Method and device for detecting performance of batch MOS (Metal Oxide Semiconductor) transistors
By conducting standard factory testing, classification, and sampling tests on batches of MOSFETs, the problems of low accuracy and inefficiency in performance evaluation of batch MOSFETs were solved, achieving efficient and accurate performance evaluation.
Patent Information
- Application Number
- CN202511161635.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-19
- Publication Date
- 2025-11-11
AI Technical Summary
Existing technologies suffer from low accuracy and inefficiency in batch MOSFET performance evaluation, making it difficult to achieve efficient and accurate performance evaluation in large-scale production environments.
By performing standard factory tests on each MOSFET to obtain basic characteristic data, classifying them into multiple clusters, determining a reasonable sampling quantity for sampling, conducting power cycling tests, and comprehensively evaluating the performance of the entire batch of MOSFETs.
This enables efficient and accurate evaluation of the performance of batch MOSFETs, improves the representativeness and comprehensiveness of the evaluation results, and reduces testing costs and time.
Smart Images
Figure CN120928145A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of MOSFET testing, and in particular to a method and apparatus for testing the performance of batch MOSFETs. Background Technology
[0002] With the rapid development of the electronics market, the performance evaluation of MOSFETs, as core components of electronic devices, has become increasingly important in production and application. Especially in large-scale production environments, accurate evaluation of the performance of batches of MOSFETs is crucial to the reliability and stability of the entire batch of products.
[0003] Currently, the performance evaluation of batch MOSFETs mainly employs a general sampling inspection method. This method typically involves randomly selecting a certain number of samples from the entire batch of MOSFETs for testing, and then using the test results of these samples to evaluate the performance level of the entire batch. However, due to factors such as process fluctuations and material differences during the production process, the performance of MOSFETs exhibits a certain degree of dispersion, and simple random sampling methods often fail to comprehensively reflect the true performance level of the entire batch. Another common evaluation method is to conduct comprehensive and in-depth testing on each MOSFET. While this method can obtain relatively accurate performance data, it is time-consuming, costly, and inefficient, making it unsuitable for the needs of large-scale production environments. Therefore, existing technologies for evaluating the performance of batch MOSFETs suffer from low accuracy and insufficient efficiency. Summary of the Invention
[0004] This application provides a method and apparatus for testing the performance of batch MOSFETs, aiming to solve the technical problems of low accuracy and insufficient efficiency in evaluating the performance of batch MOSFETs in the prior art.
[0005] The first aspect disclosed in this application provides a performance testing method for batch MOSFETs. The method includes: performing standard factory testing on each MOSFET in the batch to obtain basic characteristic data for each MOSFET; classifying the batch MOSFETs according to the basic characteristic data to obtain multiple MOSFET clusters; determining the sampling quantity for each MOSFET cluster, and sampling within each MOSFET cluster based on the sampling quantity to obtain multiple groups of MOSFETs under test; performing power cycling tests on each MOSFET in the multiple groups of MOSFETs under test to obtain multiple groups of MOSFET performance test results; and determining the performance evaluation result of the batch MOSFETs based on the multiple groups of MOSFET performance test results.
[0006] Optionally, standard factory testing is performed on each MOSFET in the batch to obtain basic characteristic data of each MOSFET, including: obtaining the application requirements of the batch of MOSFETs, and determining the factory testing scheme of the batch of MOSFETs according to the application requirements, wherein the factory testing scheme includes multiple factory testing items; and testing each MOSFET in the batch of MOSFETs according to the multiple factory testing items to obtain basic characteristic data of each MOSFET.
[0007] Optionally, the batch of MOSFETs is classified according to the basic characteristic data of each MOSFET to obtain multiple MOSFET clusters, including: determining multiple device levels according to the application requirements, wherein each device level corresponds to a basic characteristic parameter range; after obtaining the basic characteristic data of any MOSFET, comparing the basic characteristic data with the basic characteristic parameter range of each device level, and classifying the MOSFET into the corresponding MOSFET cluster according to the comparison result; classifying each MOSFET in the batch of MOSFETs to obtain multiple MOSFET clusters.
[0008] Optionally, the sampling quantity of each MOS transistor cluster is determined, and sampling is performed in each MOS transistor cluster based on the sampling quantity to obtain multiple groups of MOS transistors to be tested, including: counting the total number of MOS transistors in the batch and obtaining the number of MOS transistor clusters in each MOS transistor cluster; determining the cluster ratio of each MOS transistor cluster based on the number of MOS transistor clusters and the total number of MOS transistors; obtaining a preset sampling quantity of MOS transistors, and obtaining the sampling quantity of each MOS transistor cluster based on the preset sampling quantity of MOS transistors and the cluster ratio of each cluster.
