Pulse jet flow rate adjustment device, adjustment method, and program
By using pressure sensors and optimal filling time storage processing in the pulse jet flow adjustment device, the switching operation of the first and second shut-off valves is optimized, and the problem of long response time for high-temperature gas flow adjustment in the prior art is solved, and the problem of long response time for high-temperature gas flow adjustment in the prior art is realized. Faster and accurate flow control.
Patent Information
- Application Number
- CN202180026240.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-03-30
- Filing Date
- 2021-03-02
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2041-03-02
AI Technical Summary
The existing pulse jet flow adjustment device has a longer response time in high-temperature gas control, and it is impossible to quickly adjust the gas flow to the target flow.
In the pulse jet flow adjustment device, the pressure sensor is used to measure the pressure after filling and the pressure after injection, and the pulse jet method is adjusted, and the switching actions of the first and second shut-off valves are optimized in combination with the optimal filling time storage and filling time control. , improve the responsiveness of gas flow.
It significantly shortens the time from the start to the target flow to the stable response, and improves the responsiveness and accuracy of flow adjustments.
Smart Images

Figure CN115461693B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a pulse jet flow rate adjustment device, a pulse jet flow rate adjustment method, and a program for adjusting the volume flow rate of a gas. Background Art
[0002] In the past, in order to accurately control the gas flow rate in gas supply systems such as semiconductor manufacturing equipment, thermal mass flow controllers were used, for example. However, in recent years, high-temperature gases above 100°C have become the control targets. Thermal mass flow controllers have difficulty controlling such high-temperature gases.
[0003] For example, Patent Document 1 discloses a pulse jet flow control device capable of controlling the flow rate of high temperature gas. The pulse jet flow control device includes a first stop valve connected to a gas source, a second stop valve connected to the first stop valve, a gas-filled container located between the first stop valve and the second stop valve, a pressure sensor for measuring the pressure of the gas-filled container, and a controller.
[0004] Controllers for pulse jet flow control devices, such as Fig. 9 As shown, after the first stop valve is switched, the second stop valve is switched repeatedly. At the same time, the controller uses a pressure sensor to change the filling time of the gas filling by maintaining the opening operation of the first stop valve in the next pulse injection based on the pressure difference D1 between the pressure of the gas filling container filled with gas after the first stop valve is switched (pressure after filling) P1 and the pressure of the gas filling container from which the gas has been ejected after the second stop valve is switched (pressure after ejection) P2, thereby changing the pulse injection method and adjusting the gas flow rate to the target flow rate.
[0005] Existing patent literature
[0006] Patent Literature
[0007] Patent Document 1: Japanese Patent No. 4197648 Summary of the invention
[0008] Problems to be solved by the present invention:
[0009] However, there are problems in the prior art. In other words, in the case of a process in which the gas is repeatedly supplied by repeating the pulse jet, the gas usage conditions are different in each process. On the other hand, the prior art pulse jet flow control device makes the initial pulse jet in each process the same, such as Fig.10As shown, the pulse jetting mode is slowly changed according to the pressure difference D1 during the pulse jetting. Therefore, the conventional pulse jetting flow rate control device has a long response time from the start of the process to the adjustment of the gas flow rate to the target flow rate.
[0010] The present invention is intended to solve the above-mentioned problem, and its object is to provide a technology capable of improving responsiveness for a pulse jet flow rate control device that controls a gas flow rate to a target flow rate by repeated pulse jetting.
