Gas spray header detection device and detection method
By designing a gas spray head testing device and a standardized testing process, the problem of the inability to comprehensively test the performance of CVD spray heads in existing technologies has been solved. This enables accurate measurement of gas distribution in spray heads and meets high-precision process requirements, allowing the testing standards to evolve in response to process development.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XIAN AEROSPACE YUANZHENG FLUID CONTROL
- Filing Date
- 2026-01-22
- Publication Date
- 2026-05-12
AI Technical Summary
Existing technologies cannot comprehensively and accurately detect the overall performance of CVD spray heads, especially the surface quality of small holes, geometric tolerances, burrs, etc., which affect the gas flow distribution of the spray head under actual working conditions and cannot meet the stringent requirements of semiconductor CVD processes.
A gas spray head detection device is adopted, including a data acquisition component and a gas source delivery component. Through the design of pressure sensor, gas flow meter and sealing gasket, it realizes independent acquisition and accurate measurement of the gas flow rate in different local areas of the spray head. Combined with machine learning model to optimize the detection standard, it provides a standardized detection process.
It significantly improves the representativeness and accuracy of spray head testing, ensuring that the performance of spray heads meets high-precision process requirements, and provides cross-batch and cross-model data comparison and analysis functions to adapt to equipment aging and process changes.
Smart Images
Figure CN122016273A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor manufacturing technology. Background Technology
[0002] In the semiconductor manufacturing field, chemical vapor deposition (CVD) equipment is a crucial core component. The spray head, as a key part of the CVD equipment, directly affects the uniformity and quality of the deposited thin film, as well as the final performance of the semiconductor device. With the rapid development of the semiconductor industry, the performance requirements for CVD spray heads are increasingly stringent.
[0003] Currently, CVD spray heads are generally disc-shaped, divided into upper and lower layers, and have a porous structure. After the CVD spray head is machined, the inspection method is to detect the aperture through optical equipment. Since the quality of the small holes determines the quality of the gas at the spray head outlet, dimensional inspection cannot represent the overall performance of the spray head. It often lacks systematicness and specificity, and cannot comprehensively and accurately detect and evaluate key performance indicators such as gas flow distribution of the spray head under actual working conditions, such as the surface quality, form and position tolerances, and burrs of the small holes inside the spray disc. It cannot guarantee that the machined spray head can meet the stringent requirements of semiconductor CVD processes, which may affect the quality of the entire CVD equipment and the yield of semiconductor manufacturing. Summary of the Invention
[0004] To address the aforementioned technical problems, this invention provides a gas spray head detection device and detection method.
[0005] To achieve the above objectives, the present invention specifically adopts the following technical solution: A gas spray head detection device includes a data acquisition component, a gas source delivery component, and a detection fixture; The gas supply delivery assembly includes a pump connected to a gas source, and the pump output end is sequentially connected to a hand valve, a pressure reducing valve, and a needle valve. The needle valve output end is connected to the detection fixture. The data acquisition component includes a data acquisition system, a pressure sensor, an inlet gas flow meter, and an outlet gas flow meter. There are two pressure sensors, one of which is located between the hand valve and the pressure reducing valve, and the other is located between the pressure reducing valve and the needle valve. The inlet gas flow meter is located between the output end of the needle valve and the detection fixture, and the outlet gas flow meter is located at the output end of the detection fixture. The detection fixture includes an air inlet fixture, a clamping assembly, and an outlet collection fixture. The air inlet fixture is connected to the needle valve via a pipe, and the outlet gas flow meter is located inside the outlet collection fixture. Through the above scheme, the gas source is driven by a pump and flows sequentially through a manual valve, a pressure reducing valve for temperature and pressure, and a needle valve for fine flow regulation, finally entering the testing fixture. Two pressure sensors monitor the pressure before and after the pressure reducing valve to ensure the stability and accuracy of pressure regulation. An inlet gas flow meter is installed after the needle valve and before the testing fixture to directly measure the total gas flow rate entering the testing fixture. An outlet gas flow meter is installed in the outlet collection fixture of the testing fixture to measure the gas flow rate exiting the spray head. The data acquisition system is responsible for collecting data from all sensors in real time. This provides a complete testing process framework, enabling precise control and regulation of the input gas pressure and flow rate, and synchronously and accurately collecting total inlet flow rate, total outlet flow rate, and key node pressure data, laying the foundation for evaluating the overall performance of the spray head.
