A high-precision semiconductor testing and sorting method and device

Abstract

The application discloses a high-precision semiconductor test sorting method and equipment, and belongs to the technical field of semiconductor testing, and comprises the following steps: executing a test process of a target chip, recording a target marking chip which is qualified in the test as a target marking chip, analyzing a laser marking energy fluctuation characteristic parameter, determining a first execution path of a laser marking system, combining the laser marking energy fluctuation characteristic parameter to execute adaptive adjustment of the laser marking system, analyzing a second execution path of the laser marking system after the adjustment is completed, analyzing a laser marking carbonization characteristic parameter, and determining a third execution path of the laser marking system. The application can realize real-time monitoring and dynamic adjustment of energy fluctuation and carbonization quality in the laser marking process, not only improves the response speed and regulation and control precision of marking abnormalities, but also effectively avoids the misjudgment or good product NG problem caused by laser parameter drift, so that the stability, yield and automation level of the whole machine test sorting process are significantly enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor testing, in particular to a high-precision semiconductor testing and sorting method and device. BACKGROUND

[0002] During the packaging and testing of semiconductor chips, automated equipment is often used to complete the functional testing and sorting process. The existing chip testing and sorting process generally includes the following steps: feeding, automatic grabbing, functional testing, defect screening, classification feeding, laser marking, visual recognition confirmation, and subsequent tape packaging.

[0003] In practical applications, chips are usually fed by a vibrating disc or a linear track mechanism, and are then transferred one by one to the testing station by a distribution mechanism. The testing process mainly includes pin electrical property detection, package shape and size detection, and chip MARK recognition in visual recognition. The testing data is often collected in real time by the control system and compared with the product qualification standard. According to the test results, the chips are sent to the sorting module, different types of unqualified products (NG) are classified and fed into different containers, and qualified chips enter the marking stage. In the marking stage, a laser marking system is often used to mark the model, batch, and other information on the surface of the chip. A CCD vision module is also integrated to detect the integrity and clarity of the marked characters. If it is determined that the marking is NG, the chip will be rejected. After completing the marking and detection, the chips are usually packaged by a plastic encapsulation assembly or a tape system for final packaging.

[0004] For example, the Chinese invention patent with the publication number CN114690024B discloses a full-automatic chip testing machine, which includes a first machine table, a second machine table, and a third machine table. The second and third machine tables are parallelly arranged and vertically connected to the first machine table. Each of the second and third machine tables is provided with a mounting bracket, and the mounting bracket is provided with a plurality of testing mechanism installation layers. Each testing mechanism installation layer is provided with a plurality of testing mechanisms. The corresponding sides of the second and third machine tables are provided with transfer feeding and transplanting mechanical hands.

[0005] For example, the Chinese invention patent with the publication number CN118275863B discloses a chip placement tray, a manufacturing method, and a chip-level testing method. The chip is placed on the chip placement tray by using the sorting machine function, and finally tested by the testing machine.

[0006] The above-mentioned technology at least has the following technical problems:

[0007] The existing semiconductor test sorting method focuses on the automatic detection and NG sorting process of the semiconductor chip in the aspects of electrical performance, pin structure, appearance, etc., but often ignores the deep influence of laser marking as a subsequent key identification step on the yield of the final product. After the laser operates continuously for a long time or performs a high-frequency marking task, due to the thermal field disturbance inside the laser cavity, the response lag of temperature control and the micro-vibration of the cavity light path, the short-term instability of the output power is caused. Such power fluctuations can cause small differences in thermal input in the marking area. After the accumulation of the small differences, surface quality problems such as local ablation depth variation, inconsistent carbonization degree, blurred character edge burr, or residual particles of the marked pattern may occur. Especially in the continuous marking, high repetition rate application scenarios or the case where the thermal diffusion capacity of the material to be marked is poor, it is difficult for heat to be released in time, and local overheating is easily caused, such as slight surface scorching, dark spots or excessive carbonization. In severe cases, the pattern may be distorted or difficult to identify. Since the system lacks a dynamic monitoring and adjustment mechanism for real-time energy fluctuations and imaging carbonization characteristics during the marking process, these micro-defects cannot be corrected in time during the marking process, and ultimately the chips that are completely qualified in the test may be mistakenly rejected as NG products due to marking quality problems, directly affecting the machine test efficiency and product yield. SUMMARY

[0008] To solve the above technical problems existing in the prior art, the embodiment of the present application provides a high-precision semiconductor test sorting method and equipment. The technical scheme is as follows:

[0009] On the one hand, a high-precision semiconductor test sorting method is provided, comprising the following steps:

[0010] The semiconductor chip to be tested is recorded as a target chip, the test process of the target chip is performed, and the target chip that passes the test is recorded as a target marking chip.

[0011] The laser marking process of the target marking chip is performed and monitored in real time, the laser marking energy fluctuation characteristic parameter is analyzed, and the first execution path of the laser marking system is determined.

[0012] When the first execution path is to execute marking adjustment, the adaptive adjustment of the laser marking system is performed in combination with the laser marking energy fluctuation characteristic parameter, and the second execution path of the laser marking system is analyzed after the adjustment is completed.

[0013] A laser marking image set is collected, a laser marking carbonization characteristic parameter is analyzed, and a third execution path of the laser marking system is determined.

[0014] On the other hand, a high-precision semiconductor test sorting equipment is provided, which comprises a rack, a feeding mechanism, a test mechanism, a sorting assembly and a plastic sealing assembly.

[0015] A rack is used for supporting an upper loading mechanism, a testing mechanism, a sorting assembly and a plastic packaging assembly.

[0016] The loading mechanism is used for transmitting target chips to the testing mechanism.

[0017] The testing mechanism is used for executing a testing process of the target chips.

[0018] The sorting assembly is used for executing sorting, laser marking and laser marking testing processes of the target marking chips.

[0019] The plastic packaging assembly is used for executing a plastic packaging process of qualified chips.

[0020] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects:

[0021] 1. The high-precision semiconductor testing and sorting method provided by the present application can realize real-time monitoring and dynamic adjustment of energy fluctuation and carbonization quality in the laser marking process, not only improves the response speed and control precision of marking abnormalities, but also effectively avoids the misjudgment or good product NG problem caused by laser parameter drift, thereby significantly enhancing the stability, yield and automation level of the whole machine testing and sorting process.

[0022] 2. The adaptive adjustment of the laser marking system is performed by combining the laser marking energy fluctuation characteristic parameter, which can realize accurate identification and dynamic response of the laser power output state, real-time optimization and control of the marking deviation problem caused by unstable energy, effectively improve the consistency and reliability of the marking process, reduce the probability of marking failure caused by energy abnormalities, and improve the overall product yield.