[0011] Means used to solve the problem:
[0012] One aspect of the present invention has the following structure. (1) It is characterized in that in a pulse jet flow control device including a first stop valve connected to a gas source, a second stop valve connected to the first stop valve, a gas filling container located between the first stop valve and the second stop valve, and a pressure sensor for measuring the pressure of the gas filling container, a process of controlling the gas to a target flow rate is performed multiple times. The first stop valve, the second stop valve, and the pressure sensor are communicatively connected to a controller for controlling the operation of the pulse jet flow control device. The controller performs, in each process, a flow control process for adjusting the gas flow rate to a target flow rate by calculating the volume flow rate of the gas ejected from the gas filling container and changing the manner of repeating the pulse jetting while repeating the pulse jetting of the second stop valve after the first stop valve is opened and closed, and by adjusting the gas flow rate to a target flow rate by changing the manner of repeating the pulse jetting while repeating the pulse jetting of the second stop valve after the first stop valve is opened and closed, and by changing the manner of repeating the pulse jetting while repeating the pulse jetting of the second stop valve. In addition, the controller: performs an optimal filling time storage process, in which, in the flow control process, when a pulse injection in which the flow rate has been adjusted to a target flow rate is performed, the first shutoff valve is opened and closed, and the filling time of the gas filling container filled with the gas is stored as the optimal filling time; and performs a filling time control process, in which, in the flow control process before the next process to the process in which the optimal filling time storage process has been performed, in the case of an initial pulse injection, the first shutoff valve is opened and closed using the optimal filling time stored in the optimal filling time storage process.
[0013] The pulse flow rate control device having the above structure, in the case of executing the first pulse injection of the process, uses the optimal filling time stored in the previous process to open and close the first stop valve. As a result, in the first pulse injection, the gas suitable for adjusting the volume flow rate of the gas ejected from the second stop valve to the target flow rate is filled into the gas filling container, and the volume flow rate of the gas ejected from the second stop valve is more likely to be adjusted to the target flow rate or to a value close to the target flow rate. Therefore, according to the pulse injection type flow rate control device having the above structure, the response time from the start of the process to the gas being stabilized to the target flow rate is shortened, and the responsiveness can be improved.
[0014] (2) In the pulse-jet flow rate control device as described in (1), preferably, in the optimal filling time storage process, the filling time of the last pulse-jet is stored as the optimal filling time.
[0015] The pulse jet flow control device having the above structure can reduce the processing load and memory load when storing the optimal filling time by storing the filling time measured in the last pulse jet that is likely to adjust the volume flow to the target flow as the optimal filling time.
[0016] (3) In the pulse jet flow rate adjustment device as described in (1) or (2), preferably, the controller performs a gas replacement process to replace the gas remaining in the gas filling container by simultaneously opening the first stop valve and the second stop valve.
[0017] In the pulse jet flow control device having the above structure, when gas replacement is performed, the first stop valve and the second stop valve are opened simultaneously, so that the gas supplied from the gas source flows into the gas filling container and is easily replaced with the gas remaining in the gas filling container. Thus, the gas can be replaced faster than when the first stop valve and the second stop valve are opened and closed alternately to perform gas replacement.
[0018] (4) In the pulse jet flow rate adjustment device as described in (3), preferably, the controller performs a pressure adjustment process, and after the replacement process, the first stop valve and the second stop valve that are in the opening action are closed in such a manner that the pressure of the gas filling container is adjusted to the target pressure of the gas filling container when the flow rate control process is performed.
[0019] The pulse jet flow rate control device having the above structure adjusts the pressure of the gas filling container to the target pressure of the gas filling container when the flow rate control process is performed, depending on the time point when the first stop valve and the second stop valve are closed, after the first stop valve and the second stop valve are opened at the same time to replace the gas in the gas filling container. Therefore, in the process after the gas replacement, since the pressure of the gas filling container is controlled to the target pressure from the first pulse jet, the delay of the response time due to the pressure deviation of the gas filling container can be suppressed.
[0020] (5) In the pulse jet flow rate adjustment device as described in any one of (1) to (4), preferably, the controller: performs a learning process, in which the first stop valve is opened and closed to make the filling time before the gas filling container is filled with the gas variable, and pulse jet is repeatedly performed while the second stop valve is opened and closed to make the ejection time of the gas from the gas filling container constant, and the filling time for each volume flow rate of the gas ejected from the second stop valve is calculated, and the relationship between the filling time and the volume flow rate is learned; in the flow rate control process, a correction process is performed, in which, when the target flow rate is changed from the target flow rate used in the previous process, or when the target flow rate is changed during the execution of the flow rate control process, the optimal filling time is corrected based on the relationship between the filling time and the volume flow rate learned in the learning process according to the difference between the target flow rate and the changed target flow rate.