[0006] Furthermore, a filter is provided on the pipe connecting the air intake fixture and the needle valve. The air intake fixture is in the form of a cover and is connected to the filter through a pipe. The air intake flow meter is located between the output end of the needle valve and the filter. With the above method, the gas passes through a filter installed on the pipeline before entering the inlet fixture. The filter is located after the needle valve output and before the inlet fixture, and its main function is to intercept any particulate matter or other impurities that may be present in the gas. This effectively protects the subsequent detection fixtures and flow meters from contamination or blockage, improving the reliability of the detection and extending the equipment's lifespan.
[0007] Furthermore, the outlet collection fixture includes a sealing gasket, which is adapted to fit the lower end cap of the porous structure of the spray head. The sealing gasket has several sampling ports, and the outlet gas flow meter is connected to the sampling ports through a pipe joint. The above-described design utilizes a porous structure on the spray head disc to allow gas passage, with a sealing gasket mounted on its lower end. Multiple sampling ports corresponding to the openings in the lower end cap are located on the sealing gasket. When the spray head is installed and clamped onto the testing fixture, the outlet surface of the spray head is in close contact with the sealing gasket. Gas escaping from different areas of the spray head is guided through the porous structure of the lower end cap to the corresponding specific sampling port on the sealing gasket. Each sampling port is connected to an independent outlet gas flow meter via a pipe connector. This specific sampling port design on the sealing gasket enables independent collection of the outlet gas flow rate from different local areas of the spray head. The porous lower end cap and sealing gasket together form a zoned collection chamber, ensuring that gas exiting from specific locations on the spray head can be accurately captured and measured by the corresponding sampling port. This is a key structural feature for detecting the uniformity of gas distribution in the spray head.
[0008] Furthermore, taking the center of the lower end cap of the porous structure of the spray head as a reference, one sampling area is set at the center position. Around the central area, three rings of sampling areas are set from the inside out, and four, four, and six sampling areas are set from the inside out respectively. The number of sampling ports opened on the sealing gasket is the same as the number of sampling areas to be tested and they correspond one-to-one. Using the above scheme, when opening sampling ports on the sealing gasket, a specific area division is strictly followed, with the center of the spray head as the reference: one sampling port at the center, four sampling ports evenly distributed around the center in the first ring, four sampling ports evenly distributed in the second ring, and six sampling ports evenly distributed in the outermost ring, totaling 15 sampling areas. Each sampling port precisely corresponds to a test area of the spray head, simulating the typical characteristics of gas distribution in the spray head (center, transition zone, edge zone), making the distribution of sampling points more scientific and reasonable, and able to more comprehensively reflect the flow distribution on the entire gas outlet surface of the spray head. Targeted setting of the number of sampling points helps to more accurately capture potential edge effects or uneven distribution problems, significantly improving the representativeness and accuracy of the detection.
[0009] Furthermore, the clamping assembly includes a locating pin located at the edge of the sealing gasket, and also includes a screw and a nut, wherein the screw passes through the locating pin and the air intake fixture and is secured by the nut. In the above method, a locating pin is installed on the edge of the sealing gasket for precise positioning of the spray head. During installation, the spray head is placed between the sealing gasket and the air inlet fixture, with the air inlet fixture covering the top of the spray head. Screws are passed through the holes on the locating pin and the air inlet fixture in sequence, and then tightened with nuts. By tightening the nuts, the air inlet fixture and the spray head are securely clamped and fixed together.