[0009] Optionally, a power cycling test is performed on each of the multiple groups of MOSFETs under test to obtain multiple sets of MOSFET performance test results, including: setting a threshold for the number of cycles, a threshold for saturation on-state voltage drop, and a threshold for thermal resistance in the power cycling test; performing a power cycling test on the first MOSFET under test and recording the temperature change and electrical parameters of the MOSFET in each cycle; ending the power cycling test when the number of cycles reaches the threshold for the number of cycles, or the saturation on-state voltage drop of the first MOSFET under test exceeds the threshold for saturation on-state voltage drop, or the thermal resistance of the first MOSFET under test exceeds the threshold for thermal resistance; determining a first performance test result for the first MOSFET under test based on the recorded temperature change and electrical parameters; and performing a power cycling test on each of the multiple groups of MOSFETs under test in the same manner as determining the first performance test result to obtain multiple sets of MOSFET performance test results.
[0010] Optionally, the performance evaluation results of the batch of MOSFETs are determined based on the performance test results of multiple groups of MOSFETs, including: performing statistical analysis on the performance test results of each group of MOSFETs to obtain the comprehensive test results of multiple cluster MOSFETs; weighting the comprehensive test results of multiple cluster MOSFETs based on the cluster ratio to obtain the comprehensive test results of the batch of MOSFETs; and generating the performance evaluation results of the batch of MOSFETs based on the comprehensive test results of the batch of MOSFETs.
[0011] Optionally, the method further includes: determining the performance index requirements of the batch of MOSFETs according to the application requirements; comparing each performance index in the performance evaluation results of the batch of MOSFETs with the performance index requirements; and triggering a production line alarm when any performance index of the batch of MOSFETs fails to meet the corresponding index requirements.
[0012] Another aspect of this application discloses a performance testing device for batch MOSFETs, comprising: a basic testing unit for performing standard factory testing on each MOSFET in the batch to obtain basic characteristic data of each MOSFET; a classification processing unit for classifying the batch MOSFETs according to the basic characteristic data of each MOSFET to obtain multiple MOSFET clusters; a sampling determination unit for determining the sampling quantity of each MOSFET cluster and sampling in each MOSFET cluster based on the sampling quantity to obtain multiple groups of MOSFETs to be tested; a cycle testing unit for performing power cycle tests on each MOSFET in the multiple groups of MOSFETs to be tested to obtain multiple groups of MOSFET performance test results; and a performance evaluation unit for determining the performance evaluation result of the batch MOSFETs based on the multiple groups of MOSFET performance test results.
[0013] By performing standard factory testing on each MOSFET in a batch, basic characteristic data of each MOSFET is obtained, providing a data foundation for subsequent classification. Based on this basic characteristic data, the batch of MOSFETs is classified into multiple MOSFET clusters. MOSFETs with similar performance characteristics are grouped together, avoiding the accuracy reduction caused by mixed evaluation of MOSFETs with different characteristics and improving the representativeness of subsequent sampling. The sampling quantity for each MOSFET cluster is determined, and sampling is performed within each cluster based on this quantity to obtain multiple sets of MOSFETs to be tested. By determining a reasonable sampling quantity for each cluster and extracting samples from them, a targeted sampling strategy is implemented, ensuring that corresponding samples of each type of MOSFET are included in the testing, enhancing the representativeness and comprehensiveness of the sampling. Power cycling tests are performed on each MOSFET in the multiple sets of MOSFETs to be tested, obtaining performance test results for multiple sets of MOSFETs. The power cycling test method is used to conduct in-depth testing of the sampled MOSFETs, simulating performance changes under actual operating conditions, obtaining performance data of each cluster of MOSFETs under high load conditions, improving the practicality of the test results. Based on the performance test results of multiple groups of MOSFETs, the performance evaluation results of the batch of MOSFETs are determined. By comprehensively analyzing the test results of each cluster, a performance evaluation of the entire batch of MOSFETs is formed, which not only ensures the comprehensiveness of the evaluation, but also improves the evaluation efficiency, and at the same time ensures the accuracy of the evaluation results.
[0014] The above technical solution enables efficient and accurate evaluation of the performance of batch MOSFETs, solving the technical problems of low evaluation accuracy and insufficient efficiency in the existing technology. Attached Figure Description
[0015] Figure 1 This application provides a flowchart illustrating a method for testing the performance of batch MOSFETs. Figure 2 This application provides a schematic diagram of the structure of a performance testing device for batch MOSFETs.
[0016] Explanation of reference numerals in the attached figures: 11. Basic testing unit; 12. Classification and processing unit; 13. Sampling determination unit; 14. Cyclic testing unit; 15. Performance evaluation unit. Detailed Implementation
[0017] The overall concept of the technical solution provided in this application is as follows: This application provides a method and apparatus for performance testing of batch MOSFETs. Specifically, the method first performs standard factory testing on the batch of MOSFETs to obtain basic characteristic data. Then, based on this data, the MOSFETs are classified into multiple clusters with similar characteristics. Next, a reasonable sampling quantity is determined for each cluster, and samples are taken. Power cycling tests are then performed on the sampled samples. Finally, the performance evaluation results of the entire batch of MOSFETs are determined by combining multiple test results. By fully considering the characteristic differences between MOSFETs, MOSFETs with similar performance characteristics are grouped into the same cluster and evaluated separately, avoiding the accuracy reduction problem caused by mixed evaluation of MOSFETs with different characteristics. Simultaneously, by determining the sampling quantity for each cluster in a targeted manner, both the representativeness and comprehensiveness of the sampling are ensured, while avoiding the inefficiency problem of full inspection. Furthermore, the power cycling test method is used to conduct in-depth testing on the sampled MOSFETs, effectively simulating performance changes under actual working conditions, enhancing the practicality and accuracy of the test results. Through this systematic and hierarchical testing method, efficient and accurate performance evaluation of batch MOSFETs is achieved, solving the technical problems of low evaluation accuracy and insufficient efficiency in the prior art.
