Detection simulation and life evaluation system and method for chip test probe

Abstract

The invention relates to the field of chip test probe detection and life evaluation, in particular to a chip test probe detection simulation and life evaluation system and method. The system comprises a programmable controller, a test needle execution mechanism, a power supply module, a signal generator and an oscilloscope, and further comprises a temperature monitoring module and a deformation monitoring module which are used for controlling and monitoring the test process. The method comprises the following steps: presetting test parameters, controlling a probe and a test piece to circularly act to apply mechanical and electric stress, acquiring response data, performing data fusion analysis to diagnose a failure mode and evaluate the service life, and performing accelerated test and result judgment. According to the invention, the technical effects of carrying out comprehensive detection simulation and accurate life evaluation on the chip test probe and providing a reliable basis for the quality judgment of the probe are achieved.

Description

Technical Field

[0001] This application relates to the field of chip testing technology, and in particular to a system and method for simulating the detection and lifespan assessment of chip test probes. Background Technology

[0002] In the chip manufacturing and testing industry, test probes are crucial components for signal and current transmission between chips and testing equipment. As chip technology continues to advance, the requirements for chip performance and quality are increasing, making the lifespan and electrical performance of test probes increasingly critical to the accuracy and stability of chip testing. Accurately assessing the lifespan and electrical performance of test probes ensures efficient and stable operation of the chip testing process, thereby improving overall chip quality and production efficiency, and driving the sustainable development of the chip industry.

[0003] Currently, the industry primarily employs traditional methods for verifying the lifetime and evaluating the electrical performance of test probes. This approach relies on feedback from clients during actual use to obtain probe lifetime and electrical performance data, while also utilizing traditional complete testing machines for chip inspection. The former reflects the probe's performance in real-world applications to some extent, while the latter provides a more comprehensive inspection of the chip, indirectly revealing the probe's performance. However, traditional testing machines are designed with the core objective of "testing multiple chips simultaneously," which limits their control over the test probes and their ability to simulate real-world production scenarios.

[0004] Traditional methods for lifetime verification and electrical performance evaluation have significant drawbacks. Relying on customer feedback for data acquisition results in verification lag, meaning that performance issues with probes cannot be detected before shipment. This not only increases user risk for customers but also raises after-sales costs and brand risk for companies. Furthermore, traditional testing equipment is expensive and requires a large footprint, making it only suitable for batch chip testing on large-scale production lines. It cannot meet the needs of companies for specialized probe testing, resulting in poor economic efficiency. Additionally, traditional testing equipment cannot precisely control the movement parameters of the test probes and cannot simulate the cyclical scenarios on actual production lines, making it impossible to repeatedly test the same probe to evaluate its lifetime degradation under different current conditions. Moreover, the lack of an internal, independent verification platform means that probe performance optimization can only begin after receiving customer feedback, extending the R&D cycle and hindering technological innovation efficiency. Summary of the Invention

[0005] The purpose of this application is to overcome the above-mentioned technical problems and provide a system and method for simulating the detection and life assessment of chip test probes.

[0006] A chip test probe detection simulation and lifetime assessment system includes a programmable controller, a test probe actuator, a power module, a signal generator, and an oscilloscope; The programmable controller is electrically connected to the test needle actuator, the power module and the signal generator respectively, and is used to preset test parameters, control the coordinated operation of each component and receive feedback signals. The test probe actuator is used to carry the test probe and drive the test probe to perform a tapping action of contact and separation; The power module is used to output a preset current to the test circuit formed by the probe under test and the test piece when triggered by the programmable controller. The signal generator is used to provide the excitation signal required for the test; The oscilloscope is electrically connected to the test circuit and is used to acquire current waveform data and voltage waveform data in the test circuit. Upon receiving a contact signal generated after the probe under test comes into contact with the test piece, the programmable controller triggers the power module and the signal generator to operate.

[0007] By adopting the above technical solution, probe performance issues can be identified in advance, reducing customer usage risks and enterprise after-sales costs and risks. It meets the enterprise's internal needs for specialized probe testing, precisely controlling probe movement parameters, simulating actual production line scenarios, and repeatedly testing the same probe to evaluate its lifespan degradation under different conditions. It achieves automatic control and synchronization of probe testing actions, electrical signal application, and data acquisition, eliminating reliance on complete testing machines and providing a low-cost, high-efficiency hardware foundation for internal specialized testing.

[0008] Preferably, the system further includes a temperature monitoring module, which is communicatively connected to the programmable controller and is used to monitor the temperature data of the probe under test during the test process; the programmable controller is configured to monitor and analyze the test process based on the temperature data.

[0009] By adopting the above technical solution and integrating a temperature monitoring module, the real-time quantification of probe temperature rise caused by electrothermal effect during the test is realized, providing key data for analyzing temperature-related failures (such as oxidation), and extending the evaluation from the purely electrical field to the thermoelectric coupling field.

[0010] Preferably, the system further includes a deformation monitoring module, which is a laser displacement sensor and is communicatively connected to the programmable controller for non-contact monitoring of the microscopic deformation data of the probe tip under test; the programmable controller is configured to evaluate the mechanical wear of the probe under test based on the deformation data.