[0010] Furthermore, the sampling fixture is made of corrosion-resistant metal material, and its inner wall is polished with a roughness Ra≤0.8μm. The above-described scheme provides a smooth flow path through the polished surface (Ra≤0.8μm) of the inner wall of the testing fixture as the gas flows through and out of the internal channels and chambers. This ensures the long-term durability of the fixture in potentially corrosive process gas environments. The high-grade polishing significantly reduces gas adsorption and turbulence on the wall surface, lowers flow resistance, and guarantees the stability of gas flow and the accuracy of measurement results.
[0011] A detection method using the above-mentioned device is characterized by comprising the following steps: Step S1: Assemble the detection device, set the inlet pressure as needed, preliminarily adjust the inlet and outlet flow rates, test the compatibility of each sensor with the data acquisition system, and verify the zero drift and range error of the pressure sensor, the inlet gas flow meter and the outlet gas flow meter through the data acquisition system. Step S2: Match the corresponding sealing gasket according to the area to be tested, fix the spray head to be tested on the testing fixture through the clamping assembly, calibrate the sampling port and the sampling area to be tested, and connect the sampling port and the outlet gas flow meter. Step S3: Turn on the gas supply system, initially test the pressure and introduce gas. After the airflow is stable, the data acquisition system collects data from the pressure sensor, the inlet gas flow meter and the outlet gas flow meter, and performs initial screening on the collected data to remove outliers. Step S4: Collect statistical sampling data, calculate the average and variance of these data, compare the calculated average and variance with the preset standard values, and only when the calculation results meet the set values can the sprinkler head be judged as qualified. Step S5: Display the gas flow data, average value, variance, and comparison results with the standard value of each sampling area in the form of charts, and provide data comparison and analysis functions to make horizontal comparisons of the test data of different batches and different models of spray heads. The above method involves several steps: First, S1: Assemble the device, set the target inlet pressure, adjust the valves to ensure adequate flow, and verify the accuracy of all sensors. Then, S2: Select matching sealing gaskets based on the spray head size and the area to be tested, install the spray head, and secure it with clamping components, ensuring the sampling port is aligned with the target area. Connect the outlet flow meters to each sampling port. Next, S3: Turn on the air source and slowly increase the pressure to the set value. Once the airflow stabilizes, the data acquisition system begins recording all pressure and flow data and performs initial screening to remove outliers. Then, S4: Perform statistical analysis on the valid data from each sampling area and compare it with the preset standard range to comprehensively determine whether the spray head is qualified. Finally, S5: Display the test results in graphical form and provide cross-batch and cross-model data comparison and analysis functions. This method provides a standardized and operable testing process, covering the entire process from preparation, installation, testing to result analysis and reporting.
[0012] Furthermore, in step S3, the initial screening of data specifically includes outlier detection and processing of several collected data points. The three-standard-deviation principle is adopted. If the difference between a certain data point and the average value exceeds three times the standard deviation, the data point is determined to be an outlier and is removed to avoid outlier data having a significant impact on the final result. The system automatically optimizes the outlier determination threshold based on multiple detection data.
[0013] The above approach significantly improves the robustness of data processing. The three-standard-deviation principle is a statistically effective method for identifying significantly deviating data. It automatically filters out invalid or erroneous data points caused by transient disturbances (such as minor vibrations, electromagnetic interference, and instantaneous airflow fluctuations), preventing these outliers from distorting the final statistical results (especially variance), thus obtaining data that better represents the true flow rate level under stable conditions. The system's ability to self-learn and optimize thresholds further enhances the adaptability and intelligence of the method, enabling it to maintain optimal outlier identification performance under different operating conditions or equipment states.
[0014] Furthermore, in step S4, if the average value is within ±5% of the set standard average value and the variance is less than the set maximum variance value, the spray head is deemed qualified. The pre-set standard value is derived from the statistical analysis of multiple measurement data of imported spray heads under the same test conditions. At the same time, combined with the actual requirements of the semiconductor CVD process for gas flow, a reasonable fluctuation range is set to make the verification results more referential and practical. The standard value threshold is automatically corrected by analyzing historical qualified data through a machine learning model, and the standard parameters are adjusted in a timely manner according to industry technology development and process improvement.