[0018] After introducing the basic principles of this application, various non-limiting embodiments of this application will be described in detail below with reference to the accompanying drawings.
[0019] Example 1, as Figure 1 As shown in the figure, this application provides a method for performance testing of batch MOSFETs, including: S100: Perform standard factory testing on each MOSFET in the batch to obtain basic characteristic data of each MOSFET.
[0020] Specifically, standardized factory testing is performed on each MOSFET device in the batch manufactured on the production line. This testing includes, but is not limited to, measuring electrical parameters such as threshold voltage, drain-source saturation voltage, drain-source breakdown voltage, gate-source breakdown voltage, leakage current, and transient response characteristics. Using specialized testing equipment, such as curve tracers and semiconductor parameter analyzers, preset test conditions are applied to each MOSFET, and its response characteristics are recorded to obtain the fundamental characteristic data of each MOSFET.
[0021] The acquired basic characteristic data not only reflects whether the MOSFETs are functioning correctly, but more importantly, this data forms the basis for subsequent classification and evaluation of batch MOSFETs. By acquiring the basic characteristic data of each MOSFET, a reference basis is provided for more in-depth performance evaluation.
[0022] S200: Classify batches of MOSFETs based on their basic characteristic data to obtain multiple MOSFET clusters.
[0023] Specifically, after obtaining the basic characteristic data of each MOSFET, the batch of MOSFETs is classified based on this data. The classification process groups MOSFETs with similar performance characteristics into the same cluster based on the similarity or difference in their basic characteristic data. Specifically, key electrical parameters such as threshold voltage, on-resistance, and breakdown voltage are selected as classification criteria, and corresponding parameter ranges are set. Then, MOSFETs whose parameters fall within the same range are grouped into one cluster, thus obtaining multiple MOSFET clusters.
[0024] By classifying the batch of MOSFETs, they are divided into multiple different MOSFET clusters. The MOSFETs in each cluster have similar electrical characteristics, which can more accurately reflect the performance distribution of the batch of MOSFETs and provide a more targeted data basis for subsequent sampling tests and performance evaluations.
[0025] S300: Determine the sampling quantity for each MOS transistor cluster, and sample from each MOS transistor cluster based on the sampling quantity to obtain multiple sets of MOS transistors to be tested.
[0026] Specifically, after completing the batch classification of MOSFETs and obtaining multiple MOSFET clusters, the number of MOSFET samples to be extracted from each cluster is determined, and sampling is performed accordingly.
[0027] First, for each MOSFET cluster, the sampling size for each cluster is determined based on its size, importance, and other information. After determining the sampling size, a corresponding number of MOSFETs are randomly selected from each cluster as test samples. Through cluster-based sampling, multiple groups of MOSFETs can be obtained, each corresponding to a specific MOSFET cluster. These MOSFETs will be used for subsequent performance testing. This sampling method ensures that the selected samples are representative of the characteristics of different clusters, improving the accuracy and reliability of subsequent performance evaluation results, while also avoiding the waste of resources associated with testing all MOSFETs.
[0028] S400: Performs power cycling tests on each MOSFET in multiple groups of MOSFETs under test to obtain the performance test results of multiple groups of MOSFETs.
[0029] Specifically, after sampling each MOSFET cluster and obtaining multiple groups of MOSFETs under test, more in-depth power cycling tests are conducted on these samples to evaluate their actual performance.
[0030] The power cycling test is a reliability testing method that simulates the repeated switching operations of a MOSFET under actual operating conditions. It applies repeated current and voltage stresses to the MOSFET, causing it to cycle between on and off states, and subjecting it to periodic temperature changes during this process. For each MOSFET under test, the changes in its electrical parameters and temperature response characteristics are recorded during the power cycling test, until a preset number of cycles is reached or a parameter of the MOSFET exceeds the allowable range.
[0031] By conducting power cycling tests on each MOSFET in multiple groups of MOSFETs under test, the performance test results of multiple groups of MOSFETs are obtained. These results can comprehensively reflect the reliability and stability of different clusters of MOSFETs under long-term operating conditions, providing data support for the performance evaluation of batch MOSFETs.
[0032] S500: Based on the performance test results of multiple sets of MOSFETs, determine the performance evaluation results of the batch of MOSFETs.