[0011] By adopting the above technical solution and introducing high-precision laser displacement sensing, non-contact and accurate measurement of the probe's microscopic deformation is achieved, extending the dimension of life assessment from electrical performance to mechanical morphology, and providing direct evidence for diagnosing purely mechanical failures.

[0012] A method for simulating the detection and assessing the lifetime of a chip test probe includes the following steps: Test parameters are preset by a programmable controller. These test parameters include motion parameters for generating mechanical stress, electrical parameters for generating electrical stress, and cyclic parameters. The control test probe actuator drives the contact and separation cycle between the test probe and the test piece according to the test parameters to apply the mechanical stress; and when the contact action occurs, an electrical excitation signal is applied to the test circuit formed by the test probe and the test piece to apply the electrical stress synchronously. Collect the first response data generated by the probe under test based on the electrical stress and the second response data generated based on the mechanical stress, and summarize them to generate multi-performance response data; Based on the multi-performance response data, data fusion analysis is performed to diagnose the dominant failure mode of the probe under test and evaluate the lifetime of the probe under test.

[0013] By adopting the above technical solution, composite stress can be applied simultaneously, which can realistically simulate the working state of the probe under the simultaneous mechanical impact and electrothermal stress in actual working conditions. This makes the simulation of the failure mechanism more realistic from the root, solves the problems of single test conditions and distorted scenarios in traditional testing, and allows probe performance problems to be detected in advance before leaving the factory, reducing customer usage risks and enterprise after-sales costs.

[0014] Preferably, the first response data includes at least real-time contact resistance data calculated based on the current and voltage waveforms of the test circuit, and probe temperature response data acquired by a temperature sensor; the current and voltage waveforms are acquired by an oscilloscope after the electrical excitation signal is applied to the test circuit. The second response data includes at least the probe tip micro-deformation data acquired by the laser displacement sensor; The failure mode includes at least one of oxidation failure, wear failure, and plastic deformation failure.

[0015] By adopting the above technical solution, the specific composition of the multidimensional data (resistance, temperature, deformation) on which the diagnosis depends is clarified, providing more comprehensive data support for subsequent analysis. It also clarifies the possible failure modes of the probe under test, such as oxidation failure, wear failure, and plastic deformation failure. By anchoring the abstract data fusion analysis to specific physical quantities and engineering problems, it helps to accurately diagnose the dominant failure mode of the probe under test and makes the technical solution clearly feasible.

[0016] Preferably, the data fusion analysis based on the multi-performance response data specifically includes the following steps: Based on the real-time contact resistance data, the probe temperature response data, and the probe tip micro-deformation data, a multi-parameter performance degradation dataset with the number of cycles as the independent variable is constructed. Establish an internal correlation model for the aforementioned multi-parameter performance degradation dataset, specifically as follows: Using the number of cycles as a time index, the real-time contact resistance, the temperature response data, and the micro-deformation data are respectively constructed into discrete data sequences that vary with the number of cycles, including resistance sequence, temperature sequence, and deformation sequence; The rate of change of each discrete data sequence relative to the number of cycles is calculated to obtain a rate of change sequence, which includes a resistance rate of change sequence, a temperature rise rate sequence, and a deformation rate sequence. Calculate the correlation coefficient between every two of the rate of change sequences to form a coupling matrix, where any pair of parameters in the coupling matrix represents the correlation strength between the corresponding two parameter rate of change during the test process; Analyze the coupling matrix and select the parameter pairs whose absolute values ​​of the correlation coefficients exceed a preset correlation threshold as the key coupling parameter pairs that dominate the current performance degradation process.

[0017] By adopting the above technical solution, a multi-parameter performance degradation dataset with the number of cycles as the independent variable was constructed, and an internal correlation model was established for this dataset. By constructing a coupling matrix and identifying key coupling parameter pairs, the complex interaction relationship between multiple parameters was quantitatively characterized. This enables a more scientific analysis of the correlation and changes between various performance parameters of the test probe under different stresses, laying a solid and computable mathematical foundation for subsequent intelligent diagnosis.

[0018] Preferably, determining the failure mode of the probe under test based on the key coupling parameters specifically includes the following steps: Calculate the first correlation coefficient between the resistance change rate sequence and the temperature rise rate sequence, the second correlation coefficient between the resistance change rate sequence and the deformation rate sequence, and the third coefficient of determination for the linear fit of the deformation data sequence with respect to the number of cycles. The absolute values ​​of the first correlation coefficient and the second correlation coefficient are compared with the first preset threshold, and the third determination coefficient is compared with the second preset threshold; If the first correlation coefficient is positive and has the largest absolute value, and the absolute value exceeds the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as oxidation failure. If the absolute value of the second correlation coefficient is the largest and exceeds the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as wear failure. If the third determination coefficient exceeds the second preset threshold, and the absolute values ​​of both the first correlation coefficient and the second correlation coefficient are lower than the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as plastic deformation failure.

[0019] By adopting the above technical solution, an automated diagnostic rule for non-obvious failure modes based on explicit mathematical criteria (comparison of three coefficients and thresholds) is defined. This rule can determine the failure mode of the probe under test based on key coupling parameters, thereby achieving objectivity, standardization, and intelligence in the diagnostic process and providing a basis for subsequent probe lifetime assessment and performance optimization.