[0015] The aforementioned scheme ensures a scientifically rigorous judgment standard that effectively controls flow rates within a reasonable range while strictly managing flow fluctuations, guaranteeing that the spray head performance meets high-precision process requirements. The standard values are based on reliable sample data and actual process needs, ensuring the authority and practicality of the test results. The introduction of a machine learning model enables the standard to "self-evolve," adapting to equipment aging, environmental changes, or natural drift in test data patterns, keeping the judgment threshold optimal and reducing false positives and false negatives. Combined with a proactive adjustment mechanism for industry and process development, this ensures that the testing standard keeps pace with the times and continuously meets the requirements of advanced processes.
[0016] The beneficial effects of this invention are as follows: 1. This invention features a simple structure. The porous structure of the spray head disc allows gas to pass through, and a sealing gasket is installed on its lower end face. Multiple sampling ports corresponding to the holes in the lower end cover are provided on the sealing gasket. When the spray head is installed and clamped onto the testing fixture, the outlet surface of the spray head is in close contact with the sealing gasket. Gas escaping from different areas of the spray head is guided to the corresponding specific sampling port on the sealing gasket after passing through the porous structure of the lower end cover. Each sampling port is connected to an independent outlet gas flow meter via a pipe connector. The specific sampling port design on the sealing gasket enables independent collection of the outlet gas flow rate from different local areas of the spray head. The porous lower end cover and the sealing gasket together form a partitioned collection chamber, ensuring that gas exiting from a specific location on the spray head can be accurately captured and measured by the corresponding sampling port. This is a key structural feature for detecting the uniformity of gas distribution in the spray head. 2. First, in step S1, assemble the device, set the target inlet pressure, initially adjust the valves to achieve a suitable flow rate, and verify the accuracy of all sensors. Then, in step S2, select matching sealing gaskets based on the spray head size and the area to be tested, install the spray head, and secure it with the clamping assembly, ensuring the sampling port is aligned with the target area. Connect the outlet flow meters to each sampling port. Next, in step S3, turn on the air source and slowly increase the pressure to the set value. After the airflow stabilizes, the data acquisition system begins recording all pressure and flow data and performs initial screening to remove outliers. Then, in step S4, statistically analyze the valid data from each sampling area and compare it with the preset standard range to comprehensively determine whether the spray head is qualified. Finally, in step S5, display the test results in graphical form and provide cross-batch and cross-model data comparison and analysis functions. This method provides a standardized and operable testing process, covering the entire process from preparation, installation, testing to result analysis and reporting. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the structure of the present invention, in which the detection tooling part is... Figure 2 The text appears to be a mix of Chinese characters and symbols, possibly representing a corrupted or incomplete document. A direct translation wouldn't be meaningful. Figure 2 This is a schematic diagram of the cross-sectional structure of the detection tooling part of the present invention; Figure 3 This is a schematic diagram of the sampling area division of the present invention.
[0018] Reference numerals: 11. Pump; 12. Hand valve; 13. Pressure reducing valve; 14. Needle valve; 15. Data acquisition system; 16. Pressure sensor; 17. Inlet gas flow meter; 18. Outlet gas flow meter; 19. Inlet fixture; 20. Filter; 21. Sealing gasket; 22. Sampling port; 23. Positioning pin; 24. Screw; 25. Nut; 26. Sampling area. Detailed Implementation
[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0020] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.