[0023] 3. The third execution path of the laser marking system is determined by analyzing the laser marking carbonization characteristic parameter, which can quantitatively evaluate the image quality after laser marking, identify the carbonization abnormal state in time, and dynamically decide whether to perform re-adjustment operation, thereby realizing closed-loop control of the marking effect, effectively improving the clarity and consistency of the marking pattern, and preventing identification failure caused by too deep or too shallow carbonization. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical scheme in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0025] Figure 1 is a high-precision semiconductor testing and sorting method process schematic diagram provided by the embodiment of the present application.

[0026] Figure 2 is a high-precision semiconductor test and sorting equipment overall structure diagram provided by the embodiment of the application.

[0027] Figure 3 is a high-precision semiconductor test and sorting equipment feeding mechanism structure diagram provided by the embodiment of the application.

[0028] Figure 4 is a high-precision semiconductor test and sorting equipment test mechanism structure diagram provided by the embodiment of the application.

[0029] Figure 5 is a high-precision semiconductor test and sorting equipment lifting power mechanism structure diagram provided by the embodiment of the application.

[0030] Figure 6 is a high-precision semiconductor test and sorting equipment sorting assembly structure diagram provided by the embodiment of the application.

[0031] Figure 7 is a flow chart of the execution of the marking supplement adjustment and early warning trigger judgment provided by the embodiment of the application.

[0032] Figure 8 is a third execution path judgment flow chart provided by the embodiment of the application.

[0033] Figure 9 is a high-precision semiconductor test and sorting equipment CCD test result display provided by the embodiment of the application. Figure 1 .

[0034] Figure 10 is a high-precision semiconductor test and sorting equipment CCD test result display provided by the embodiment of the application. Figure 2 .

[0035] Reference signs: 1, rack; 2, feeding mechanism; 3, test mechanism; 4, sorting assembly; 5, plastic packaging assembly; 1-01, bottom plate; 2-01, circular vibrating disc; 2-02, linear vibrating disc; 2-03, material distribution mechanism; 3-01, lifting power mechanism; 3-01-01, stand column; 3-01-02, servo motor; 3-01-03, screw guide rail assembly; 3-01-04, compression cylinder; 3-01-05, support; 3-02, suction head rotating mechanism; 3-02-01, precision cam divider; 3-02-02, support piece; 3-02-03, suction head assembly; 3-02-04, synchronous belt; 3-02-05, test jig; 4-01, precision cam divider; 4-02, support piece; 4-03, material carrier disc; 4-04, material box; 4-05, laser marking system; 4-06, CCD test system. DETAILED DESCRIPTION

[0036] The technical solutions in the present application will be described below with reference to the drawings.

[0037] In the embodiments of the present application, the words such as "example", "for example" are used to represent an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific manner. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or one of the two.

[0038] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that the meanings expressed are consistent when the distinction is not emphasized. "Of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that the meanings expressed are consistent when the distinction is not emphasized.

[0039] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1. The meanings expressed are consistent when the distinction is not emphasized.

[0040] In order to make the technical problems, technical solutions and advantages of the present application clearer, specific embodiments will be described in detail below with reference to the drawings.

[0041] The embodiments of the present application provide a high-precision semiconductor test sorting method and equipment.

[0042] As shown in a high-precision semiconductor test sorting method flowchart, comprising the following steps: Figure 1

[0043] The semiconductor chip to be tested is recorded as a target chip, the test process of the target chip is executed, and the target chip that passes the test is recorded as a target marked chip.

[0044] The laser marking process of the target marked chip is executed and monitored in real time, the laser marking energy fluctuation characteristic parameter is analyzed, and the first execution path of the laser marking system is determined.

[0045] Further, the laser marking energy fluctuation characteristic parameter is analyzed, and the analysis process is as follows:

[0046] The laser marking process of the target marked chip is executed and monitored in real time, and within a preset time window, the marking energy fluctuation characteristic parameter of the laser marking system is collected, including the average output power, power fluctuation standard deviation, instantaneous power maximum deviation value and power slope change of the laser marking system. ​

[0047] In this embodiment, the average output power refers to the average value of the energy power outputted by the laser to the outside during the process of maintaining the marking work of the laser marking system in the preset time window, reflecting the overall stability level of the marking energy in this stage. Through the energy monitoring module built in the laser controller, the energy output value of the laser in the marking process is read in real time, the energy output value is continuously sampled in the preset time window, and the average value calculated from the energy output value sequence in this time period is taken as the average output power of the current period.

[0048] The power fluctuation standard deviation is used to measure the dispersion degree of the energy output in the laser marking process, which is a statistical measure of the fluctuation range of the output power. The larger the power fluctuation standard deviation, the worse the running stability of the laser system.

[0049] The maximum instantaneous power deviation value refers to the maximum difference between the energy output value at a certain time in the laser marking process and the average power in this time period, representing the most severe energy fluctuation point in the marking process.

[0050] The power slope change amount refers to the fluctuation amplitude of the slope of the laser output power with respect to time in the laser marking process, which is used to measure the change degree of the power change rate. In the preset time window in the laser marking process, the output power of the laser is continuously monitored. At each sampling time point, the change rate between the power and the previous sampling point is calculated, and the fluctuation of these change rates is recorded, and the maximum value is taken as the power slope change amount.

[0051] It should be noted that in specific embodiments, to avoid errors caused by single-point anomalies or isolated fluctuations in the overall judgment, the system has excluded obvious interference noise signals through continuous sampling and pre-filtering mechanism before calculating the maximum change rate, thereby ensuring that the maximum value is representative.

[0052] According to the marking energy fluctuation characteristic parameters, the laser marking energy fluctuation characteristic parameters are analyzed.

[0053] The defined average output power, the defined power fluctuation standard deviation, the defined maximum instantaneous power deviation value and the defined power slope change amount stored in the database are extracted.

[0054] The preset average output power fluctuation distribution factor, the power fluctuation standard deviation fluctuation distribution factor, the maximum instantaneous power deviation value fluctuation distribution factor and the power slope change amount fluctuation distribution factor in the database are extracted.

[0055] In the embodiment, a plurality of sets of fluctuation distribution factor mapping relationships of energy fluctuation characteristics have been pre-configured in the system database, which are used to associate different energy fluctuation characteristics with fluctuation evaluation weights. The mapping relationships are managed in the form of a structured data table (for example, a fluctuation distribution table), and combined with monitoring data and historical experimental results to form a complete power fluctuation analysis system. Based on the system, the fluctuation distribution factor corresponding to each energy fluctuation characteristic (such as average output power, power fluctuation standard deviation, maximum instantaneous power deviation value, and power slope change amount) can be directly obtained from the database. The numerical range of these fluctuation distribution factors is limited to 0 to 1, and the sum of all factors is 1, thereby ensuring reasonable weight distribution of each fluctuation characteristic in the total fluctuation evaluation.