[0020] Preferably, the lifetime assessment process for the probe under test specifically includes the following steps: Based on the dominant failure mode obtained from the diagnosis, the target data sequence is determined according to the corresponding discrete data sequence. For the selected target data sequence, a fitting function is obtained by performing a mathematical function fitting on the target data sequence, using the number of loops already executed as the independent variable; Calculate the value of the independent variable when the function value of the fitting function reaches the failure threshold preset for the current target data sequence, and use the value of the independent variable as the predicted lifetime cycle number; The predicted lifetime cycle number is used as the evaluated lifetime of the probe under test.

[0021] By adopting the above technical solution and through mathematical fitting and threshold solving, it is possible to quantitatively predict the remaining lifespan of the probe before its performance completely fails, realizing a leap from "post-judgment" to "pre-warning", which greatly enhances the predictive value of the test.

[0022] Preferably, after the predicted lifetime cycle number is obtained through evaluation, the following steps are further included: If the predicted lifetime cycle count is greater than the preset expected verification cycle count, the accelerated test is initiated. The accelerated testing is achieved by enhancing the corresponding test parameters based on the dominant failure mode; Specifically, if the dominant failure mode is oxidation failure, then the electrical parameters corresponding to the electrical stress are enhanced; If the dominant failure mode is wear failure or plastic deformation failure, then the motion parameters corresponding to the mechanical stress are enhanced.

[0023] By adopting the above technical solution, a closed-loop control function of "adaptive accelerated testing" is added on the basis of diagnosis and prediction. Based on the preliminary prediction results and diagnostic mode, the key stress can be intelligently strengthened to accelerate life verification. Thus, while ensuring that the failure mechanism remains unchanged, the verification cycle of long-life products is significantly shortened, solving the core pain point of low R&D testing efficiency.

[0024] Preferably, after evaluating the lifetime of the probe under test, the following steps are further included: The predicted lifetime cycle count obtained from the assessment is compared with the preset qualified lifetime standard; If the predicted lifetime cycle count reaches the qualified lifetime standard, the probe is deemed qualified, and a test report containing the determination result and lifetime data is output. If the predicted lifetime cycle count is lower than the acceptable lifetime standard, the probe is deemed unqualified, and a test report containing the determination result, failure mode diagnostic information, and lifetime data is output.

[0025] By adopting the above technical solution, the life assessment results can be automatically compared with the quality standards, and a comprehensive report containing diagnostic conclusions can be generated, which directly serves the product quality determination (qualified / unqualified). This transforms complex test data into an intuitive and actionable basis for decision-making, and realizes the ultimate realization of the value of testing.

[0026] In summary, this application includes at least one of the following beneficial technical effects: (1) This application can accurately diagnose the fundamental failure mechanism that leads to performance degradation by simultaneously loading mechanical and electrical stress, simultaneously collecting multi-dimensional response data of electro-thermal-mechanical systems, and based on innovative data fusion and correlation model algorithms, thus upgrading the test from judging whether it has failed to analyzing why it failed.

[0027] (2) This application can not only predict the lifespan of the probe before it completely fails, but also intelligently adjust the test parameters based on the diagnostic results to conduct targeted accelerated testing of the dominant failure mechanism. It transforms the passive verification that originally relied on long-term natural failure or customer feedback into an active, fast, and goal-oriented internal R&D tool, which greatly shortens the product development and quality certification cycle. Attached Figure Description

[0028] Figure 1 This is a flowchart of the detection simulation and lifetime assessment method for chip test probes in the embodiment. Detailed Implementation

[0029] The technical solutions in the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. The described embodiments are only possible technical implementations of the present invention, but are not limited thereto. Other embodiments obtained by those skilled in the art in conjunction with the embodiments of the present invention without creative effort are also within the protection scope of the present invention.

[0030] This application mainly uses a programmable controller to coordinate multiple components to simulate test probes, which achieves the effect of accurately evaluating the lifespan and performance of chip test probes. The following is a further detailed description of this application.

[0031] Example 1 The chip test probe detection simulation and lifetime assessment system provided in this application includes a programmable controller, a test probe actuator, a power supply module, a signal generator, and an oscilloscope. The programmable controller is electrically connected to the test probe actuator, the power supply module, and the signal generator, and is used to preset test parameters, control the coordinated operation of each component, and receive feedback signals. This achieves the effect of accurately controlling the test process and comprehensively acquiring test data. As the core control component, the programmable controller performs unified scheduling and management of each component to ensure that the test process is carried out according to the preset parameters.

[0032] Specifically, The programmable controller is electrically connected to the test probe actuator, power module, and signal generator, respectively, and is used to preset test parameters, control the coordinated operation of each component, and receive feedback signals.

[0033] The programmable controller (PCC) can be an industrial-grade programmable logic controller (PLC), which features high stability and strong anti-interference capabilities, or it can use a microcontroller or other chip with programmable control functions. Internally, the PCC contains a storage unit and a processing unit. The storage unit stores preset test parameters and programs, while the processing unit controls and processes data from various components according to the preset programs. The PCC connects to other components via data cables. For example, when connected to a test probe actuator, it can send control signals to drive the actuator to operate according to preset motion parameters.