[0021] Example 1 like Figure 1 As shown, this embodiment provides a gas spray head detection device, including a data acquisition component, a gas source delivery component, and a detection fixture; The gas supply delivery assembly includes a pump 11 connected to a gas source. The output end of the pump 11 is sequentially connected to a hand valve 12, a pressure reducing valve 13, and a needle valve 14. The output end of the needle valve 14 is connected to a testing fixture. The data acquisition components include a data acquisition system 15, a pressure sensor 16, an inlet gas flow meter 17, and an outlet gas flow meter 18. There are two pressure sensors 16, one of which is located between the hand valve 12 and the pressure reducing valve 13, and the other is located between the pressure reducing valve 13 and the needle valve 14. The inlet gas flow meter 17 is located between the output end of the needle valve 14 and the detection fixture, and the outlet gas flow meter 18 is located at the output end of the detection fixture. The testing fixture includes an inlet fixture 19, a clamping assembly, and an outlet collection fixture. The inlet fixture 19 is connected to the needle valve 14 via a pipe, and the outlet gas flow meter 18 is located inside the outlet collection fixture.
[0022] Therefore, the gas source is driven by pump 11, flowing sequentially through hand valve 12, pressure reducing valve 13 (temperature and pressure), and needle valve 14 (fine flow adjustment), finally entering the testing fixture. Two pressure sensors 16 monitor the pressure before and after pressure reducing valve 13 to ensure the stability and accuracy of pressure regulation. An inlet gas flow meter 17 is positioned after needle valve 14 and before the testing fixture, directly measuring the total gas flow entering the testing fixture. An outlet gas flow meter 18 is installed in the outlet collection fixture of the testing fixture to measure the gas flow exiting the spray head. The data acquisition system 15 is responsible for real-time acquisition of data from all sensors. This provides a complete testing process framework, enabling precise control and regulation of the input gas pressure and flow rate, and synchronously and accurately acquiring total inlet flow rate, total outlet flow rate, and key node pressure data, laying the foundation for evaluating the overall performance of the spray head.
[0023] Reference Figure 1 The clamping assembly includes a locating pin 23 located on the edge of the sealing gasket 21, and also includes a screw 24 and a nut 25. The screw 24 passes through the locating pin 23 and the air inlet fixture 19 and is secured by the nut 25. The locating pin 23 is installed on the edge of the sealing gasket 21 for precise positioning of the spray head. During installation, the spray head is placed between the sealing gasket 21 and the air inlet fixture 19, with the air inlet fixture 19 covering the top of the spray head. The screw 24 passes through the holes in the locating pin 23 and the air inlet fixture 19 in sequence, and then is tightened using the nut 25. By tightening the nut 25, the air inlet fixture 19 and the spray head are securely clamped and fixed together.
[0024] Reference Figure 1The outlet collection fixture includes a sealing gasket 21, which fits snugly against the lower end cap of the porous structure of the spray head. Several sampling ports 22 are provided on the sealing gasket 21. The outlet gas flow meter 18 is connected to the sampling ports 22 via a pipe connector. A sampling area 26 is set at the center of the lower end cap of the porous structure of the spray head, and three concentric rings of sampling areas 26 are set around the center, with 4, 4, and 6 sampling areas 26 respectively from the inside out. The number of sampling areas 26 can be flexibly adjusted as needed, from 1 to 15. The number of sampling ports 22 on the sealing gasket 21 corresponds one-to-one with the number of sampling areas 26 to be tested. Therefore, the porous structure of the spray head allows gas to pass through, and the lower end face is fitted with the sealing gasket 21. Multiple sampling ports 22 corresponding to the holes in the lower end cap are provided on the sealing gasket 21. When the spray head is installed and clamped onto the testing fixture, the outlet surface of the spray head is tightly fitted with the sealing gasket 21. Gas escaping from different areas of the spray head is guided to specific sampling ports 22 on the sealing gasket 21 after passing through the porous structure of the lower end cap. Each sampling port 22 is connected to an independent outlet gas flow meter 18 via a pipe connector. The design of specific sampling ports 22 on the sealing gasket 21 enables independent collection of the outlet gas flow rate from different local areas of the spray head. The porous lower end cap and the sealing gasket 21 together form a partitioned collection chamber, ensuring that gas exiting from specific locations on the spray head can be accurately captured and measured by the corresponding sampling port 22. This is a key structure for detecting the uniformity of gas distribution in the spray head. When opening the sampling ports 22 on the sealing gasket 21, a specific area division is strictly followed, with the center of the spray head as the reference: one sampling port 22 at the center, four sampling ports 22 evenly distributed around the center in the first ring, four sampling ports 22 evenly distributed in the second ring, and six sampling ports 22 evenly distributed in the outermost ring, totaling 15 sampling areas 26. Each sampling port 22 precisely corresponds to a test area of the spray head, simulating the typical characteristics of gas distribution in the spray head (center, transition zone, edge zone). This makes the distribution of sampling points more scientific and reasonable, and can more comprehensively reflect the flow distribution on the entire outlet surface of the spray head. Targeted setting of the number of sampling points helps to more accurately capture potential edge effects or uneven distribution problems, significantly improving the representativeness and accuracy of the detection.