[0056] The laser marking energy fluctuation characteristic parameter is a quantitative representation of the influence degree of the average output power, the power fluctuation standard deviation, the maximum instantaneous power deviation value, and the power slope change amount on the energy fluctuation state of the laser marking system. Specifically, the average output power is deviated from the defined average output power, the deviation result is compared with the defined average output power, the power fluctuation standard deviation, the maximum instantaneous power deviation value, and the power slope change amount are compared with the corresponding defined values respectively, and each comparison result is coupled and weighted with the corresponding fluctuation distribution factor, thereby obtaining the laser marking energy fluctuation characteristic parameter.

[0057] In specific embodiments, the laser marking energy fluctuation characteristic parameter is specifically represented as follows:

[0058] ,

[0059] Wherein, A is the laser marking energy fluctuation characteristic parameter, a is the average output power, b is the power fluctuation standard deviation, c is the maximum instantaneous power deviation value, d is the power slope change amount, a0 is the defined average output power, b0 is the defined power fluctuation standard deviation, c0 is the defined maximum instantaneous power deviation value, d0 is the defined power slope change amount, x1 is the average output power fluctuation distribution factor, x2 is the power fluctuation standard deviation fluctuation distribution factor, x3 is the maximum instantaneous power deviation value fluctuation distribution factor, and x4 is the power slope change amount fluctuation distribution factor.

[0060] Further, a first execution path of the laser marking system is determined, and the analysis process is as follows:

[0061] The pre-set laser marking energy fluctuation threshold in the database is extracted.

[0062] If the laser marking energy fluctuation characteristic parameter is less than or equal to the laser marking energy fluctuation threshold, the first execution path of the laser marking system is recorded as continuing laser marking.

[0063] In this embodiment, if the laser marking energy fluctuation characterization parameter is less than or equal to the laser marking energy fluctuation threshold, it indicates that the power output of the laser marking system is stable, and the energy fluctuation is within an acceptable range. This means that the current laser marking state meets expectations, with no excessive energy fluctuation or power instability, so the regular marking process can continue without additional adjustment measures. Therefore, the first execution path of the laser marking system is set to continue laser marking.

[0064] If the laser marking energy fluctuation characterization parameter is greater than the laser marking energy fluctuation threshold, the first execution path of the laser marking system is recorded as executing marking adjustment.

[0065] In this embodiment, if the laser marking energy fluctuation characterization parameter is greater than the laser marking energy fluctuation threshold, it indicates that there is a large fluctuation in the power output of the laser marking system, which exceeds the preset stable range. Such energy fluctuation may affect the marking quality, leading to inconsistent character depth or surface defects, etc. Therefore, in order to ensure the marking quality and stability, appropriate adjustment measures need to be taken, at this time, the first execution path of the laser marking system is set to execute marking adjustment to correct the current unstable condition.

[0066] When the first execution path is to execute marking adjustment, the adaptive adjustment of the laser marking system is executed in combination with the laser marking energy fluctuation characterization parameter, and the second execution path of the laser marking system is analyzed after the adjustment is completed.

[0067] Further, the adaptive adjustment of the laser marking system is executed, and the specific analysis process is as follows:

[0068] According to the laser marking energy fluctuation characterization parameter and the laser marking energy fluctuation threshold, a laser marking energy fluctuation deviation factor is obtained.

[0069] Based on the laser marking energy fluctuation deviation factor, the working parameter adjustment set of the laser marking system is matched.

[0070] In this embodiment, the mapping relationship between the laser marking energy fluctuation deviation factor and the corresponding laser marking system working parameter adjustment set is pre-stored in the database, and the mapping relationship is managed through a structured mapping table (such as a working parameter configuration table). In the matching process, the system first extracts the energy fluctuation deviation factor of the laser marking system, and uses the deviation factor as a key value to search for the corresponding working parameter adjustment set in the mapping table. The adjustment set includes a laser cooling waiting time supplement value, a power setting reduction value, and a pulse energy reduction value. These parameters are adjusted according to the different amplitudes of energy fluctuation to ensure the stability of power output during laser marking, and to avoid excessive fluctuation or power mutation. The mapping table is constructed based on long-term monitoring data and historical experimental analysis results of the laser marking system, and the values can be dynamically adjusted according to the real-time laser marking energy fluctuation, thereby effectively improving the adaptability and accuracy of the system under different working conditions. Through this process, the system can quickly determine and apply the corresponding working parameter adjustment set according to the specific situation of each laser energy fluctuation, thereby ensuring the quality and stability of the marking process.

[0071] The laser marking system working parameter adjustment set includes a laser cooling waiting time supplement value, a power setting reduction value, and a pulse energy reduction value.

[0072] It should be explained that the laser cooling waiting time supplement value refers to the additional delay amount added to the original cooling waiting time during the adaptive adjustment process of the laser marking system. This supplement value allows the laser to continue working in a more stable thermal state, which helps to improve the marking consistency.

[0073] The power setting reduction value refers to the numerical range of the required reduction in the originally set laser marking power based on the currently detected laser energy fluctuation. By reducing the laser power output, the risk of energy instability and abnormal marking quality caused by excessive power can be effectively reduced.

[0074] The pulse energy reduction value refers to the energy value reduced from the originally set pulse energy. This reduction helps to suppress the energy mutation effect caused by excessive laser pulse.

[0075] It should be added that the larger the laser marking energy fluctuation deviation factor, the greater the amplitude of energy fluctuation during laser marking, which may cause instability of the laser output power and affect the marking quality and accuracy. Therefore, in order to effectively compensate for such fluctuations, the system needs to appropriately adjust the marking process to reduce the negative impact of power fluctuation on the final marking effect. The larger the laser cooling waiting time supplement value, the greater the absolute value of the power setting reduction value, and the greater the absolute value of the pulse energy reduction value, thereby prolonging the laser cooling time, reducing the power output, and reducing the pulse energy, to ensure the stability of the power output during marking.

[0076] If the average output power exceeds the preset average output power threshold in the database, no pulse energy reduction is performed.

[0077] In this embodiment, if the average output power of the laser marking system exceeds the preset average output power threshold in the database, it indicates that the power output during the laser marking process is too high, which may cause overheating, carbonization or other quality problems. Therefore, in order to ensure the marking quality and protect the stable operation of the system, it is necessary to reduce the power setting value, reduce the output power of the laser, avoid the influence on the marking effect due to excessive energy, and improve the safety and adaptability of the system.

[0078] If the average output power does not exceed the preset average output power threshold in the database, no power setting reduction is performed.