[0033] Specifically, after completing power cycle tests on multiple groups of MOSFETs under test and obtaining the corresponding performance test results, these test data are comprehensively analyzed and evaluated to obtain a performance evaluation of the entire batch of MOSFETs.
[0034] First, the performance test results of each group of MOSFETs are statistically processed to analyze the performance characteristics of each cluster of MOSFETs in power cycling tests, such as parameter drift, failure rate, and average lifetime. Then, considering the distribution ratio of each MOSFET cluster in the entire batch of MOSFETs, the test results of different clusters are weighted and calculated to obtain performance evaluation results that can represent the performance level of the entire batch of MOSFETs.
[0035] By using sampling tests and statistical analysis, the overall performance and reliability of batch MOSFETs can be effectively evaluated, providing an important basis for production quality control and product application.
[0036] Furthermore, standard factory testing is performed on each MOSFET in the batch to obtain basic characteristic data for each MOSFET, including: S110: Obtain the application requirements of the batch of MOSFETs, and determine the factory testing scheme for the batch of MOSFETs based on the application requirements, wherein the factory testing scheme includes multiple factory testing items; S120: Each MOSFET in the batch is tested according to multiple factory testing items to obtain the basic characteristic data of each MOSFET.
[0037] In one feasible implementation, when performing standard factory testing on each MOSFET in a batch to obtain its basic characteristic data, the application requirements for the batch of MOSFETs are first determined, and a factory testing plan is constructed based on these specific application requirements. The factory testing plan includes multiple test items set according to the application requirements of the batch of MOSFETs, such as detailed test specifications for electrical parameters like gate threshold voltage, drain-source breakdown voltage, on-resistance, and switching time characteristics. This pre-planning approach ensures that subsequent testing accurately corresponds to the performance requirements of the MOSFETs in the actual application environment.
[0038] After the factory testing plan was established, comprehensive and standardized testing was conducted on each MOSFET device produced in batches, according to the established multiple factory testing items. During this process, professional testing instruments such as semiconductor parameter analyzers were used to accurately measure and systematically record various parameters of each MOSFET, ultimately obtaining a series of fundamental characteristic data reflecting the basic electrical characteristics of the MOSFET. This comprehensive fundamental characteristic data will serve as the objective basis and data foundation for subsequent MOSFET performance classification and in-depth evaluation.
[0039] Furthermore, based on the basic characteristic data of each MOSFET, the batch of MOSFETs are classified to obtain multiple MOSFET clusters, including: S210: Determine multiple device levels according to the application requirements, wherein each device level corresponds to a range of basic characteristic parameters; S220: After obtaining the basic characteristic data of any MOSFET, compare the basic characteristic data with the basic characteristic parameter range of each device level, and classify the MOSFET into the corresponding MOSFET cluster according to the comparison result. S230: Each MOS transistor in the batch is classified into multiple MOS transistor clusters.
[0040] In one feasible implementation, when acquiring multiple MOSFET clusters, firstly, several clear device grade standards are established based on the application requirements of the batch of MOSFETs. These device grades are divided according to product performance requirements and quality control requirements, and each grade corresponds to specific ranges of basic characteristic parameters, such as threshold voltage range, on-resistance range, and breakdown voltage range. This application-demand-based grading method ensures that the classification standards are closely related to the actual usage environment and performance expectations of the MOSFETs, improving the practicality and relevance of the classification.
[0041] After acquiring the basic characteristic data of any MOSFET, a parameter comparison process is performed, which involves systematically comparing the measured basic characteristic data of the MOSFET with the determined basic characteristic parameter ranges corresponding to each device level. Through data comparison and analysis, it is determined which level of parameter range the MOSFET's performance parameters fall into, and accordingly, the MOSFET is classified into the corresponding MOSFET cluster. This parameter comparison-based classification method ensures the objectivity and accuracy of the classification process.
[0042] Then, the classification process in step S220 is repeated for each MOSFET in the batch to ensure that all MOSFETs are accurately assigned to the appropriate cluster. Through a comprehensive and systematic classification process, multiple MOSFET clusters with different performance characteristics are ultimately formed, with MOSFETs in each cluster exhibiting similar characteristics in key electrical parameters. This clustered classification provides a structured data foundation for subsequent sampling tests and performance evaluations.
[0043] The MOSFET classification process enables accurate classification of batches of MOSFETs, creating favorable conditions for subsequent cluster-based sampling tests and comprehensive performance evaluation.
[0044] Furthermore, the sampling quantity for each MOS transistor cluster is determined, and sampling is performed in each MOS transistor cluster based on the sampling quantity to obtain multiple sets of MOS transistors under test, including: S310: Count the total number of MOSFETs in the batch and obtain the number of MOSFET clusters in each MOSFET cluster; S320: Determine the cluster ratio of each MOSFET cluster based on the number of MOSFETs in each cluster and the total number of MOSFETs; S330: Obtain the preset MOS transistor sampling quantity, and based on the preset MOS transistor sampling quantity and the ratio of each cluster, obtain the sampling quantity of each MOS transistor cluster.