[0021] The pulse jet flow adjustment device having the above structure can quickly adjust the gas flow to the subsequent target flow by correcting the optimal filling time based on the pre-learned relationship between the filling time and the volume flow rate, such as when the target flow rate is changed from the target flow rate used in the previous process or is changed during the execution of the flow control process.
[0022] (6) In the pulse jet flow rate control device as described in (3) or (4), preferably, the controller: performs a learning process, in which the first shutoff valve is opened and closed to fill the gas to the gas filling volume with a variable filling time, and pulse jet is repeated while the second shutoff valve is opened and closed to fill the gas from the gas filling container with a constant ejection time, and the filling time is calculated for each volume flow rate of the gas ejected from the second shutoff valve, and the relationship between the filling time and the volume flow rate is learned; in the flow control process, a correction process is performed, in which, when the target flow rate is changed from the target flow rate used in the previous process, or when the target flow rate is changed during the execution of the flow control process, based on the relationship between the filling time and the volume flow rate learned in the learning process, the optimal filling time is corrected according to the difference between the target flow rate and the changed target flow rate. In addition, the controller: performs a mode setting process, in which a flow control mode for executing the flow control process, a gas replacement mode for executing the replacement process, and a learning mode for executing the learning process are set; and a process according to the mode set in the mode setting process is performed.
[0023] The pulse jet flow rate adjustment device having the above structure performs flow control processing, replacement processing, and learning processing according to the settings of the flow control mode, gas replacement mode, and learning mode. Each processing can be performed at any time point, and has good usability.
[0024] Another aspect of the present invention is (7) a pulse jet flow rate adjustment method, characterized in that it is used in a pulse jet flow rate adjustment device, wherein the pulse jet flow rate adjustment device includes a first stop valve connected to a gas source, a second stop valve connected to the first stop valve, a gas filling container located between the first stop valve and the second stop valve, and a pressure sensor for measuring the pressure of the gas filling container, wherein the pulse jet flow rate adjustment device uses a controller that is communicably connected to the first stop valve, the second stop valve, and the pressure sensor to control the operation and perform a plurality of processes of controlling the gas to a target flow rate. In each process, a flow rate control step is performed, based on the post-filling pressure measured by the pressure sensor after filling the gas filling container with gas by switching the first stop valve and repeating the pulse jet of switching the second stop valve after switching the first stop valve, and the post-ejection pressure measured by the pressure sensor after switching the second stop valve and ejecting the gas from the gas filling volume, by calculating the volume flow rate of the gas ejected from the gas filling container and changing the method of the repeated pulse jet, thereby adjusting the volume flow rate to the target flow rate. In addition, an optimal filling time storage step is performed, in which, in the flow rate control step, when a pulse injection in which the volume flow rate has been adjusted to the target flow rate is performed, the first stop valve is opened and closed, and the filling time of the gas filling container filled with the gas is stored as the optimal filling time; and a filling time control step is performed, in which, in the flow rate control step of a process performed next to the process in which the optimal filling time storage step has been performed, when an initial pulse injection is performed, the first stop valve is opened and closed using the optimal filling time stored in the optimal filling time storage step.