[0025] Reference Figure 1To improve the stability of the device, the sampling fixture is made of corrosion-resistant metal material, and its inner wall is polished to a roughness Ra≤0.8μm. A filter 20 is installed on the pipe connecting the inlet fixture 19 and the needle valve 14. The inlet fixture 19 is a cover and is connected to the filter 20 through a pipe. The inlet gas flow meter 17 is located between the output end of the needle valve 14 and the filter 20. Before entering the inlet fixture 19, the gas passes through the filter 20 installed on the pipe. The filter 20 is located after the output end of the needle valve 14 and before the inlet fixture 19. Its main function is to intercept any particulate matter or other impurities that may be present in the gas. This effectively protects the subsequent detection fixture and flow meter from contamination or blockage, improving the reliability of the detection and the lifespan of the equipment. As the gas flows through the internal channels and chambers of the detection fixture and flows out, the polished surface (Ra≤0.8μm) of the inner wall of the fixture provides a smooth flow path. This ensures the long-term durability of the fixture in potentially corrosive process gas environments. High-grade polishing greatly reduces gas adsorption and turbulence on the wall surface, lowers flow resistance, and ensures the stability of gas flow and the accuracy of measurement results.
[0026] The present invention also provides a detection method using the above-described device, comprising the following steps: Step S1: Assemble the detection device, set the inlet pressure as needed, perform preliminary debugging of the inlet and outlet flow rates, test the compatibility of each sensor with the data acquisition system 15, and verify the zero drift and range error of the pressure sensor 16, the inlet gas flow meter 17 and the outlet gas flow meter 18 through the data acquisition system 15. Step S2: Match the corresponding sealing gasket 21 according to the area to be tested, fix the spray head to be tested on the testing fixture through the clamping assembly, calibrate the sampling port 22 and the sampling area 26 to be tested, and connect the sampling port 22 and the gas flow meter 18. Step S3: Turn on the gas supply system, initially test the pressure and introduce gas. After the airflow is stable, the data acquisition system 15 collects data from the pressure sensor 16, the inlet gas flow meter 17 and the outlet gas flow meter 18, and performs initial screening on the collected data to remove outliers. Step S4: Collect statistical sampling data, calculate the average and variance of these data, compare the calculated average and variance with the preset standard values, and only when the calculation results meet the set values can the sprinkler head be judged as qualified. Step S5: Display the gas flow data, average value, variance, and comparison results with the standard value of each sampling area 26 in the form of charts, and provide data comparison and analysis functions to make horizontal comparisons of the test data of different batches and different models of spray heads.