[0079] In this embodiment, if the average output power does not exceed the preset average output power threshold in the database, it indicates that the power output during the laser marking process is within a normal range, but in order to further ensure the stability of the power and improve the precision of the marking, certain adjustment is still needed. Therefore, the system will reduce the pulse energy, further avoid the occurrence of local excessive heat input or power fluctuation, ensure that the marking process is more fine and stable, and reduce unnecessary fluctuation influence.

[0080] The adaptive adjustment of the laser marking system is performed according to the working parameter adjustment set of the laser marking system.

[0081] In one specific embodiment, if the average output power exceeds the preset average output power threshold in the database, the current laser cooling waiting time and the current power setting value recorded in the system program log are obtained, the sum of the current laser cooling waiting time and the laser cooling waiting time supplement value is taken as the execution laser cooling waiting time, and the sum of the current power setting value and the power setting reduction value is taken as the execution power setting value.

[0082] In another specific embodiment, if the average output power does not exceed the preset average output power threshold in the database, the current laser cooling waiting time and the current pulse energy recorded in the system program log are obtained, the sum of the current laser cooling waiting time and the laser cooling waiting time supplement value is taken as the execution laser cooling waiting time, and the sum of the current pulse energy and the pulse energy reduction value is taken as the execution pulse energy.

[0083] Further, the second execution path of the laser marking system is analyzed, and the specific analysis process is as follows:

[0084] The first waiting time is matched with the laser marking energy fluctuation allowable deviation interval based on the laser marking energy fluctuation deviation factor.

[0085] In this embodiment, the system database pre-stores a mapping rule between the laser marking energy fluctuation deviation factor and its corresponding first waiting time and allowed deviation interval, which is managed in the form of a structured configuration table. In actual operation, the system will extract the energy fluctuation deviation factor in the current laser marking process in real time, and use this factor as an index item to retrieve the matching first waiting time and corresponding energy fluctuation tolerance interval in the mapping table.

[0086] It should be noted that the larger the laser marking energy fluctuation deviation factor, the stronger the instability of the current power output of the laser system, the more intense the previously performed adaptive adjustment intervention, and in order to ensure that the system has sufficient stable recovery time after high-amplitude adjustment, the system will extract a longer first waiting time to extend the cooling period or transition period after adjustment is completed; at the same time, in order to avoid misjudgment during adjustment, the system will extract a wider laser marking energy fluctuation allowed deviation interval to accommodate possible slight fluctuations in the early stage of adjustment. The setting of this allowed deviation interval is mainly used to determine whether the current laser system state is within a reasonable energy fluctuation range after adjustment is completed. When the energy fluctuation deviation factor obtained by subsequent monitoring falls within this interval, it can be determined that the system has recovered to a stable operating state; otherwise, subsequent supplemental adjustment strategies need to be continued to prevent laser marking quality abnormalities.

[0087] After the first waiting time ends, the laser marking energy fluctuation deviation factor is reanalyzed, denoted as the first energy fluctuation deviation factor.

[0088] If the first energy fluctuation deviation factor is within the laser marking energy fluctuation allowed deviation interval, the second execution path of the laser marking system is recorded as continuous monitoring, and a stable signal is generated.

[0089] If the first energy fluctuation deviation factor is within the laser marking energy fluctuation allowed deviation interval, it means that after the adaptive adjustment of the previous stage, the power output of the laser system has recovered to the expected stable state, and the energy fluctuation is within the system tolerance range. Therefore, the system does not need to continue additional adjustment operations, and the second execution path can be set to continuous monitoring state, and a stable signal is generated as the basis for subsequent image acquisition and carbonization analysis.

[0090] If the first energy fluctuation deviation factor is not within the laser marking energy fluctuation allowed deviation interval, the second execution path of the laser marking system is recorded as performing marking supplemental adjustment.

[0091] If the first energy fluctuation deviation factor is not within the laser marking energy fluctuation allowable deviation interval, it indicates that the previous adjustment has not effectively suppressed the energy fluctuation of the laser system, and the system is still in an abnormal fluctuation state, and continuing to mark will have a higher quality risk. Therefore, the system sets the second execution path to perform a marking supplementary adjustment to gradually return the system to a stable working interval, ensuring the accuracy and reliability of the subsequent marking process.

[0092] As shown in Figure 7 is the execution marking supplementary adjustment and early warning trigger judgment flowchart involved in the embodiment of the application, the system analyzes the first energy fluctuation deviation factor, extracts the laser marking frequency reduction value and the second waiting time length according to the deviation factor, and performs a marking supplementary adjustment. After the second waiting time length ends, the system analyzes the second energy fluctuation deviation factor again. Then, it is judged whether the second energy fluctuation deviation factor is within the laser marking energy fluctuation allowable deviation interval: if it is within the interval, it indicates that the adjustment effect is good, and the system will generate a stable signal; if it is not within the interval, it indicates that the adjustment effect has not yet reached the expected stable state, and the system will generate a warning signal to trigger subsequent fault response or safety protection processes.

[0093] Further, the marking supplementary adjustment is performed, and the specific execution process is as follows:

[0094] Based on the first energy fluctuation deviation factor, the laser marking frequency reduction value and the second waiting time length are extracted.

[0095] In this embodiment, the mapping rule between the first energy fluctuation deviation factor and the laser marking frequency reduction value and the second waiting time length is pre-set in the system database, and the mapping rule is maintained in the form of a parameter control reference table. In the actual adjustment process, the system will first extract the first energy fluctuation deviation factor of the current stage, and use it as an index variable to find the frequency reduction reference value and the recommended waiting time setting value that match it in the mapping table.

[0096] It should be noted that the larger the first energy fluctuation deviation factor is, the more significant the energy output instability of the laser marking system after completing the previous stage of adjustment, the higher the current energy disturbance intensity of the system, and the less obvious the recovery trend, indicating that the existing control means is not enough to suppress energy fluctuation. Therefore, the larger the extracted laser marking frequency reduction value is, the lower the pulse frequency of the laser is, and the fewer the energy excitation times per unit time, thereby alleviating the risk of energy accumulation or divergence caused by excessive excitation frequency. At the same time, the second waiting time length will also be extracted as a larger value to provide a longer system buffer and cooling period, so that the laser has sufficient recovery time.

[0097] The marking supplement adjustment is performed based on the laser marking frequency reduction value and the current laser marking frequency, i.e., the sum of the current laser marking frequency extracted from the system program log and the laser marking frequency reduction value is taken as the executed laser marking frequency.

[0098] After the adjustment is completed, after the second waiting duration ends, the laser marking energy fluctuation deviation factor is analyzed again, denoted as a second energy fluctuation deviation factor.