[0025] In addition, another aspect of the present invention is (8), a program, characterized in that it is built into a controller for controlling the operation of a pulse jet flow rate adjustment device, wherein the pulse jet flow rate adjustment device includes a first stop valve connected to a gas source, a second stop valve connected to the first stop valve, a gas filling container located between the first stop valve and the second stop valve, and a pressure sensor for measuring the pressure of the gas filling container. In the case where the flow rate adjustment device performs a plurality of processes for controlling the gas to a target flow rate, in each process, the controller performs: a flow rate control process for adjusting the volume flow rate to a target flow rate by calculating the volume flow rate of gas ejected from the gas filling volume and changing the manner in which the pulse jetting is repeated while the first stop valve is opened and closed, and the pressure after ejection is measured using the pressure sensor after the gas is filled into the gas filling container by opening and closing the first stop valve, and the pressure after ejection is measured using the pressure sensor after the gas is ejected from the gas filling volume by opening and closing the second stop valve. In addition, the controller: performs an optimal filling time storage process, in which, in the flow control process, when a pulse injection in which the volume flow rate has been adjusted to the target flow rate is performed, the first stop valve is opened and closed, and the filling time of the gas filling container filled with the gas is stored as the optimal filling time; and performs a filling time control process, in which, in the flow control process before the next process to the process in which the optimal filling time process has been performed, in the case of an initial pulse injection, the first stop valve is opened and closed using the optimal filling time stored in the optimal filling time storage process.
[0026] Effects of the Invention
[0027] Therefore, according to the present invention, a technique for improving the responsiveness of a pulse jet type flow rate control device that controls the flow rate of a gas to a target flow rate by repeatedly performing pulse jets can be realized.
[0028] Brief description of the attached figure
[0029] Figure 1 It is a schematic structural diagram of a pulse jet flow rate adjustment device according to an embodiment of the present invention.
[0030] Figure 2 It is a flowchart explaining the procedure of control processing.
[0031] Figure 3 : is a flowchart explaining the process of gas replacement treatment.
[0032] Figure 4Ais an example of a control mode.
[0033] Figure 4B is an example of a control mode.
[0034] Figure 4C is an example of a control mode.
[0035] Figure 5 is a flowchart illustrating the process of flow control processing.
[0036] Figure 6 It is the functional block diagram of the controller.
[0037] Fig. 7A It is a schematic diagram explaining the correction process.
[0038] Figure 7B It is a schematic diagram explaining the correction process.
[0039] Figure 7C It is a schematic diagram explaining the correction process.
[0040] Figure 8 is a flowchart illustrating the procedure of the learning process.
[0041] Fig. 9 This is a sequence diagram of a conventional pulse jet flow rate regulating device.
[0042] Fig.10 This is a diagram showing an example of flow rate control by a conventional pulse jet flow rate adjustment device. DETAILED DESCRIPTION
[0043] Hereinafter, embodiments of the pulse jet flow rate adjustment device, the pulse jet flow rate adjustment method and the program of the present invention will be described based on the drawings.
[0044] Schematic structure of a pulse jet flow rate control device.
[0045] like Figure 1 As shown, in this embodiment, the present invention is applied to a pulse jet flow rate control device (hereinafter referred to as a “flow rate control device”) for controlling the flow rate supplied to the chamber 230 .
[0046] The chamber 230 is made into a vacuum environment by, for example, a vacuum pump 240, and an atomic layer deposition (ALD) method is used to form a predetermined film on the wafer. In ALD, by repeating the cycle of (a) introducing a raw material gas, (b) purging with an inert gas, (c) introducing a reaction gas, and (d) purging with an inert gas, the thickness of the film formed on the wafer can be adjusted in units of 0.1 μm.
[0047] In ALD, different types of gases coagulate when mixed. Therefore, a pipe 200 connected to a gas source 211 of a raw gas, a pipe 300 connected to a gas source 311 of a reaction gas, and a pipe 400 connected to a gas source 411 of an inert gas are connected to the chamber 230 .
[0048] In addition, the raw material gas is, for example, TMA (trimethylaluminum). The reaction gas is, for example, H 2 O (water vapor). Inert gases such as N 2 Gas. TMA is a solid material at room temperature, and is vaporized to become a high-temperature gas of 120° C. or higher when supplied to the chamber 230 .