[0045] In a preferred embodiment, when acquiring multiple groups of MOSFETs to be tested, firstly, a comprehensive statistical analysis of the batch-produced MOSFETs is performed to determine the total number of MOSFETs in the batch. At the same time, each MOSFET cluster is counted to obtain the number of MOSFETs contained in each MOSFET cluster. This provides the necessary data basis for subsequent proportion calculations and sampling decisions, ensuring that the sampling work is based on an accurate grasp of the overall quantity distribution.
[0046] Then, based on the obtained number of each MOSFET cluster and the total number of MOSFETs in the batch, mathematical calculations are performed to determine the proportion of each MOSFET cluster in the entire batch. Specifically, the proportion of each cluster is obtained by dividing the number of each MOSFET cluster by the total number of MOSFETs. These cluster proportion data objectively reflect the distribution of MOSFETs with different performance characteristics in the batch of products, providing a basis for subsequent sampling quantity allocation.
[0047] Next, a pre-set total number of MOSFETs to be sampled is obtained, which is typically determined based on statistical principles and actual testing resource considerations. Then, the pre-set number of MOSFETs to be sampled is multiplied by the calculated proportion of each cluster, thereby allocating the specific number of samples to be drawn from each MOSFET cluster proportionally. This method of determining the sampling number based on cluster proportions ensures the representativeness of the sampling test and the rationality of resource utilization.
[0048] By using the above sampling quantity determination process, reasonable sampling of batch MOSFETs is achieved, ensuring that the sampled test samples can accurately reflect the performance distribution characteristics of the entire batch of MOSFETs, thus laying a sample foundation for subsequent performance testing.
[0049] Furthermore, power cycling tests were performed on each MOSFET in multiple groups of test MOSFETs to obtain performance test results for multiple groups of MOSFETs, including: S410: Set the threshold for the number of cycles, the threshold for saturation conduction voltage drop, and the threshold for thermal resistance in the power cycling test; S420: Perform a power cycle test on the first MOSFET under test, and record the temperature change and electrical parameters of the MOSFET in each cycle; S430: When the number of cycles reaches the cycle number threshold, or the saturation on-state voltage drop of the first MOSFET under test exceeds the saturation on-state voltage drop threshold, or the thermal resistance of the first MOSFET under test exceeds the thermal resistance threshold, the power cycle test ends. S440: Based on the recorded temperature changes and electrical parameters, determine the first performance test result of the first MOSFET under test; S450: Following the method for determining the first performance test result, power cycling tests are performed on each of the multiple groups of MOSFETs under test to obtain the performance test results of multiple groups of MOSFETs.
[0050] In a preferred embodiment, firstly, several key control parameters for the power cycling test are set, including a cycle count threshold, a saturation on-state voltage drop threshold, and a thermal resistance threshold. The cycle count threshold is typically set to tens of thousands to hundreds of thousands of cycles to evaluate the long-term reliability of the MOSFET; the saturation on-state voltage drop threshold is generally set to approximately 105% of the initial value as a criterion for judging MOSFET performance degradation; and the thermal resistance threshold is often set to approximately 120% of the initial value to monitor changes in the MOSFET's thermal performance. These three threshold parameters together constitute the termination condition and evaluation criteria for the power cycling test.
[0051] Then, a power cycle test is performed on the first MOSFET under test. This test involves applying periodic current and temperature stress to the MOSFET, and recording the temperature changes and electrical parameters of the MOSFET in detail during each cycle. Here, "first MOSFET under test" refers to any MOSFET under test. Specifically, the gate voltage of the control device is first set to 15V to bring the first MOSFET under test into saturation conduction. Then, a heating current switch is turned on to allow current to flow through the first MOSFET under test, causing its temperature to rise. During this period, the saturation conduction voltage drop of the first MOSFET under test is monitored by a voltage detection circuit. When the temperature reaches a certain level, the heating current switch is turned off, and the first MOSFET under test enters water cooling mode (this moment is recorded as t0). Simultaneously, the gate voltage is adjusted to 0V to bring the first MOSFET under test into the off state. At this time, the test current flows in reverse through the body diode of the first MOSFET under test. The body diode voltage drop of the first MOSFET under test is measured at two moments, t1 and t2. The device temperatures T1 and T2 at these two moments are calculated using a pre-calibrated relationship between the body diode voltage drop and temperature, and then the temperature T0 at moment t0 is deduced as the maximum cycle temperature. After a period of heat dissipation, the body diode voltage drop is measured again to calculate the minimum cycle temperature, thereby obtaining key parameters such as the device junction temperature difference and thermal resistance. Simultaneously, three key indicators are continuously monitored during the power cycling test: the number of cycles, the saturation on-state voltage drop of the MOSFET, and the thermal resistance. The power cycling test is terminated when one of the following three conditions is met: the number of cycles reaches a preset threshold, indicating that the device has completed its predetermined lifespan test; or the saturation on-state voltage drop of the first MOSFET under test exceeds a preset threshold, indicating that the device's electrical performance has significantly deteriorated; or the thermal resistance of the first MOSFET under test exceeds a preset threshold, indicating that the device's thermal performance has significantly worsened. This multi-condition termination mechanism ensures the effectiveness of the test and the rational use of resources.