[0099] If the second energy fluctuation deviation factor is within the laser marking energy fluctuation allowable deviation interval, a stable signal is generated.

[0100] In this embodiment, if the second energy fluctuation deviation factor is within the laser marking energy fluctuation allowable deviation interval, it indicates that after the two rounds of adjustment and recovery periods, the energy output of the laser marking system has recovered to the system acceptable fluctuation range, and the current power stability meets the requirements of performing normal marking procedures. At this time, the system considers that the current running state has reached stability and can support the normal progress of subsequent marking image acquisition, carbonization quality judgment and other links, and therefore generates a stable signal as a trigger condition for executing the next process.

[0101] If the second energy fluctuation deviation factor is not within the laser marking energy fluctuation allowable deviation interval, a warning signal is generated.

[0102] In this embodiment, if the second energy fluctuation deviation factor is not within the laser marking energy fluctuation allowable deviation interval, it indicates that even after performing the marking supplement adjustment and the second waiting period, the energy fluctuation of the laser marking system still significantly deviates and cannot meet the marking stability requirements, and there is a high quality risk. At this time, the system judges that the current state has exceeded the self-adjustment control range and needs to attract the attention of the system, and therefore generates a warning signal to prompt the operator to intervene or trigger the backup marking strategy to avoid continuing work in an unstable state.

[0103] The laser marking image set is collected, the laser marking carbonization characteristic parameters are analyzed, and the third execution path of the laser marking system is determined.

[0104] Further, the laser marking carbonization characteristic parameters are analyzed, and the specific analysis process is as follows:

[0105] After receiving the stable signal, the laser marking image set is collected, and the image carbonization parameters are obtained.

[0106] The image carbonization parameters include the average median gray value, the maximum edge-center gray difference value, the average edge gradient intensity, and the Laplacian variance minimum value.

[0107] The average median gray value is a statistical quantity obtained by calculating the median gray value of each image in the laser marking image set collected from the laser marking process, and then performing average operation on the median gray values, and is used to reflect the median trend of the gray distribution of the marking area as a whole.

[0108] The maximum edge-center gray difference value is a parameter obtained by calculating the difference between the edge region and the center region gray value of each image in the laser marking image set, taking the maximum value of the difference values as the parameter, and reflecting the contrast intensity between the carbonized edge and the center produced by the laser marking.

[0109] It should be noted that in specific embodiments, noise removal will be performed before extracting the maximum value to remove incidental factors and make the extracted maximum value have true representativeness.

[0110] The average edge gradient intensity is a parameter obtained by calculating the gradient intensity of the edge region of the collected laser marking image set and taking the average value, and is used to measure the sharpness and change intensity of the carbonized region edge.

[0111] The Laplacian variance minimum value is a parameter obtained by calculating the variance value of the Laplacian operator response of each image in the laser marking image set and selecting the smallest variance value, and is used to reflect the smoothness of the texture details and the uniformity of carbonization in the marking area.

[0112] It should be noted that the image carbonization parameter can be obtained by image analysis algorithm, and the openCV is used in the embodiment.

[0113] The laser marking carbonization characterization parameter is analyzed based on the image carbonization parameter.

[0114] The reference average median gray value, the reference maximum edge-center gray difference value, the reference average edge gradient intensity and the reference Laplacian variance minimum value stored in the database are extracted.

[0115] The average median gray value carbonization measurement coefficient, the maximum edge-center gray difference value carbonization measurement coefficient, the average edge gradient intensity carbonization measurement coefficient and the Laplacian variance minimum value carbonization measurement coefficient preset in the database are extracted.

[0116] In this embodiment, the system database has pre-stored mapping relationships of multiple sets of carbonization characteristic metric coefficients, which are used to associate different image carbonization parameters with their weight contributions in overall carbonization quality evaluation. The mapping relationships are managed through a structured data table (e.g., a carbonization metric coefficient table), and combined with collected image data and historical carbonization quality feedback, a complete image carbonization characterization system is constructed. Based on this system, the system can directly obtain the carbonization metric coefficients corresponding to each carbonization characteristic parameter (including the average median gray value, the maximum edge-center gray difference value, the average edge gradient intensity, and the Laplacian variance minimum value) from the database. The numerical range of these metric coefficients is limited to 0 to 1, and the sum of each metric coefficient is 1, to ensure reasonable weight distribution of different carbonization characteristics in comprehensive carbonization state evaluation, which helps to achieve comprehensive and accurate marking carbonization quality judgment.

[0117] The laser marking carbonization characterization parameter is a quantitative representation of the degree of influence of the average median gray value, the maximum edge-center gray difference value, the average edge gradient intensity, and the Laplacian variance minimum value on the laser marking carbonization state, which is specifically represented as: comparing the average median gray value and the maximum edge-center gray difference value with the corresponding reference values, respectively, comparing the reference values of the average edge gradient intensity and the Laplacian variance minimum value with the average edge gradient intensity and the Laplacian variance minimum value, and coupling the results of each comparison with the corresponding carbonization metric coefficient for weighted processing, thereby obtaining the laser marking carbonization characterization parameter.

[0118] In specific embodiments, the laser marking carbonization characterization parameter is specifically represented as:

[0119] ,

[0120] wherein B is the laser marking carbonization characterization parameter, f is the average median gray value, g is the maximum edge-center gray difference value, h is the average edge gradient intensity, k is the Laplacian variance minimum value, f0 is the reference average median gray value, g0 is the reference maximum edge-center gray difference value, h0 is the reference average edge gradient intensity, k0 is the reference Laplacian variance minimum value, y1 is the average median gray value carbonization metric coefficient, y2 is the maximum edge-center gray difference value carbonization metric coefficient, y3 is the average edge gradient intensity carbonization metric coefficient, and y4 is the Laplacian variance minimum value carbonization metric coefficient.

[0121] As Figure 8As shown, the third execution path judgment flowchart related to the embodiment of the present application is shown. First, a laser marking image set is collected, and an image carbonization parameter is extracted. A laser marking carbonization characterization parameter is calculated therefrom, and the characterization parameter is compared with a preset laser marking carbonization characterization threshold in a database. If the laser marking carbonization characterization parameter is greater than or equal to the carbonization characterization threshold, it is indicated that the marking effect deviates from the normal carbonization state, and the third execution path is determined as executing the readjustment of the laser marking system. Otherwise, it is considered that the marking effect meets the carbonization expectation, and the third execution path is determined as executing the normal laser marking process.

[0122] Further, the third execution path of the laser marking system is determined, and the specific analysis process is as follows:

[0123] The preset laser marking carbonization characterization threshold in the database is extracted.