[0034] The test probe actuator is used to carry the test probe and drive it to perform a tapping action that involves contact and separation.

[0035] The power module is used to output a preset current to the test circuit formed by the probe under test and the test piece when triggered by the programmable controller.

[0036] The power module is connected to the test circuit via wires. Triggered by the programmable controller, it outputs a preset current to the test circuit formed by the probe under test and the test piece. The power module has an internal current regulation circuit that can precisely adjust the output current according to the instructions of the programmable controller.

[0037] A signal generator is used to provide the excitation signal required for testing.

[0038] A signal generator can be a function signal generator or an arbitrary waveform generator. The signal generator connects to the test circuit via signal lines to provide the necessary excitation signal for the test, ensuring the consistency of the test environment with the actual chip testing scenario. Internally, the signal generator contains waveform generation circuitry and frequency adjustment circuitry, enabling precise control of the waveform and frequency of the output signal.

[0039] An oscilloscope, electrically connected to a test circuit via a probe, is used to acquire current and voltage waveform data from the test circuit and transmit the data to a programmable controller or external storage device. Internally, an oscilloscope contains amplification and sampling circuits, enabling it to amplify and sample weak signals, thus improving the accuracy of data acquisition.

[0040] After receiving the contact signal generated when the probe under test comes into contact with the test piece, the programmable controller triggers the power module and signal generator to work.

[0041] In another specific implementation, the system also includes a temperature monitoring module and a deformation monitoring module.

[0042] Specifically, the temperature monitoring module is installed near the probe under test and communicates with the programmable controller via a data cable to monitor the temperature data of the probe under test during the test process; the programmable controller is configured to monitor and analyze the test process based on the temperature data.

[0043] The system also includes a deformation monitoring module, which is a laser displacement sensor. Laser displacement sensors offer advantages such as non-contact measurement, high accuracy, and fast response. The laser displacement sensor is installed near the tip of the probe under test and communicates with the programmable controller via a data cable. It is used for non-contact monitoring of the microscopic deformation data of the probe tip. The programmable controller is configured to evaluate the mechanical wear of the probe under test based on the deformation data. For example, if the deformation exceeds a certain range, it indicates that the probe may have already experienced wear.

[0044] In another specific implementation, the system further includes a high-frequency performance testing module; the high-frequency performance testing module is a vector network analyzer, which is communicatively connected to the programmable controller and is used to measure the S-parameters of the probe under test during the test. The programmable controller is configured to evaluate the high-frequency signal transmission performance of the probe under test based on the S-parameters.

[0045] The implementation principle of this embodiment is as follows: The system uses a programmable controller to precisely control each component, simulating actual chip testing scenarios and performing comprehensive testing simulation and lifespan assessment of chip test probes. The addition of temperature and deformation monitoring modules allows the system to acquire more test data, thereby more accurately evaluating probe performance and lifespan. Compared to traditional testing methods, this system can detect probe performance issues early, reducing customer usage risks and after-sales costs for enterprises, while also shortening R&D cycles and improving technological innovation efficiency.

[0046] Example 2 The chip test probe detection simulation and lifetime assessment method provided in this application embodiment is based on the above system, such as... Figure 1 As shown, it includes the following steps: S1. Preset test parameters through the programmable controller. The test parameters include motion parameters for generating mechanical stress, electrical parameters for generating electrical stress, and cycle parameters.

[0047] In one specific implementation, test parameters are preset through the human-machine interface of the programmable controller, such as: Motion parameters include the striking speed of the probe under test, the stroke (the contact distance between the probe under test and the test piece), and the contact pressure; Electrical parameters include the output current and voltage of the power module; The cyclic parameters include the number of "contact-separation" cycles and the cycle frequency for simulating high-speed, high-volume detection.

[0048] S2. Simulated impact test: The programmable controller sends a control signal to the test probe actuator, which drives the test probe to contact and separate the test chip in a cyclical action according to the test parameters, so as to apply mechanical stress. Through the repeated impact action of high frequency contact and separation, the cyclical scenario of continuous chip feeding, testing and delivery on the actual production line is simulated. When the contact action occurs, an electrical excitation signal is applied to the test circuit formed by the probe under test and the test piece to apply electrical stress synchronously.

[0049] When the probe under test contacts the test piece to form a test circuit, the test probe actuator or the test piece carrier feeds back a contact signal to the programmable controller. The programmable controller triggers the power module and signal generator to output an electrical excitation signal to the test circuit based on the contact signal.

[0050] Feedback of contact signals can be achieved in various ways, such as... In one specific implementation, there is a feedback circuit between the programmable controller, the test piece, and the probe under test. When the test piece contacts the test probe, the feedback circuit is activated and feeds back the contact signal to the programmable controller.

[0051] In another specific implementation, the programmable controller can also calculate the time when the test probe contacts the test piece according to the preset test parameters, and feed this time point back to the programmable controller as a feedback signal. After receiving the feedback signal, the programmable controller triggers the power module to output a preset current to the test circuit, and at the same time controls the signal generator to output a coordinated excitation signal to form a complete test circuit.