[0049] When ALD is used for film formation, the supply amount of each gas is managed by supply time. In addition, in ALD, the above-mentioned cycle (a) to (d) is repeated hundreds of times on each wafer. In this case, for example, TMA and H are supplied in (a) and (c). 2 The supply time of O to the chamber 230 is several tens of milliseconds (msec), and the supply time of N in (b) and (d) is several tens of milliseconds (msec). 2 The supply time of gas to the chamber 230 is from several seconds to tens of seconds. Therefore, the gas supplied to the chamber 230 needs to be controlled with high precision, high frequency, and high speed. Here, flow control devices 1A, 1B, and 1C are provided in the pipes 200, 300, and 400 to control the flow rate by alternately opening and closing the first stop valve 11 and the second stop valve 13.
[0050] The structure of the flow rate regulating device 1 is shown in FIG. Figure 1 Since the flow control devices 1A to 1C are similarly configured, the following description will focus on the flow control device 1A disposed in the pipe 200. In addition, when there is no need to distinguish them in particular, they are collectively referred to as "flow control devices 1".
[0051] The flow control device 1 includes a first stop valve 11, a groove 12, a second stop valve 13, and a pressure sensor 14. The first stop valve 11 is arranged on the upstream side of the groove 12 and is connected to a pressurized gas source 211. The second stop valve 13 is arranged on the downstream side of the groove 12 and is connected to a chamber 230 that has become a vacuum environment. The groove 12 becomes a closed space by closing the first stop valve 11 and the second stop valve 13. The groove 12 of this embodiment has a volume of 5 to 1000cc, for example. The groove 12 is an example of a "gas-filled container". In addition, it is also possible to form a gas-filled container by piping instead of the groove 12. The pressure sensor 14 is a vacuum pressure gauge for high temperature. The filter 16 is arranged on the upstream side of the first stop valve 11 to remove foreign matter from the gas flowing into the flow control device 1.
[0052] The first stop valve 11 and the second stop valve 13 are air-operated on-off valves. Then, the first stop valve 11 and the second stop valve 13 are valves that can be controlled even in high-temperature gas of 120°C or more. In addition, the first stop valve 11 and the second stop valve 13 are high-speed valves that can be switched using a cycle of several milliseconds. In addition, the diameter of the valve used for piping is a quarter of an inch of the general piping diameter. However, the first stop valve 11 and the second stop valve 13 are selected to have a diameter of, for example, three-eighths of an inch of the piping diameter, and the value of the flow coefficient (Cv) is increased. The value of the flow coefficient (Cv) of the first stop valve 11 and the second stop valve 13 is, for example, 0.6. In this way, by arranging the first stop valve 11 and the second stop valve 13, which have a large value of the flow coefficient (Cv) and can be operated at high speed and high frequency, on the upstream and downstream sides of the groove 12, the flow control device 1 can suppress the pulsation accompanying the switching operation of the first stop valve 11 and the second stop valve 13. In addition, in the flow control device 1 , the gas can flow easily through the thermal mass flow controller having the throttle portion, and the gas can be supplied to the chamber 230 quickly.
[0053] During the process, the flow control device 1 controls the gas to a target flow rate while repeatedly performing pulse injection at a high speed to open and close the first stop valve 11 and then the second stop valve 13. In this specification, "target flow rate" is defined as the volume flow rate of the gas ejected from the second stop valve 13 per unit time. Therefore, the flow control device 1 is arranged near the chamber 230. For example, the second stop valve 13 and the chamber 230 are directly connected, or connected via a pipe of less than 2m. In addition, the first stop valve 11 and the second stop valve 13 are not limited to gas-operated switch valves, and electromagnetic switch valves may also be used.
[0054] The flow rate regulating device 1 of this embodiment is controlled in operation by a controller 15. The controller 15 is communicably connected to the first stop valve 11, the second stop valve 13, and the pressure sensor 14. The controller 15 is an example of a "controller".