[0124] If the laser marking carbonization characterization parameter is greater than or equal to the laser marking carbonization characterization threshold, the third execution path of the laser marking system is recorded as executing the readjustment of the laser marking system.

[0125] In the embodiment, if the laser marking carbonization characterization parameter is greater than or equal to the laser marking carbonization characterization threshold, it is indicated that the carbonization degree generated in the laser marking process has exceeded the preset quality control range, there is a risk of excessive carbonization or uneven carbonization, and the clarity and reliability of the chip identification may be affected. Therefore, the system triggers the readjustment program of the laser marking, optimizes the marking process by adjusting the related parameters, reduces the carbonization abnormality, and ensures that the subsequent marking effect meets the quality requirements.

[0126] It needs to be understood that although the early power fluctuation has been controlled, the single-point heat accumulation may be caused due to the reduction of the frequency, the extension of the marking dwell time, or the overall high operation of the laser power, which may cause the material surface to be ablated excessively, the character edge to be carbonized, or the character edge to be blackened.

[0127] If the laser marking carbonization characterization parameter is less than the laser marking carbonization characterization threshold, the third execution path of the laser marking system is recorded as executing the normal laser marking process.

[0128] In the embodiment, if the laser marking carbonization characterization parameter is less than the laser marking carbonization characterization threshold, it is indicated that the carbonization state in the current laser marking process is within the reasonable control range, the carbonization degree is moderate and uniform, and the predetermined quality standard is met. The system therefore confirms that the laser marking process is in a stable state, continues to execute the normal laser marking operation, does not need additional adjustment, and ensures efficient and consistent marking quality.

[0129] Further, the readjustment of the laser marking system is executed, and the specific execution process is as follows:

[0130] Subtracting the laser marking carbonization characterization threshold from the laser marking carbonization characterization parameter to obtain a laser marking carbonization characterization deviation factor.

[0131] If there is a laser marking frequency reduction value in the laser marking system in the current monitoring period, a laser duty cycle reduction value is extracted based on the laser marking carbonization characterization deviation factor, and based on this, readjustment of the laser marking system is performed, that is, the current laser duty cycle in the system program log is added to the laser duty cycle reduction value to obtain the executed laser duty cycle.

[0132] It needs to be explained that frequency reduction means to reduce the number of markings per unit time to avoid heat accumulation. If carbonization still occurs, it means that the single laser action time / energy is still too high. At this time, the duty cycle is reduced to reduce the effective energy release of each pulse.

[0133] In this embodiment, the correspondence between the laser marking carbonization characterization deviation factor and the laser duty cycle reduction value is established in the system database in advance, and the relationship is maintained through a structured parameter mapping table (such as a duty cycle adjustment configuration table). During extraction, the system first obtains the current laser marking carbonization characterization deviation factor and uses it as a search key to find a matching laser duty cycle reduction value in the mapping table. The mapping table is dynamically optimized based on the carbonization deviation data and adjustment effect feedback in the previous marking process to ensure that the extracted duty cycle reduction value can accurately reflect the current deviation degree, thereby effectively guiding the parameter adjustment of the laser marking device and realizing targeted correction of carbonization abnormalities.

[0134] It needs to be supplemented that the larger the laser marking carbonization characterization deviation factor, the higher the degree of carbonization abnormality in the current laser marking process. In order to reduce the quality risk caused by excessive carbonization, it is necessary to adjust the laser parameters more significantly. Therefore, the absolute value of the corresponding laser duty cycle reduction value extracted by the system is also larger to strengthen the control of laser energy output, thereby effectively inhibiting excessive carbonization and improving the stability of marking quality.

[0135] If there is no laser marking frequency reduction value in the laser marking system in the current monitoring period, and there is a pulse energy reduction value, a Z-axis defocusing value is extracted based on the laser marking carbonization characterization deviation factor, and based on this, readjustment of the laser marking system is performed, that is, the current Z-axis focal length value in the system program log is added to the Z-axis defocusing value to obtain the executed focal length.

[0136] It needs to be explained that the pulse energy has been reduced, which means that the energy overload has been preliminarily solved. If carbonization still occurs, the Z-axis defocusing is adjusted to smooth the spot and soften the marking edge.

[0137] In this embodiment, the mapping relationship between the laser marking carbonization characterization deviation factor and the corresponding Z-axis defocusing value is pre-configured in the system database, which is managed through a structured adjustment table (such as a focal length adjustment configuration table). During extraction, the system first obtains the current laser marking carbonization characterization deviation factor and uses it as a key parameter to match the corresponding Z-axis defocusing value in the mapping table. This mapping is constructed based on historical marking quality data and focal length adjustment effects, and can accurately reflect the demand for focal length adjustment due to carbonization deviation, thereby achieving targeted focal point optimization and improving the overall quality and consistency of laser marking.

[0138] It should be noted that the larger the laser marking carbonization characterization deviation factor, the more significant the carbonization deviation during laser marking. In order to improve the problem of uneven carbonization caused by focal point position, it is necessary to adjust the laser focal length more significantly. Therefore, the Z-axis defocusing value extracted by the system also increases to achieve fine adjustment of the laser focal point, promoting the improvement of carbonization uniformity and marking quality.

[0139] If there is no laser marking frequency reduction value for the laser marking system in the current monitoring period, and there is a power setting reduction value, then based on the laser marking carbonization characterization deviation factor, the pulse duration reduction value is extracted, and based on this, the re-adjustment of the laser marking system is performed, that is, the current pulse duration in the system program log is added to the pulse duration reduction value to obtain the execution pulse duration.

[0140] It should be noted that if carbonization still occurs after power adjustment, it means that heat is excessively concentrated in a certain point, causing scorching and carbonization, so the pulse duration needs to be shortened.

[0141] In this embodiment, the system database pre-saves the corresponding relationship between the laser marking carbonization characterization deviation factor and the pulse duration reduction value, which is maintained in the form of a structured adjustment mapping table (such as a pulse duration adjustment table). During extraction, the system uses the current laser marking carbonization characterization deviation factor as the basis for querying to locate and match the pulse duration reduction value in the mapping table. This mapping relationship is continuously optimized based on historical marking data and adjustment effect experience, ensuring that the obtained reduction value can accurately regulate the duration of the laser pulse, thereby effectively reducing carbonization abnormalities and improving the uniformity and stability of marking quality.

[0142] It should be noted that the larger the laser marking carbonization characterization deviation factor, the higher the degree of carbonization abnormality in the current laser marking process. In order to effectively reduce the phenomenon of excessive carbonization and optimize marking quality, the absolute value of the pulse duration reduction value corresponding to the system extraction also increases to reduce the duration of the laser pulse by reducing the input of laser energy, thereby improving carbonization uniformity and enhancing overall marking effect.