[0052] S3. Collect the first response data generated by the probe under test based on electrical stress and the second response data generated based on mechanical stress, and summarize them to generate multi-performance response data.

[0053] The first response data includes at least the real-time contact resistance data calculated based on the current and voltage waveforms of the test circuit, and the probe temperature response data acquired by the temperature sensor; the current and voltage waveforms are acquired by an oscilloscope after an electrical excitation signal is applied to the test circuit.

[0054] The second response data includes at least the probe tip micro-deformation data acquired by the laser displacement sensor.

[0055] S4. Based on multi-performance response data, perform data fusion analysis to diagnose the dominant failure mode of the probe under test and assess its lifetime. This includes the following steps: The failure mode must include at least one of oxidation failure, wear failure, or plastic deformation failure.

[0056] S41. Data fusion analysis of multiple performance response data includes the following steps: S411. Based on real-time contact resistance data, probe temperature response data, and probe tip micro-deformation data, a multi-parameter performance degradation dataset with the number of cycles as the independent variable is constructed.

[0057] S412. Establish a correlation model within the multi-parameter performance degradation dataset. This correlation model is used to quantify the interaction and evolutionary synergy of different performance parameters during the probe's degradation process. Specifically, Generate discrete data sequences: Using the number of cycles as the time index, the real-time contact resistance, temperature response data, and micro-deformation data are respectively constructed into discrete data sequences that change with the number of cycles, including the resistance sequence {R(n)}, the temperature sequence {T(n)}, and the deformation sequence {D(n)}, where n is the number of cycles.

[0058] Calculate the rate of change: Calculate the rate of change of each discrete data sequence relative to the number of cycles to obtain the rate of change sequence, which includes the resistance rate of change sequence {ΔR(n)}, the temperature rise rate sequence {ΔT(n)}, and the deformation rate sequence {ΔD(n)}; Calculate the correlation coefficient between every two rate of change sequences to form a coupling matrix M. Any parameter pair in the coupling matrix represents the correlation strength between the corresponding two parameter rates of change during the test process. That is, for the matrix element M_ij in the coupling matrix M, it represents the linear correlation strength between the rate of change of the i-th parameter and the rate of change of the j-th parameter throughout the entire test process.

[0059] Analyze the coupling matrix and select parameter pairs whose absolute values ​​of correlation coefficients exceed a preset correlation threshold as the key coupling parameter pairs that dominate the current performance degradation process.

[0060] S413. Determine the failure mode of the probe under test based on the key coupling parameters, specifically including the following steps: In this embodiment, the Pearson correlation coefficient algorithm, which is conventional in the art, is used to calculate the correlation between the two sequences, and the least squares method is used for linear fitting and to calculate the coefficient of determination.

[0061] The first correlation coefficient between the resistance change rate sequence and the temperature rise rate sequence, the second correlation coefficient between the resistance change rate sequence and the deformation rate sequence, and the third coefficient of determination for the linear fit of the deformation data sequence with respect to the number of cycles were calculated respectively.

[0062] The first correlation coefficient is used to quantify the degree of linear correlation between resistance change and temperature change, while the third coefficient of determination is used to quantify the degree of agreement between the trend of deformation change with the number of cycles and the linear model.

[0063] The absolute values ​​of the first correlation coefficient and the second correlation coefficient are compared with the first preset threshold, and the third deterministic coefficient is compared with the second preset threshold. If the first correlation coefficient is positive and has the largest absolute value, and the absolute value exceeds the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as oxidation failure.

[0064] This is because oxidation is dominated by electro-thermal coupling. The current generates Joule heat, and the high temperature accelerates the growth of the oxide film on the contact surface. The oxide film increases the resistance, which in turn generates more heat, forming a positive feedback loop. Under complex operating conditions, multiple couplings may coexist, but when electro-thermal coupling is the most prominent among all coupling relationships, oxidation becomes the dominant failure mode.

[0065] If the absolute value of the second correlation coefficient is the largest and exceeds the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as wear failure.

[0066] This is because wear is dominated by electromechanical coupling. Mechanical impact causes changes in the micro-geometry of the needle tip, resulting in unstable contact area, which directly manifests as drastic fluctuations in contact resistance. Deformation is the cause, and resistance fluctuation is the result. Wear may lead to poor contact (increased resistance) or instantaneous micro-welding (decreased resistance). Therefore, the correlation may be positive or negative. Thus, we only need to focus on the strength of the correlation, that is, the absolute value. Therefore, when the strength of the electromechanical coupling is the highest, it is judged that wear is dominant.

[0067] If the third coefficient of determination exceeds the second preset threshold, and the absolute values ​​of the first correlation coefficient and the second correlation coefficient are both lower than the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as plastic deformation failure.

[0068] Plastic deformation is purely driven by mechanical fatigue. Under cyclic stress, the material undergoes irreversible elongation or bending, which is a cumulative process mainly related to the number of cycles (time / stress cycles) and minimally affected by electricity. Setting a threshold for the determination coefficient confirms that deformation is a stable and predictable cumulative process. At the same time, significant interferences from electro-thermal coupling and electromechanical coupling must be eliminated to ensure that plastic deformation is a relatively independent process primarily driven by mechanical cycles.