[0052] Then, based on all temperature change data and electrical parameter data recorded during the power cycling test, the comprehensive performance test results of the first MOSFET under test were determined through analysis and calculation. These results typically include multiple indicators such as the MOSFET's thermal stability, electrical parameter drift, estimated lifetime, and failure modes, comprehensively reflecting the MOSFET's performance and reliability level under long-term operating conditions. Following the same test methods and evaluation procedures as the first MOSFET under test, power cycling tests were sequentially conducted on each MOSFET in multiple groups of test MOSFETs, and the corresponding performance test results were obtained. This standardized testing method ensured the comparability and consistency of the test results for each MOSFET, ultimately obtaining multiple sets of statistically significant MOSFET performance test results, providing comprehensive and objective data support for subsequent comprehensive evaluation.
[0053] Through power cycling tests, in-depth testing and evaluation of the performance of multiple groups of MOSFETs were achieved, providing reliable experimental data for the final determination of the performance evaluation results of batch MOSFETs.
[0054] Furthermore, based on the performance test results of multiple sets of MOSFETs, the performance evaluation results of the batch of MOSFETs are determined, including: S510: Perform statistical analysis on the performance test results of each group of MOSFETs to obtain the comprehensive test results of multiple clusters of MOSFETs; S520: Based on the cluster ratio, the comprehensive test results of multiple cluster MOS transistors are weighted to obtain the comprehensive test results of batch MOS transistors; S530: Based on the comprehensive test results of the batch of MOSFETs, generate the performance evaluation results of the batch of MOSFETs.
[0055] In a preferred embodiment, firstly, statistical analysis is performed on the performance test results of each group of MOSFETs to obtain comprehensive test results for multiple clusters of MOSFETs. Specifically, for the obtained performance test results of multiple groups of MOSFETs, systematic statistical analysis is performed according to the cluster to which the MOSFETs belong. The analysis includes, but is not limited to, calculating key indicators such as the mean failure time, parameter drift rate, and thermal performance change trend of the MOSFETs in each cluster, and synthesizing these statistical data to derive the overall performance and reliability level of the MOSFETs in each cluster. This cluster-based statistical analysis method can accurately reflect the differences in test results of MOSFETs with different performance characteristics, providing a clearly categorized data foundation for subsequent weighted processing.
[0056] Then, based on the cluster ratio, the comprehensive test results of multiple clusters of MOSFETs are weighted to obtain the comprehensive test results of the batch of MOSFETs. Specifically, the proportion of each MOSFET cluster in the entire batch of MOSFETs is used as a weighting coefficient to calculate the comprehensive test results of each cluster. In practice, the comprehensive test result of each cluster is multiplied by its corresponding cluster ratio, and then all weighted results are summed to obtain the comprehensive test result that represents the performance level of the entire batch of MOSFETs, i.e., the comprehensive test result of the batch of MOSFETs. By using a weighting method based on cluster ratio, it is ensured that the final comprehensive test result accurately reflects the overall performance distribution characteristics of the batch of MOSFETs.
[0057] Subsequently, based on the comprehensive test results of the batch of MOSFETs, performance evaluation results for the batch of MOSFETs are generated. Specifically, based on the obtained comprehensive test results of the batch of MOSFETs, combined with product quality standards and application requirements, the performance of the batch of MOSFETs is comprehensively evaluated, and formal performance evaluation results are formed. These performance evaluation results include key evaluation indicators such as the average reliability level, expected lifespan, and performance stability of the batch of MOSFETs, as well as a comprehensive judgment on the overall quality of the batch of MOSFETs, providing a scientific basis for product quality control and application decisions.
[0058] The performance evaluation results confirmed a comprehensive assessment of the performance of batch MOSFETs, ensuring the scientific rigor, accuracy, and reliability of the evaluation results, and providing strong support for the quality management and application of batch MOSFETs.
[0059] Furthermore, embodiments of this application also include: S610: Determine the performance requirements of the batch of MOS transistors according to the application requirements; S620: Compare the performance indicators in the performance evaluation results of the batch of MOSFETs with the performance indicator requirements. If any performance indicator of the batch of MOSFETs fails to meet the corresponding indicator requirements, trigger an alarm on the production line.
[0060] In a preferred embodiment, firstly, the performance requirements for the batch of MOSFETs are determined based on application requirements. Specifically, based on the actual application scenarios and operating environments of the batch of MOSFETs, a series of clear performance requirements are established, such as reliability requirements, lifespan requirements, and stability requirements. These performance requirements are typically given in the form of specific numerical ranges or thresholds, serving as objective standards for judging whether the performance of the batch of MOSFETs meets application requirements. This method of setting performance indicators based on application requirements ensures the goal-oriented and practical nature of performance evaluation, effectively avoiding the problem of evaluation results being out of touch with actual needs.
