as a sensor, converts these electrical signals into a processable electrical signal form, and transmits the measured electrical signal to the signal processing circuit through a low-noise, high-precision transmission line.

[0063] Finally, the data analysis and output stage (S500) is entered. During the wafer test process, the signal processing circuit captures the electrical parameter values ​​from the MEMS probe card in real time. The captured raw electrical parameter values ​​are preprocessed to eliminate noise, offset and inconsistency. For the electrical parameter values ​​of each test point, statistical analysis is applied to verify the electrical parameter values. By calculating the mean, the average level of the parameter is understood. The mean calculation formula is: Where: X i represents the i-th electrical parameter measurement value, n is the total number of measurements, and the variance is calculated to understand the degree of discreteness of the data: the expression of the variance is: The Shapiro-Wilk test method is used to test the normality of the electrical parameter values ​​at each test point. When the confidence level is 95%, it is determined whether the electrical parameter values ​​obey the normal distribution. The confidence interval calculation formula is: Where: z α / 2 It is the standard normal distribution quantile corresponding to the confidence level. The electrical parameter value of the test point is compared with the expected standard range. If the electrical parameter value of a test point exceeds the range of the confidence interval, it is considered that the performance of the test point is abnormal and there may be a potential fault. The control unit receives the output results of the signal processing circuit, including the electrical parameter value of each test point, statistical analysis results and performance evaluation information. The control unit integrates this information to form a comprehensive wafer performance evaluation report. The control unit automatically generates a formatted test report based on the integrated evaluation results.

[0064] Embodiment 2:

[0065] This example particularly emphasizes the dynamic adjustment of the extraction method:

[0066] The control unit receives data from dynamic sensors in real time. These data include morphological changes, vibration frequency and amplitude on the wafer surface. The sensors are installed on the test platform and can accurately capture tiny changes in the wafer during the test process.

[0067] The received raw data first undergoes data validation and preprocessing steps, including checking data integrity, removing noise and outliers, and data standardization to ensure the accuracy and reliability of subsequent analysis.

[0068] Based on the preprocessed data, the decision tree model is used for data analysis. The decision tree model extracts key features from the data by learning the complex mapping relationship between the wafer surface features and the optimal contact parameters. For each feature, the information gain after splitting is calculated. The formula for information gain is: in: Here, K is the number of categories, p k is the proportion of samples belonging to category k in the parent node, p ik is the proportion of samples belonging to category k in child node i. The feature with the largest information gain is selected as the split feature. By continuously iterating this process, a complete decision tree is constructed. The tree structure is simplified by removing subtrees, and the performance of the decision tree is evaluated using validation data. The decision tree model is used to identify the mapping relationship between surface features and optimal contact parameters.

[0069] After determining the contact parameters that need to be optimized (including the horizontal position, vertical position, angle and contact force of the probe), a genetic algorithm is used for further optimization. These parameters are encoded into a chromosome using real number coding: X = [x, y, z, θ, F], and the initial population P is randomly generated, containing N chromosomes: P = {X1, X2, …, X N}, define the fitness function f(X) to evaluate the quality of each chromosome. If the goal is to minimize the contact resistance R(X), the fitness function is: Among them, α is a weight coefficient, φ(X) is a penalty function used to penalize chromosomes that exceed the limit, and a new population is generated through roulette selection, crossover and mutation operations. The iterative process is repeated until the predetermined number of iterations is reached. The chromosome with the highest fitness value is selected from the current population as the optimal solution, that is, the optimal probe contact parameter combination, which is used to guide the adjustment operation of the probe card.

[0070] Based on the optimization results of the genetic algorithm, the optimal contact parameters that the probe should adjust under the current wafer surface state are calculated. These parameters include the precise position, angle and contact force of the probe.

[0071] The control unit generates precise adjustment instructions based on the optimal contact parameters and sends these instructions to the MEMS probe card in real time. After receiving the instructions, the probe card makes corresponding adjustments to the probe to ensure stable contact between the probe and the wafer.

[0072] After the adjustment is completed, the control unit continues to receive real-time monitoring data from the dynamic sensor, which will be used to verify the adjustment effect and trigger a new round of adjustment process.