[0069] In one specific implementation, the above calculation process is simplified to: First, calculate the Pearson correlation coefficient between the resistance change rate sequence and the temperature rise rate sequence, denoted as Corr_RT; then calculate the Pearson correlation coefficient between the resistance change rate sequence and the deformation rate sequence, denoted as Corr_RD; perform a linear regression on the deformation sequence with respect to the number of cycles to obtain the coefficient of determination R2_D; compare the absolute values ​​of Corr_RT and Corr_RD, |Corr_RD|, with the first preset threshold Th_Corr; and compare R2_D with the second preset threshold Th_R2.

[0070] For example, the calculated values ​​are Corr_RT = +0.72, Corr_RD = -0.45, and R2_D = 0.65; With Th_Corr=0.6 and Th_R2=0.75, |Corr_RT|>Th_Corr and is the maximum, therefore the dominant failure mode is diagnosed as oxidation failure.

[0071] S414. The lifetime assessment process for the probe under test includes the following steps: Based on the dominant failure mode obtained from the diagnosis, the target data sequence is determined according to the corresponding discrete data sequence.

[0072] Specifically, if the diagnosis is oxidation failure, the target data sequences to be determined are the resistance sequence and the temperature sequence. If the diagnosis is wear failure, the determined target data sequence is the resistance sequence and the deformation sequence; If the diagnosis is plastic deformation failure, then the determined target data sequence is the deformation sequence.

[0073] For a selected target data sequence, the target data sequence is fitted with a mathematical function F(n) using the number of loops executed n as the independent variable, where i = 1, 2, ..., k, and k is the number of loops completed so far. This can be achieved through linear fitting or exponential fitting.

[0074] When oxidation failure is diagnosed and the resistance sequence {R(n)} and temperature sequence {T(n)} are selected, the model fitting includes: R_fit(n) is obtained by fitting a function to the resistance sequence {R(n)}, and T_fit(n) is obtained by fitting a function to the temperature sequence {T(n)}. When wear is diagnosed and the resistance sequence {R(n)} and deformation sequence {D(n)} are selected, the model fitting includes: R_fit(n) is obtained by fitting a function to the resistance sequence {R(n)}, and D_fit(n) is obtained by fitting a function to the deformation sequence {D(n)}. When the diagnosis is plastic deformation and the deformation sequence {D(n)} is selected, the model fit is: D_fit(n) is obtained by fitting a function to the deformation sequence {D(n)}.

[0075] Calculate the value of the independent variable when the function value of the fitted function reaches the failure threshold preset for the current target data sequence, and use the value of the independent variable as the predicted lifetime cycle number.

[0076] The predicted lifetime cycle count is used as the evaluated lifetime of the probe under test.

[0077] In one specific implementation, when two target data sequences are selected, one target data sequence is designated as the main prediction sequence and the other as the constraint sequence. Function fitting is then performed on the main prediction sequence and the constraint sequence respectively. When calculating the number of predicted lifetime cycles, the solved independent variable values ​​must simultaneously satisfy the following: A. The fitting function value of the main prediction sequence reaches its failure threshold; B. And the fitted function value of the constrained sequence under the same independent variable does not exceed its warning threshold; The smallest independent variable value that simultaneously satisfies both conditions is used as the predicted lifetime cycle number.

[0078] In another specific implementation, when the contact and separation cycle test reaches a preset cycle node, the mechanical impact action is paused, and the test probe actuator is controlled to keep the test probe and the test piece in stable contact. A high-frequency test signal is injected into the test loop using a vector network analyzer, and the S-parameters of the probe under test at the current loop node are measured. Record the changes in S-parameters with the number of cycles. When any defined key S-parameter deteriorates beyond the preset high-frequency performance threshold, the probe under test is determined to have reached the end of its high-frequency performance lifespan. For the final lifetime assessment result of the probe under test, the smaller value between the predicted lifetime cycle number and the high-frequency performance lifetime endpoint cycle number is taken.

[0079] In another specific implementation, the correlation between the degradation characteristics of the S-parameters and the dominant failure mode can be further analyzed: If the degradation of S-parameters is mainly manifested as a slow and monotonically increasing insertion loss, and the trend of increasing resistance is synchronized at the cyclic node, then it is determined that the high-frequency performance degradation is mainly caused by the increase in contact impedance due to tip oxidation. If the S-parameter degradation manifests as a sharp deterioration in return loss or a jump in insertion loss, and is correlated with abrupt changes in deformation data, then the high-frequency performance degradation is determined to be mainly caused by geometric deformation of the contact surface due to mechanical wear.

[0080] In real-world applications, some failures only manifest at high frequencies. For example, a tiny crack in the internal structure of a probe may have a negligible impact on DC resistance, but it can severely disrupt the transmission of high-frequency signals. Through correlation analysis, failure modes for such high-frequency structural failures can be defined.