[0061] Then, the performance indicators in the batch MOSFET performance evaluation results are compared with the performance requirements. If any performance indicator of the batch MOSFETs fails to meet the corresponding requirement, a production line alarm is triggered. Specifically, each specific indicator in the obtained batch MOSFET performance evaluation results is systematically compared one by one with the determined performance requirements. When any performance indicator of the batch MOSFETs fails to meet the corresponding requirement, the production line alarm mechanism is immediately triggered, and relevant personnel are promptly notified for intervention. This real-time monitoring and early warning mechanism can detect potential quality problems of batch MOSFETs at the first moment, preventing unqualified products from flowing into subsequent stages, effectively improving the timeliness and effectiveness of quality control.
[0062] By triggering a production line alarm when any performance indicator of a batch of MOSFETs fails to meet the corresponding requirements, the production line performance of batch MOSFETs is monitored, providing strong support for quality control in the production process.
[0063] Example 2, based on the same inventive concept as the batch MOS transistor performance testing method in the foregoing examples, such as... Figure 2 As shown in the figure, this application provides a performance testing device for batch MOSFETs, the device comprising: The basic testing unit 11 is used to perform standard factory testing on each MOSFET in the batch and obtain basic characteristic data of each MOSFET. The classification processing unit 12 is used to classify batch MOSFETs according to the basic characteristic data of each MOSFET and obtain multiple MOSFET clusters; The sampling determination unit 13 is used to determine the sampling quantity of each MOS transistor cluster, and to perform sampling in each MOS transistor cluster based on the sampling quantity to obtain multiple sets of MOS transistors to be tested. The cycle test unit 14 is used to perform power cycle tests on each MOSFET in multiple groups of MOSFETs under test and obtain the performance test results of multiple groups of MOSFETs. The performance evaluation unit 15 is used to determine the performance evaluation results of the batch of MOSFETs based on the performance test results of multiple groups of MOSFETs.
[0064] Furthermore, the execution steps of the basic detection unit 11 include: Obtain the application requirements of the batch of MOSFETs, and determine the factory testing scheme for the batch of MOSFETs based on the application requirements, wherein the factory testing scheme includes multiple factory testing items; Each MOSFET in the batch is tested according to multiple factory testing items to obtain the basic characteristic data of each MOSFET.
[0065] Furthermore, the execution steps of the classification processing unit 12 include: Multiple device levels are determined based on the application requirements, and each device level corresponds to a range of basic characteristic parameters. After obtaining the basic characteristic data of any MOSFET, the basic characteristic data is compared with the basic characteristic parameter range of each device level, and the MOSFET is classified into the corresponding MOSFET cluster according to the comparison result. Each MOSFET in the batch is categorized to obtain multiple MOSFET clusters.
[0066] Furthermore, the execution steps of sampling determination unit 13 include: Count the total number of MOSFETs in the batch and obtain the number of MOSFET clusters in each MOSFET cluster; The cluster ratio of each MOSFET cluster is determined based on the number of MOSFETs in each cluster and the total number of MOSFETs. Obtain the preset number of MOS transistor samples, and based on the preset number of MOS transistor samples and the proportion of each cluster, obtain the number of samples for each MOS transistor cluster.
[0067] Furthermore, the execution steps of the loop test unit 14 include: Set the threshold values for the number of cycles, saturation voltage drop, and thermal resistance in the power cycling test; A power cycling test was performed on the first MOSFET under test, and the temperature change and electrical parameters of the MOSFET were recorded in each cycle. The power cycling test ends when the number of cycles reaches the cycle number threshold, or the saturation on-state voltage drop of the first MOSFET under test exceeds the saturation on-state voltage drop threshold, or the thermal resistance of the first MOSFET under test exceeds the thermal resistance threshold. Based on the recorded temperature changes and electrical parameters, the first performance test results of the first MOSFET under test are determined. Following the method for determining the first performance test result, power cycling tests were performed on each of the multiple groups of MOSFETs under test to obtain the performance test results of multiple groups of MOSFETs.
[0068] Furthermore, the execution steps of the performance evaluation unit 15 include: Statistical analysis was performed on the performance test results of each group of MOSFETs to obtain the comprehensive test results of multiple clusters of MOSFETs. Based on the cluster ratio, the comprehensive test results of multiple cluster MOSFETs are weighted to obtain the comprehensive test results of batch MOSFETs. Based on the comprehensive test results of the batch of MOSFETs, the performance evaluation results of the batch of MOSFETs are generated.
[0069] Furthermore, embodiments of this application also include an alarm triggering unit, the execution steps of which include: Determine the performance requirements for the batch of MOS transistors based on the application requirements; The performance indicators in the performance evaluation results of the batch of MOSFETs are compared with the performance requirements. When any performance indicator of the batch of MOSFETs fails to meet the corresponding requirement, an alarm is triggered on the production line.