[0055] The controller 15 is a well-known microcomputer, and includes a central processing unit (CPU) 21 and a memory 22. The memory 22 includes a non-volatile memory and a volatile memory. The memory 22 stores various programs and data. The CPU 21 temporarily stores data in the memory 22 while executing the program stored in the memory 22 to perform various processes.
[0056] For example, a control program 31 for controlling the operation of the flow control device 1 is stored in the memory 22. The control program 31 performs, for example, a gas replacement process for replacing the gas remaining in the groove 12 with the gas supplied from the gas source 211. In addition, the control program 31 performs a flow control process for controlling the flow rate of the gas. In addition, the control program 31 performs a learning process for learning the relationship between the filling time for filling the groove 12 with gas and the volume flow rate of the gas ejected from the second shutoff valve 13. With respect to the gas replacement process, the flow control process and the learning process will be described later. In addition, the control program 31 is an example of a "program".
[0057] In addition, the non-volatile memory of the memory 22 stores data used for the control program 31. For example, the memory 22 stores the optimal filling time used for the initial pulse injection in the flow control process. The optimal filling time, for example, stores an initial value when the control program 31 is installed, and is updated at any time by repeating the flow control process. For another example, the memory 22 stores the target flow rate used in the flow control process in the past process. For another example, the memory 22 stores a correction diagram of the relationship between the filling time and the volume flow rate learned in the learning process.
[0058] In addition, the controller 15 also includes a timer 23 for measuring time and a communication interface 24 (hereinafter referred to as "communication IF24") for controlling communication with the outside. The controller 15 is communicatively connected to the main controller 500 that controls the operation of the semiconductor manufacturing device via the communication IF24. The controller 15 may be connected to the main controller 500 via a line or wirelessly.
[0059] In such a flow control device 1, the filter 16, the first stop valve 11, the second stop valve 13, the pressure sensor 14 and the controller 15 are connected to each other and arranged in a housing (not shown), so that they can be unitized. In that case, the flow control device 1 is easy to operate, and there is no need to perform wiring work other than connecting the communication IF 24 to the main controller 500.
[0060] In addition, the controller 15 may be built into an external controller such as the main controller 500, or may be located outside a housing (not shown). In addition, for example, by building the control program 31 into the main controller 500, the main controller 500 may have the function of the controller 15, and the main controller 500 may be used as a "controller for controlling the operation of the flow control device".
[0061] Operation description of pulse jet flow control device: control processing
[0062] Next, the operation of the flow rate regulating device 1 will be described. Figure 2The flow control device 1 is triggered by the power being turned on, and the central processing unit 21 starts the control program 31 and executes the following steps: Figure 2 In this embodiment, the flow control device 1 receives a mode according to a command from the main controller 500 and executes a process corresponding to the mode. In addition, when the flow control device 1 has an operation unit for inputting information, the mode may be received using the operation unit.
[0063] The central processing unit 21 first determines whether a command is received (S10). For example, after the control program 31 is started, it sends a controllable notification indicating that control of the controllable flow regulating device 1 is possible to the main controller 500. The controllable notification includes identification information for identifying the flow regulating device 1. The main controller 500 that has received the controllable notification appropriately sends the command required for the flow control of the gas to the flow regulating device 1 corresponding to the identification number attached to the controllable notification. When the central processing unit 21 does not use the communication IF24 to receive the command sent from the main controller 500, it determines that the command has not been received (S10: No) and stands by.
[0064] In contrast, when the central processing unit 21 receives a command sent from the main controller 500 using the communication IF24, it determines that the command is received (S10: Yes). In this case, the central processing unit 21 analyzes the received command and determines the mode (S20). The processing of S20 is an example of "mode setting processing". Among the modes of this method, there are gas replacement mode, flow control mode and learning mode. In addition, the mode does not need to include all of these, for example, only the flow control mode is also acceptable. In addition, there are other modes other than these modes.