[0143] As Figure 2The diagram shows a high-precision semiconductor testing and sorting device. The device includes: a frame 1, a feeding mechanism 2, a testing mechanism 3, a sorting component 4, and a molding assembly 5.

[0144] The frame 1 serves as a support, and the feeding mechanism 2, testing mechanism 3, sorting component 4, and sealing component 5 are placed on top of it.

[0145] The feeding mechanism 2 is used to transfer the target chip to the testing mechanism.

[0146] like Figure 3 The diagram shown is a structural diagram of a feeding mechanism for a high-precision semiconductor testing and sorting equipment provided in an embodiment of the present invention. The feeding mechanism 2 includes a circular vibratory plate 2-01, a linear vibratory plate 2-02, and a sorting mechanism 2-03.

[0147] The feeding mechanism 2 adopts a combination of a circular vibratory feeder 2-01 and a linear vibratory feeder 2-02. The front end of the linear vibratory feeder 2-02 is equipped with a separating mechanism 2-03 to separate the product into individual products for easy gripping.

[0148] Test unit 3 is used to execute the test process for the target chip.

[0149] like Figure 4 The diagram shown is a structural diagram of a high-precision semiconductor testing and sorting equipment testing mechanism provided in an embodiment of the present invention. The testing mechanism 3 includes a lifting power mechanism 3-01 and a suction head rotation mechanism 3-02.

[0150] like Figure 5 The diagram shows the structure of a lifting power mechanism for a high-precision semiconductor testing and sorting equipment according to an embodiment of the present invention. The lifting power mechanism 3-01 is mounted on the base plate 1-01 via a column 3-01-01. The lifting power mechanism 3-01 is driven by a servo motor 3-01-02 and a lead screw guide rail assembly 3-01-03. The lead screw guide rail assembly 3-01-03 is connected to a bracket 3-01-05, and a clamping cylinder 3-01-04 is mounted on the bracket 3-01-05.

[0151] The suction head rotation mechanism 3-02 is mounted on the base plate 1-01 via a precision cam divider 3-02-01. A support component 3-02-02 is mounted on the precision cam divider 3-02-01. A suction head assembly 3-02-03 is mounted on the support component 3-02-02. A test fixture 3-02-05 is mounted below the suction head rotation mechanism 3-02, used to test different performance characteristics of the chips. Chips that pass the test are marked as target marking chips.

[0152] The aforementioned performance includes, but is not limited to, chip mark detection, shape detection, electrical performance detection, pin length detection, and pin pitch detection.

[0153] As Figure 6 shown, is a high-precision semiconductor test and sorting equipment sorting assembly structure provided by an embodiment of the application, the sorting assembly 4 is used for performing sorting, laser marking and laser marking test procedures of a target marking chip.

[0154] The sorting assembly 4 is installed on the bottom plate 1-01 through a precision cam divider 4-01. The precision cam divider 4-01 is connected with the precision cam divider 3-02-01 through a synchronous belt 3-02-04, so that the two can move synchronously. A support 4-02 is installed on the precision cam divider 4-01, and a plurality of material carriers 4-03 are installed on the support 4-02. A plurality of material boxes 4-04 are placed below the support 4-02, and are used for containing different NG materials. A laser marking system 4-05 is arranged beside the support 4-02. A CCD test system 4-06 is arranged beside the laser marking system 4-05, and is used for detecting the marking effect of the target marking chip. If the marking is NG, the chip is thrown into a waste box 4-07; if the marking is qualified, the corresponding target marking chip is marked as a qualified chip.

[0155] As Figure 9 shown, is a high-precision semiconductor test and sorting equipment CCD test result display Figure 1 provided by an embodiment of the application. The chip marking effect shown in the figure is qualified, and the chip can enter a plastic packaging assembly to perform a plastic packaging procedure.

[0156] As Figure 10 shown, is a high-precision semiconductor test and sorting equipment CCD test result display Figure 2 provided by an embodiment of the application. The chip marking effect shown in the figure is unqualified, and the CCD test system marks the chip as NG, and throws the chip into a waste box.

[0157] The above marking information includes but is not limited to a semiconductor chip model, an identifier and the like.

[0158] The plastic packaging assembly 5 is used for performing a plastic packaging procedure of a qualified chip.

[0159] In the embodiment, the high-precision semiconductor test and sorting equipment performs the following procedures: a person introduces materials to be tested and sorted into the circular vibrating disc 2-01, the materials are classified and sent into the linear vibrating disc 2-02, the materials are classified into single ones under the action of the material classification mechanism 2-03, the material suction head rotating mechanism 3-02 sucks the materials and rotates to place the materials, the materials enter a test position, after multiple tests, the materials are placed on the sorting assembly 4, are rotated in sequence, the NG products are sorted by air, are laser marked, and are subjected to CCD test after laser marking. The laser marking chip NG is thrown, the qualified products enter the plastic packaging and coding assembly to be coded and packaged, and the unqualified products are discarded at the next station.

[0160] The above-described embodiments can be implemented in part or in whole through software, hardware (e.g., circuitry), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When loaded and executed by a computer, the computer instructions or computer programs can generate the flow or function according to the embodiments of the present application in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, such as from one website site, computer, server, or data center to another website site, computer, server, or data center through a wired (e.g., infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing a set of one or more available media. The available media can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0161] It should be understood that the term "and / or" used herein is merely an association relationship between the associated objects, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents an "or" relationship between the associated objects before and after it, but it can also represent an "and / or" relationship, which can be understood in the context before and after it.

[0162] In the present application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can mean a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0163] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined by their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0164] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0165] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the devices, apparatuses and units described above can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0166] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0167] If the functions are realized in the form of software functional units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method of each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0168] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A high-precision semiconductor testing and sorting method, characterized in that, Includes the following steps: The semiconductor chip to be tested is designated as the target chip. The test process for the target chip is executed, and the target chip that passes the test is designated as the target marking chip. The laser marking process of the target marking chip is executed and monitored in real time. The laser marking energy fluctuation characterization parameters are analyzed to determine the first execution path of the laser marking system. When the first execution path is to perform marking adjustment, the laser marking system is adaptively adjusted by combining the laser marking energy fluctuation characterization parameter, and the second execution path of the laser marking system is analyzed after the adjustment is completed. Collect a set of laser marking images, analyze the laser marking carbonization characterization parameters, and determine the third execution path of the laser marking system; The analysis of the laser marking carbonization characterization parameters is as follows: After receiving a stable signal, a set of laser marking images is acquired to obtain the image carbonization parameters; The image carbonization parameters include the average median gray value, the maximum edge-to-center gray value difference, the average edge gradient intensity, and the minimum Laplacian variance. Analysis of laser marking carbonization characterization parameters based on image carbonization parameters; The laser marking carbonization characterization parameters are a quantitative representation of the influence of the average median gray value, the maximum edge-to-center gray value, the average edge gradient intensity, and the minimum Laplacian variance on the laser marking carbonization state. Specifically, the average median gray value and the maximum edge-to-center gray value are compared with their corresponding reference values. The reference values ​​of the average edge gradient intensity and the minimum Laplacian variance are compared with the minimum average edge gradient intensity and the minimum Laplacian variance. The results of each comparison are combined with the corresponding carbonization metric coefficients for weighted coupling to obtain the laser marking carbonization characterization parameters.