[0073] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for extracting electrical parameters of a semiconductor probe card based on MEMS technology, characterized in that: The specific steps of the extraction method are: S100, initialization setting: aligning the MEMS probe card with the target test area on the wafer to be tested, firmly fixing the probe card and the wafer together by using a mechanical clamping device, and starting the dynamic sensor and feedback control mechanism integrated in the probe card; S200, real-time monitoring: During the test, sensors are used to capture the tiny morphological changes and motion status of the wafer surface in real time, and these data are transmitted to the control unit in real time through the data transmission channel; S300, dynamic adjustment: The control unit receives data from the dynamic sensor, and uses the built-in advanced algorithm to quickly analyze and process the monitored data, calculate the optimal contact force, position and angle that the probe should adjust in the current state, and send the adjustment command to the probe card to drive the probe to make corresponding adjustments; S400, electrical parameter extraction: After the probe is adjusted to the optimal contact state, the MEMS probe begins to extract the electrical parameters on the wafer. The probe establishes a stable electrical contact with the pad on the wafer and transmits the measured electrical signal to the signal processing circuit; S500, data analysis and output: The signal processing circuit performs real-time analysis on the extracted electrical parameters, evaluates the performance of each test point on the wafer, and outputs the analysis results to the control unit. The control unit compiles the analysis results into a test report and outputs it to the user using the user interface.

2. According to the method for extracting electrical parameters of a semiconductor probe card based on MEMS technology according to claim 1, it is characterized in that: The S100 includes: S101, placing the MEMS probe card at a preset position on the test platform, and accurately adjusting the position of the probe card through observation under a microscope; S102, firmly fixing the probe card and the wafer to be tested together by using a vacuum suction cup; S103, after the probe card and the wafer have been firmly fixed, starting a dynamic sensor and a feedback control mechanism integrated in the probe card through a software interface of the control unit; S104, before the formal test, perform a pre-check operation to check the working status of each component; S105, after completing all the above steps, confirm that the initialization settings are completed and enter the testing phase.

3. The method for extracting electrical parameters of a semiconductor probe card based on MEMS technology according to claim 1, characterized in that: The S200 includes: S201, capturing the minute morphological changes and motion states of the wafer surface through sensors; S202, the sensor continuously scans and monitors the wafer surface, capturing the minute morphological changes and motion state data of the wafer surface in real time; S203, preprocessing the collected raw data; S204, using the data transmission channel, transmitting the preprocessed data to the control unit in real time.

4. The method for extracting electrical parameters of a semiconductor probe card based on MEMS technology according to claim 1, characterized in that: The S300 includes: S301, the control unit receives in real time the morphological changes, vibration frequency and amplitude of the wafer surface collected by the dynamic sensor; S302, performing data verification and preprocessing on the received data; S303, based on the pre-processed data, using a decision tree model to perform data analysis to understand the complex mapping relationship between wafer surface features and optimal contact parameters; S304, using a genetic algorithm to determine an optimal probe contact parameter combination under a current wafer surface state under given test conditions and constraints; S305, based on the optimal contact parameters, the control unit generates an accurate adjustment instruction, sends the adjustment instruction to the MEMS probe card in real time, and adjusts the probe accordingly; S306, after the adjustment is completed, the control unit continues to receive real-time monitoring data from the dynamic sensor to form a closed-loop control.

5. The method for extracting electrical parameters of a semiconductor probe card based on MEMS technology according to claim 4, characterized in that: The S303 uses a decision tree model to perform data analysis based on the preprocessed data, grasps the complex mapping relationship between the wafer surface features and the optimal contact parameters, extracts key features from the data of wafer surface morphology changes, vibration frequency and amplitude, and calculates the information gain after splitting for each feature. The formula of information gain is: in: Here, K is the number of categories, p k is the proportion of samples belonging to category k in the parent node, p ik is the proportion of samples belonging to category k in child node i. The feature with the largest information gain is selected as the split feature. This process is iterated continuously until the maximum depth is reached to build a complete decision tree. The tree structure is simplified by removing subtrees. The performance of the decision tree is evaluated using validation data. The decision tree model is used to identify the mapping relationship between surface features and optimal contact parameters.