[0027] Specifically, it further includes the following content. In step S3, the preliminary screening data specifically includes detecting and processing outliers in a number of collected data. Using the three-standard-deviation principle, if the difference between a certain data and the average value exceeds three standard deviations, then this data is determined to be an outlier and is excluded to avoid the large impact of abnormal data on the final result. The system automatically optimizes the outlier determination threshold according to multiple detection data. In step S4, if the average value is within the range of ±5% of the set standard average value and the variance is less than the set maximum variance value, then it is determined that the sprinkler head is qualified. The preset standard value is obtained through statistical analysis of multiple measurement data of imported sprinkler heads under the same test conditions, and at the same time, combined with the actual requirements of the semiconductor CVD process for gas flow, a reasonable fluctuation range is set to make the verification result more referenceable and practical. By analyzing historical qualified data through a machine learning model, the standard value threshold is automatically corrected, and according to the development of industry technology and process improvement, the standard parameters are adjusted in a timely manner. In step S3, this method significantly improves the robustness of data processing. The three-standard-deviation principle is an effective method in statistics to identify significantly deviating data, which can automatically filter out invalid or incorrect data points caused by instantaneous interference (such as minor vibrations, electromagnetic interference, and instantaneous fluctuations in air flow), avoiding these outliers from distorting the final statistical results (especially the variance), and thus obtaining data that can better represent the true flow level under stable conditions. The system's ability to self-learn and optimize the threshold further enhances the adaptability and intelligence level of this method, enabling it to maintain the best outlier identification effect under different working conditions or equipment states. In step S4, this determination criterion is scientific and rigorous, which can not only control the flow level within a reasonable range but also strictly control the flow fluctuation to ensure that the performance of the sprinkler head meets the high-precision process requirements. The standard value is formulated based on reliable sample data and actual process requirements, ensuring the authority and practicality of the detection results. The introduction of the machine learning model realizes the "self-evolution" of the standard, which can adapt to equipment aging, environmental changes, or the natural drift of the detection data pattern, keeping the determination threshold always optimal, reducing misjudgments and missed judgments. The active adjustment mechanism combined with industry and process development ensures that the detection standard keeps pace with the times and continuously meets the requirements of advanced processes.
[0028] Implementation Principle: First, in step S1, assemble the device, set the target inlet pressure, initially adjust the valves to achieve a suitable flow rate, and verify the accuracy of all sensors. Then, in step S2, select matching sealing gaskets 21 according to the spray head size and the area to be tested, install the spray head, and secure it with clamping components, ensuring that the sampling port 22 is aligned with the target area, and connect the outlet flow meters of each sampling port 22. Next, in step S3, turn on the air source and slowly increase the pressure to the set value. After the airflow stabilizes, the data acquisition system 15 begins recording all pressure and flow data and performs initial screening to remove outliers. Then, in step S4, statistically analyze the valid data from each sampling area 26 and compare it with the preset standard range to comprehensively determine whether the spray head is qualified. Finally, in step S5, display the test results in graphical form and provide cross-batch and cross-model data comparison and analysis functions. This method provides a standardized and operable testing process, covering the entire process from preparation, installation, testing to result analysis and reporting.
[0029] It should be noted that the connection relationships of components not specifically mentioned in this application are all assumed to be based on existing technology. Since they do not involve the inventive point and are commonly used in existing technology, the structural connection relationships are not described in detail.
Claims
1. A gas spray head detection device, characterized in that, Includes data acquisition components, gas supply delivery components, and testing fixtures; The gas supply delivery assembly includes a pump (11) connected to a gas source. The output end of the pump (11) is sequentially connected to a hand valve (12), a pressure reducing valve (13), and a needle valve (14). The output end of the needle valve (14) is connected to the detection fixture. The data acquisition components include a data acquisition system (15), a pressure sensor (16), an inlet gas flow meter (17), and an outlet gas flow meter (18). There are two pressure sensors (16), one of which is located between the hand valve (12) and the pressure reducing valve (13), and the other is located between the pressure reducing valve (13) and the needle valve (14). The inlet gas flow meter (17) is located between the output end of the needle valve (14) and the detection fixture. The outlet gas flow meter (18) is located at the output end of the detection fixture. The detection fixture includes an inlet fixture (19), a clamping assembly, and an outlet collection fixture. The inlet fixture (19) is connected to the needle valve (14) via a pipe, and the outlet gas flow meter (18) is located inside the outlet collection fixture.