2. The high-precision semiconductor testing and sorting method according to claim 1, characterized in that, The analysis of the laser marking energy fluctuation characterization parameters is as follows: The laser marking process of the target marking chip is executed and monitored in real time. Within a preset time window, the marking energy fluctuation characteristics of the laser marking system are collected, including the average output power of the laser marking system, the standard deviation of power fluctuation, the maximum instantaneous power deviation, and the power slope change. Analyze the laser marking energy fluctuation characterization parameters based on the marking energy fluctuation characterization parameters; The laser marking energy fluctuation characterization parameters are a quantitative representation of the influence of average output power, power fluctuation standard deviation, instantaneous maximum power deviation, and power slope change on the energy fluctuation state of the laser marking system. Specifically, the average output power is deviated from the defined average output power, and the deviation processing result is compared with the defined average output power. The power fluctuation standard deviation, instantaneous maximum power deviation, and power slope change are compared with the corresponding defined values. The comparison results are then weighted and coupled with the corresponding fluctuation allocation factor to obtain the laser marking energy fluctuation characterization parameters.

3. The high-precision semiconductor testing and sorting method according to claim 2, characterized in that, The process for determining the first execution path of the laser marking system is as follows: Extract the preset laser marking energy fluctuation threshold from the database; If the laser marking energy fluctuation characterization parameter is less than or equal to the laser marking energy fluctuation threshold, then the first execution path of the laser marking system is recorded as continuing laser marking; If the parameter representing the energy fluctuation of laser marking is greater than the threshold for energy fluctuation of laser marking, then the first execution path of the laser marking system is recorded as the execution of marking adjustment.

4. The high-precision semiconductor testing and sorting method according to claim 1, characterized in that, The adaptive adjustment of the laser marking system is analyzed in the following process: The laser marking energy fluctuation deviation factor is obtained based on the laser marking energy fluctuation characterization parameter and the laser marking energy fluctuation threshold. A set of adjustment parameters for the working parameters of a laser marking system based on matching the laser marking energy fluctuation deviation factor; The set of adjustable working parameters for the laser marking system includes the laser cooling waiting time supplement value, the power setting reduction value, and the pulse energy reduction value. If the average output power exceeds the preset average output power threshold in the database, pulse energy reduction will not be performed. If the average output power does not exceed the preset average output power threshold in the database, no power reduction will be performed. The laser marking system is adaptively adjusted based on the set of operating parameters.

5. The high-precision semiconductor testing and sorting method according to claim 1, characterized in that, The analysis process for the second execution path of the laser marking system is as follows: Based on the laser marking energy fluctuation deviation factor matching first waiting time and laser marking energy fluctuation allowable deviation range; After the first waiting period ends, the laser marking energy fluctuation deviation factor is re-analyzed and denoted as the first energy fluctuation deviation factor. If the first energy fluctuation deviation factor is within the allowable deviation range of laser marking energy fluctuation, then the second execution path of the laser marking system is recorded as continuous monitoring and a stable signal is generated; If the first energy fluctuation deviation factor is not within the allowable deviation range of laser marking energy fluctuation, then the second execution path of the laser marking system is recorded as executing marking supplementary adjustment.

6. The high-precision semiconductor testing and sorting method according to claim 5, characterized in that, The specific execution process for the tagging and supplementary adjustment is as follows: The laser marking frequency reduction value and the second waiting time are extracted based on the first energy fluctuation deviation factor; Perform marking supplementary adjustments based on the laser marking frequency reduction value and the current laser marking frequency; After the adjustment is completed, after the second waiting period ends, the laser marking energy fluctuation deviation factor is analyzed again and recorded as the second energy fluctuation deviation factor. If the second energy fluctuation deviation factor is within the allowable deviation range of laser marking energy fluctuation, a stable signal is generated; If the second energy fluctuation deviation factor is not within the allowable deviation range of laser marking energy fluctuation, an early warning signal will be generated.

7. The high-precision semiconductor testing and sorting method according to claim 1, characterized in that, The specific analysis process for determining the third execution path of the laser marking system is as follows: Extract the preset laser marking carbonization characterization threshold from the database; If the laser marking carbonization characterization parameter is greater than or equal to the laser marking carbonization characterization threshold, then the third execution path of the laser marking system is recorded as the readjustment of the laser marking system. If the laser marking carbonization characterization parameter is less than the laser marking carbonization characterization threshold, then the third execution path of the laser marking system is recorded as executing the normal laser marking process.

8. The high-precision semiconductor testing and sorting method according to claim 7, characterized in that, The specific process for readjusting the laser marking system is as follows: The laser marking carbonization characterization deviation factor is obtained by subtracting the laser marking carbonization characterization threshold from the laser marking carbonization characterization parameter. If there is a laser marking frequency reduction value in the laser marking system during the current monitoring period, the laser duty cycle reduction value is extracted based on the laser marking carbonization characterization deviation factor, and the laser marking system is readjusted accordingly. If, during the current monitoring period, the laser marking system does not have a laser marking frequency reduction value but does have a pulse energy reduction value, then the Z-axis defocus value is extracted based on the laser marking carbonization characterization deviation factor, and the laser marking system is readjusted accordingly. If, during the current monitoring period, the laser marking system does not have a laser marking frequency reduction value but does have a power setting reduction value, then the pulse duration reduction value is extracted based on the laser marking carbonization characterization deviation factor, and the laser marking system is readjusted accordingly.

9. A high-precision semiconductor testing and sorting device, employing the high-precision semiconductor testing and sorting method as described in any one of claims 1-8, characterized in that, The equipment includes: a frame, a feeding mechanism, a testing mechanism, a sorting component, and a sealing component; The frame serves as a support, and the feeding mechanism, testing mechanism, sorting component, and sealing component are placed on top of it. The feeding mechanism is used to transfer the target chip to the testing mechanism; The testing mechanism is used to execute the testing process for the target chip; The sorting component is used to perform the sorting, laser marking, and laser marking testing processes of the target marking chips. The molding assembly is used to perform the molding process for qualified chips.

Images