[0081] S5. After assessing the predicted lifetime cycle count, dynamically adjust the test parameters of the test probe actuator or power module based on the diagnosed dominant failure mode to execute a targeted adaptive accelerated testing strategy, including the following steps: If the predicted lifetime cycle count is greater than the preset expected verification cycle count, start the accelerated test; Accelerated testing is achieved by enhancing the corresponding test parameters based on the dominant failure mode. The aim is to specifically enhance the corresponding driving stress parameters based on the dominant failure mode, so as to shorten the number of cycles required to reach the end of the life while keeping the dominant failure mode unchanged, thereby further solving the problem of long R&D cycles.

[0082] Specifically, if the dominant failure mode is oxidation failure, the electrical parameters corresponding to the electrical stress are enhanced, specifically by increasing the current value of the test circuit or increasing the ambient temperature of the probe under test, without significantly increasing the mechanical shock frequency or pressure. If the dominant failure mode is wear failure or plastic deformation failure, then enhance the motion parameters corresponding to the mechanical stress.

[0083] Wear failure specifically increases the impact frequency or contact pressure of the probe under test, without significantly increasing the test current or ambient temperature; Plastic deformation failure specifically involves increasing the probe's stroke or impact energy.

[0084] In one specific implementation, after the adaptive accelerated testing strategy is activated, if the dominant failure mode is oxidation failure, the test environment temperature is increased by 10-30°C, or the test current amplitude is increased by 10%-30%, while keeping other test parameters unchanged, to accelerate the oxidation reaction process. If the dominant failure mode is wear, increase the contact pressure by 0.05-0.2N or increase the cycle frequency by 10-30 times / second to accelerate mechanical wear. If the dominant failure mode is plastic deformation, maintain the original test parameters and focus on increasing the acquisition frequency of the laser displacement sensor, such as acquiring deformation data once every 50 cycles, to accurately capture the critical node of plastic deformation. During accelerated testing, multiple performance response data are collected synchronously. The fitting function is corrected based on the dataset from the accelerated phase, and the corrected predicted lifetime cycle count is finally output, making the lifetime assessment results more consistent with actual working conditions.

[0085] S6. After evaluating the lifetime of the probe to be tested, the following steps are also included: The predicted lifetime cycle count obtained from the assessment is compared with the preset qualified lifetime standard; If the predicted lifetime cycle count reaches the qualified lifetime standard, the probe is deemed qualified, and a test report containing the judgment result and lifetime data is output. If the predicted lifetime cycle count is lower than the qualified lifetime standard, the probe is deemed unqualified, and a test report containing the judgment result, failure mode diagnosis information, and lifetime data is output.

[0086] In another specific implementation, based on the diagnosed dominant failure mode and predicted lifetime cycle count, design or process optimization suggestions for the probe under test can also be generated: If the dominant failure mode is oxidation failure, optimization recommendations include: recommending to enhance the anti-oxidation coating of the probe tip, or recommending in the test report to limit the maximum test current in the customer's usage scenario; If the dominant failure mode is wear, optimization recommendations include: increasing the hardness of the probe substrate material or optimizing the geometry of the probe tip to reduce contact stress. If the dominant failure mode is plastic deformation, optimization recommendations include: increasing the yield strength of the probe material or reducing the probe working stroke in the test report.

[0087] The implementation principle of this embodiment is as follows: Based on the above-mentioned detection simulation and lifetime assessment system, this method can more accurately control test conditions, simulate actual test scenarios, obtain more test data, and more effectively evaluate the lifetime decay law of probes under different current conditions by presetting test parameters, simulating actual test scenarios, performing data fusion and failure diagnosis, predicting probe lifetime, performing adaptive accelerated testing, and generating test reports.

[0088] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A chip test probe detection simulation and lifetime assessment system, characterized in that: It includes a programmable controller, a test probe actuator, a power supply module, a signal generator, and an oscilloscope; The programmable controller is electrically connected to the test needle actuator, the power module and the signal generator respectively, and is used to preset test parameters, control the coordinated operation of each component and receive feedback signals. The test probe actuator is used to carry the test probe and drive the test probe to perform a tapping action of contact and separation; The power module is used to output a preset current to the test circuit formed by the probe under test and the test piece when triggered by the programmable controller. The signal generator is used to provide the excitation signal required for the test; The oscilloscope is electrically connected to the test circuit and is used to acquire current waveform data and voltage waveform data in the test circuit. Upon receiving a contact signal generated after the probe under test comes into contact with the test piece, the programmable controller triggers the power module and the signal generator to operate.

2. The chip test probe detection simulation and lifetime assessment system according to claim 1, characterized in that: The system also includes a temperature monitoring module, which is communicatively connected to the programmable controller and is used to monitor the temperature data of the probe under test during the test process; the programmable controller is configured to monitor and analyze the test process based on the temperature data.

3. The chip test probe detection simulation and lifetime assessment system according to claim 1, characterized in that: The system also includes a deformation monitoring module, which is a laser displacement sensor and is communicatively connected to the programmable controller for non-contact monitoring of the microscopic deformation data of the probe tip under test. The programmable controller is configured to evaluate the mechanical wear of the probe under test based on the deformation data.