6. The method for extracting electrical parameters of a semiconductor probe card based on MEMS technology according to claim 4, characterized in that: S304 uses a genetic algorithm to find the optimal contact parameter combination under the limit of the probe movement range and the contact force, and determines that the contact parameters to be optimized are the horizontal position (x, y) of the probe, the vertical position z, the angle θ and the contact force F, and encodes these parameters into a chromosome using real number coding: X = [x, y, z, θ, F], and randomly generates an initial population P, which includes N chromosomes: P = {X1, X2, ..., X N }, define the fitness function f(X) to evaluate the quality of each chromosome, use the roulette wheel selection method, generate a new population through the roulette wheel selection method, crossover and mutation operations, and repeat the iterative process until the predetermined number of iterations is reached. When the termination condition is met, the current best individual is output as the optimal contact parameter combination to guide the adjustment operation of the probe card.

7. The method for extracting electrical parameters of a semiconductor probe card based on MEMS technology according to claim 1, characterized in that: The S400 includes: S401, using a contact force sensor to monitor and adjust the contact force between the probe and the pad until an optimal contact state is reached; S402, after confirming that the contact is stable, applying a predetermined test signal to the MEMS probe through a signal generator; S403, after receiving the electrical signals from the semiconductor device, the MEMS probe, as a sensor, converts the electrical signals into a processable electrical signal form; S404, transmitting the measured electrical signal to the signal processing circuit via a low-noise, high-precision transmission line.

8. The method for extracting electrical parameters of a semiconductor probe card based on MEMS technology according to claim 1, characterized in that: The S500 includes: S501, the signal processing circuit captures the electrical parameter values ​​from the wafer testing equipment in real time, and performs preprocessing operations on the received original electrical parameter values; S502, applying statistical analysis to verify the electrical parameter values, evaluating the performance of each test point on the wafer, and automatically identifying and marking test points with abnormal performance and potential failures; S503, the control unit receives the output result of the signal processing circuit, integrates the analysis data of multiple test points, and forms a comprehensive wafer performance evaluation report; S504, the control unit automatically generates a formatted test report according to the integrated evaluation results, and displays the generated test report to the user through a user-friendly interface.

9. The method for extracting electrical parameters of a semiconductor probe card based on MEMS technology according to claim 8, characterized in that: The application of statistical analysis verifies the electrical parameter values ​​and evaluates the performance of each test point on the wafer. The average level of the electrical parameters is understood by calculating the mean value in S502. The mathematical expression of the mean value is: Where: X i represents the i-th electrical parameter measurement value, n is the total number of measurements, and the variance is calculated to understand the degree of discreteness of the data: the expression of the variance is: The Shapiro-Wilk test method is used to test the normality of the electrical parameter values ​​at each test point. When the confidence level is 95%, it is determined whether the electrical parameter values ​​obey the normal distribution. The confidence interval calculation formula is: Where: z α / 2 It is the standard normal distribution quantile corresponding to the confidence level. The electrical parameter value of the test point is compared with the expected standard range. If the electrical parameter value of the test point exceeds the range of the mean plus or minus a certain multiple of the standard deviation, the performance of the test point is considered abnormal.

10. A semiconductor probe card electrical parameter extraction system based on MEMS technology, characterized in that: The system consists of the following components: MEMS probe array: composed of multiple MEMS probes, used to contact the pads and bumps on the wafer to extract electrical parameters; Dynamic sensor: composed of a micro gyroscope and an accelerometer, used to monitor the motion state and surface morphology changes of the wafer during the test in real time; Feedback control mechanism: Based on the data collected by the dynamic sensor, the algorithm calculates the optimal contact force, position and angle that the probe should adjust, and controls the probe to make corresponding adjustments; Signal processing circuit: used to receive and process the signals transmitted from the MEMS probe and dynamic sensor, and extract the required electrical parameters and motion state data; Control unit: As the core control component of the system, it is responsible for coordinating the work of various parts, executing feedback control algorithms, and outputting test results.

Images