2. The gas spray head detection device according to claim 1, characterized in that, The air intake fixture (19) is connected to the needle valve (14) by a filter (20). The air intake fixture (19) is a cover and is connected to the filter (20) by a pipe. The air intake flow meter (17) is located between the output end of the needle valve (14) and the filter (20).
3. The gas spray head detection device according to claim 2, characterized in that, The outlet collection fixture includes a sealing gasket (21), which is adapted to fit the lower end cover of the porous structure of the spray head. The sealing gasket (21) has several sampling ports (22), and the outlet gas flow meter (18) is connected to the sampling ports (22) through a pipe joint.
4. The gas spray head detection device according to claim 3, characterized in that, Taking the center of the lower end cover of the multi-hole structure of the spray head as a reference, a sampling area (26) is set at the center position. Around the center area, three sampling areas (26) are set from the inside out. Four, four, and six sampling areas (26) are set from the inside out respectively. The number of sampling ports (22) on the sealing gasket (21) is the same as the number of sampling areas (26) to be tested and they correspond one-to-one.
5. A gas spray head detection device according to claim 3, characterized in that, The clamping assembly includes a locating pin (23) located on the edge of the sealing gasket (21), and also includes a screw (24) and a nut (25). The screw (24) passes through the locating pin (23) and the air intake fixture (19) and is secured by the nut (25).
6. The gas spray head detection device according to claim 3, characterized in that, The sampling fixture is made of corrosion-resistant metal material, and its inner wall is polished with a roughness Ra≤0.8μm.
7. A detection method using the apparatus according to any one of claims 1 to 6, characterized in that, Includes the following steps: Step S1: Assemble the detection device, set the inlet pressure as needed, preliminarily adjust the inlet and outlet flow rates, test the compatibility of each sensor with the data acquisition system (15), and verify the zero drift and range error of the pressure sensor (16), inlet gas flow meter (17) and outlet gas flow meter (18) through the data acquisition system (15). Step S2: Match the corresponding sealing gasket (21) according to the area to be tested, fix the spray head to be tested on the testing fixture through the clamping assembly, calibrate the sampling port (22) and the sampling area to be tested (26), and connect the sampling port (22) and the gas flow meter (18). Step S3: Turn on the gas supply system, initially test the pressure and introduce gas. After the airflow is stable, the data acquisition system (15) collects data from the pressure sensor (16), the inlet gas flow meter (17) and the outlet gas flow meter (18), and performs initial screening on the collected data to remove outliers. Step S4: Collect statistical sampling data, calculate the average and variance of these data, compare the calculated average and variance with the preset standard values, and only when the calculation results meet the set values can the sprinkler head be judged as qualified. Step S5: Display the gas flow data, average value, variance and comparison results with the standard value of each sampling area (26) in the form of charts, provide data comparison and analysis function, and make horizontal comparison of the test data of different batches and different models of spray heads.
8. The gas spray head detection method according to claim 7, characterized in that, In step S3, the initial screening of data specifically includes outlier detection and processing of several collected data points. The three-standard-deviation principle is adopted. If the difference between a certain data point and the average value exceeds three times the standard deviation, the data point is determined to be an outlier and is removed to avoid outlier data having a significant impact on the final result. The system automatically optimizes the outlier determination threshold based on multiple detection data.
9. A method for detecting a gas spray head according to claim 7, characterized in that, In step S4, if the average value is within ±5% of the set standard average value and the variance is less than the set maximum variance value, the spray head is deemed qualified. The pre-set standard value is derived from the statistical analysis of multiple measurement data of imported spray heads under the same test conditions. At the same time, combined with the actual requirements of the semiconductor CVD process for gas flow, a reasonable fluctuation range is set to make the verification results more referential and practical. The standard value threshold is automatically corrected by analyzing historical qualified data through a machine learning model. The standard parameters are adjusted in a timely manner according to industry technology development and process improvement.