4. A method for simulating chip test probe detection and assessing lifetime, characterized in that, Includes the following steps: Test parameters are preset by a programmable controller. These test parameters include motion parameters for generating mechanical stress, electrical parameters for generating electrical stress, and cyclic parameters. The control test probe actuator drives the contact and separation cycle between the test probe and the test piece according to the test parameters to apply the mechanical stress; When the contact action occurs, an electrical excitation signal is applied to the test circuit formed by the probe under test and the test piece to synchronously apply the electrical stress. Collect the first response data generated by the probe under test based on the electrical stress and the second response data generated based on the mechanical stress, and summarize them to generate multi-performance response data; Based on the multi-performance response data, data fusion analysis is performed to diagnose the dominant failure mode of the probe under test and evaluate the lifetime of the probe under test.

5. The method for simulating chip test probe detection and assessing lifetime according to claim 4, characterized in that: The first response data includes at least real-time contact resistance data calculated based on the current and voltage waveforms of the test circuit, and probe temperature response data acquired by a temperature sensor; the current and voltage waveforms are acquired by an oscilloscope after the electrical excitation signal is applied to the test circuit. The second response data includes at least the probe tip micro-deformation data acquired by the laser displacement sensor; The failure mode includes at least one of oxidation failure, wear failure, and plastic deformation failure.

6. The method for simulating chip test probe detection and assessing lifetime according to claim 5, characterized in that, The data fusion analysis based on the multi-performance response data specifically includes the following steps: Based on the real-time contact resistance data, the probe temperature response data, and the probe tip micro-deformation data, a multi-parameter performance degradation dataset with the number of cycles as the independent variable is constructed. Establish an internal correlation model for the aforementioned multi-parameter performance degradation dataset, specifically as follows: Using the number of cycles as a time index, the real-time contact resistance, the temperature response data, and the micro-deformation data are respectively constructed into discrete data sequences that vary with the number of cycles, including resistance sequence, temperature sequence, and deformation sequence; The rate of change of each discrete data sequence relative to the number of cycles is calculated to obtain a rate of change sequence, which includes a resistance rate of change sequence, a temperature rise rate sequence, and a deformation rate sequence. Calculate the correlation coefficient between every two of the rate of change sequences to form a coupling matrix, where any pair of parameters in the coupling matrix represents the correlation strength between the corresponding two parameter rate of change during the test process; Analyze the coupling matrix and select the parameter pairs whose absolute values ​​of the correlation coefficients exceed a preset correlation threshold as the key coupling parameter pairs that dominate the current performance degradation process.

7. The method for simulating chip test probe detection and assessing lifetime according to claim 6, characterized in that, Determining the failure mode of the probe under test based on the key coupling parameters includes the following steps: Calculate the first correlation coefficient between the resistance change rate sequence and the temperature rise rate sequence, the second correlation coefficient between the resistance change rate sequence and the deformation rate sequence, and the third coefficient of determination for the linear fit of the deformation data sequence with respect to the number of cycles. The absolute values ​​of the first correlation coefficient and the second correlation coefficient are compared with the first preset threshold, and the third determination coefficient is compared with the second preset threshold; If the first correlation coefficient is positive and has the largest absolute value, and the absolute value exceeds the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as oxidation failure. If the absolute value of the second correlation coefficient is the largest and exceeds the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as wear failure. If the third determination coefficient exceeds the second preset threshold, and the absolute values ​​of both the first correlation coefficient and the second correlation coefficient are lower than the first preset threshold, then the dominant failure mode of the probe under test is diagnosed as plastic deformation failure.

8. The method for simulating chip test probe detection and assessing lifetime according to claim 6, characterized in that, The lifetime assessment process for the probe under test specifically includes the following steps: Based on the dominant failure mode obtained from the diagnosis, the target data sequence is determined according to the corresponding discrete data sequence. For the selected target data sequence, a fitting function is obtained by performing a mathematical function fitting on the target data sequence, using the number of loops already executed as the independent variable; Calculate the value of the independent variable when the function value of the fitting function reaches the failure threshold preset for the current target data sequence, and use the value of the independent variable as the predicted lifetime cycle number; The predicted lifetime cycle count is used as the evaluated lifetime of the probe under test.

9. The method for simulating chip test probe detection and assessing lifetime according to claim 8, characterized in that, After obtaining the predicted lifetime cycle count, the following steps are also included: If the predicted lifetime cycle count is greater than the preset expected verification cycle count, the accelerated test is initiated. The accelerated testing is achieved by enhancing the corresponding test parameters based on the dominant failure mode; Specifically, if the dominant failure mode is oxidation failure, then the electrical parameters corresponding to the electrical stress are enhanced; If the dominant failure mode is wear failure or plastic deformation failure, then the motion parameters corresponding to the mechanical stress are enhanced.

10. The method for simulating chip test probe detection and assessing lifetime according to claim 8, characterized in that, After evaluating the lifetime of the probe under test, the following steps are also included: The predicted lifetime cycle count obtained from the assessment is compared with the preset qualified lifetime standard; If the predicted lifetime cycle count reaches the qualified lifetime standard, the probe is deemed qualified, and a test report containing the determination result and lifetime data is output. If the predicted lifetime cycle count is lower than the acceptable lifetime standard, the probe is deemed unqualified, and a test report containing the determination result, failure mode diagnostic information, and lifetime data is output.

Images