[0070] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. A method for performance testing of batch MOSFETs, characterized in that, include: Perform standard factory testing on each MOSFET in the batch to obtain the basic characteristic data of each MOSFET; Based on the basic characteristic data of each MOSFET, the batch of MOSFETs are classified to obtain multiple MOSFET clusters; Determine the sampling quantity for each MOS transistor cluster, and sample from each MOS transistor cluster based on the sampling quantity to obtain multiple sets of MOS transistors under test; Power cycling tests were performed on each MOSFET in multiple groups of MOSFETs under test to obtain the performance test results of multiple groups of MOSFETs. Based on the performance test results of multiple sets of MOSFETs, the performance evaluation results of the batch of MOSFETs are determined.
2. The performance testing method for batch MOSFETs according to claim 1, characterized in that, Perform standard factory testing on each MOSFET in the batch to obtain basic characteristic data for each MOSFET, including: Obtain the application requirements of the batch of MOSFETs, and determine the factory testing scheme for the batch of MOSFETs based on the application requirements, wherein the factory testing scheme includes multiple factory testing items; Each MOSFET in the batch is tested according to multiple factory testing items to obtain the basic characteristic data of each MOSFET.
3. The performance testing method for batch MOSFETs according to claim 2, characterized in that, Based on the fundamental characteristic data of each MOSFET, the batch of MOSFETs are classified to obtain multiple MOSFET clusters, including: Multiple device levels are determined based on the application requirements, and each device level corresponds to a range of basic characteristic parameters. After obtaining the basic characteristic data of any MOSFET, the basic characteristic data is compared with the basic characteristic parameter range of each device level, and the MOSFET is classified into the corresponding MOSFET cluster according to the comparison result. Each MOSFET in the batch is categorized to obtain multiple MOSFET clusters.
4. The performance testing method for batch MOSFETs according to claim 1, characterized in that, Determine the sampling quantity for each MOSFET cluster, and sample from each MOSFET cluster based on the sampling quantity to obtain multiple sets of MOSFETs under test, including: Count the total number of MOSFETs in the batch and obtain the number of MOSFET clusters in each MOSFET cluster; The cluster ratio of each MOSFET cluster is determined based on the number of MOSFETs in each cluster and the total number of MOSFETs. Obtain the preset number of MOS transistor samples, and based on the preset number of MOS transistor samples and the proportion of each cluster, obtain the number of samples for each MOS transistor cluster.
5. The performance testing method for batch MOSFETs according to claim 1, characterized in that, Power cycling tests were performed on each MOSFET in multiple groups of test MOSFETs to obtain performance test results for multiple groups of MOSFETs, including: Set the threshold values for the number of cycles, saturation voltage drop, and thermal resistance in the power cycling test; A power cycling test was performed on the first MOSFET under test, and the temperature change and electrical parameters of the MOSFET were recorded in each cycle. The power cycling test ends when the number of cycles reaches the cycle number threshold, or the saturation on-state voltage drop of the first MOSFET under test exceeds the saturation on-state voltage drop threshold, or the thermal resistance of the first MOSFET under test exceeds the thermal resistance threshold. Based on the recorded temperature changes and electrical parameters, the first performance test results of the first MOSFET under test are determined. Following the method for determining the first performance test result, power cycling tests were performed on each of the multiple groups of MOSFETs under test to obtain the performance test results of multiple groups of MOSFETs.
6. The performance testing method for batch MOSFETs according to claim 4, characterized in that, Based on the performance test results of multiple sets of MOSFETs, the performance evaluation results of the batch of MOSFETs are determined, including: Statistical analysis was performed on the performance test results of each group of MOSFETs to obtain the comprehensive test results of multiple clusters of MOSFETs. Based on the cluster ratio, the comprehensive test results of multiple cluster MOSFETs are weighted to obtain the comprehensive test results of batch MOSFETs. Based on the comprehensive test results of the batch of MOSFETs, the performance evaluation results of the batch of MOSFETs are generated.
7. The performance testing method for batch MOSFETs according to claim 2, characterized in that, Also includes: The performance requirements for the batch of MOS transistors are determined based on the application requirements. The performance indicators in the performance evaluation results of the batch of MOSFETs are compared with the performance requirements. When any performance indicator of the batch of MOSFETs fails to meet the corresponding requirement, an alarm is triggered on the production line.
8. A performance testing device for batch MOSFETs, characterized in that, A method for testing the performance of batch MOSFETs according to any one of claims 1-7 includes: The basic testing unit is used to perform standard factory testing on each MOSFET in a batch of MOSFETs and obtain basic characteristic data of each MOSFET. The classification processing unit is used to classify batches of MOSFETs based on the basic characteristic data of each MOSFET, and obtain multiple MOSFET clusters; The sampling determination unit is used to determine the sampling quantity of each MOS transistor cluster, and to perform sampling in each MOS transistor cluster based on the sampling quantity to obtain multiple sets of MOS transistors to be tested. The cycle test unit is used to perform power cycle tests on each MOSFET in multiple groups of MOSFETs under test, and obtain the performance test results of multiple groups of MOSFETs. The performance evaluation unit is used to determine the performance evaluation results of the batch of MOSFETs based on the performance test results of multiple sets of MOSFETs.