[0065] The gas replacement mode is a mode for performing gas replacement processing. For example, the flow control device 1 whose power is turned on has the possibility of air being mixed into the groove 12. The gas mixed into the air has a different molecular weight from the gas component in normal times, which may reduce the film forming quality. Here, the main controller 500 sends a gas replacement instruction for instructing the gas replacement processing to be performed to the flow control device 1 that has sent the controllable notification.
[0066] When the central processing unit 21 receives the gas replacement command sent from the main controller 500 using the communication IF 24, it determines that the gas replacement mode is set (S20: gas replacement mode) and executes the gas replacement process (S30). When the gas replacement process is completed, the central processing unit 21 returns to the process of S10. The gas replacement process will be described later.
[0067] The flow control mode is a mode for executing flow control processing. For example, when the main controller 500 is executing a process of supplying TMA to the chamber 230 for a predetermined time (e.g., tens of milliseconds), the main controller 500 sends a flow control start instruction indicating the start of flow control to the flow control device 1. In addition, when the film formation of one layer of a wafer is completed, the main controller 500 sends a flow control end instruction indicating the end of flow control to the flow control device 1 to end the process.
[0068] When the CPU 21 receives the flow control start command sent from the main controller 500 using the communication IF 24, it determines that the flow control mode is set (S20: flow control mode) and executes the flow control process (S50). When the flow control process is completed, the CPU 21 returns to the process of S10. The flow control process will be described later.
[0069] The learning mode is a mode for performing learning processing. For example, TMA used in ALD is used to vaporize solid materials, but when the solid materials decrease, the evaporation amount decreases. In other words, the use conditions of the gas change with the residual amount of solid materials. Therefore, the main controller 500 changes the target flow rate according to the use conditions of the gas. In such a case, it is desired to change the pulse injection method responsively to follow the change of the target flow rate.
[0070] Here, the main controller 500 sends a learning instruction for executing the learning process to the flow control device 1 that has sent the controllable notification. The central processing unit 21, using the communication IF 24, determines that the learning mode is set (S20: learning mode) when receiving the learning instruction sent from the main controller 500, and executes the learning process (S40). When the learning process is completed, the central processing unit 21 returns to the process of S10. The learning process will be described later.
[0071] In addition, the main controller 500 may send a learning instruction only once after receiving the controllable notification, or send a learning instruction each time before controlling the gas flow. For another example, the main controller 500 may send a learning instruction each time the process is performed a predetermined number of times (e.g., 20 times).
[0072] The central processing unit 21 terminates the control program 31 when the power supply of the flow regulating device 1 is turned off. Before the control program 31 is terminated, it sends an uncontrollable notification indicating that the control of the flow regulating device 1 is impossible to the main controller 500. The uncontrollable notification includes the identification information of the flow regulating device 1. The main controller 500 that has received the uncontrollable notification does not send a command to the flow regulating device 1 corresponding to the identification information of the uncontrollable notification, and does not perform useless communication.
[0073] In addition, the flow control device 1 may turn on or off the power in response to a command from the main controller 500 or in response to turning on or off the power of the main controller 500. In this case, the CPU 21 may omit sending the controllable notification or the uncontrollable notification.
[0074] Gas replacement treatment
[0075] Next, the above-mentioned gas replacement process will be described. Figure 3 1 is a flowchart for explaining the sequence of the gas replacement process. The CPU 21 executing the gas replacement process first determines the control mode (S301). The main controller 500 attaches the control mode for flow control to the gas replacement instruction. In addition, the target pressure of the tank 12 for flow control is attached to the gas replacement instruction.
[0076] Figure 4A , Figure 4B , Figure 4C is an example of a control mode. In the control mode, for example, there are first to third control modes. Figure 4A The first control mode shown is a mode in which the flow rate of the gas reaches the target flow rate Q from the initial (first) pulse injection of the process. Figure 4B The second control mode shown is a mode in which the flow rate of gas is gradually increased until the N1-th pulse injection to reach the target flow rate Q. Figure 4C The third control mode shown is a mode in which the gas flow rate is gradually reduced until the N2th pulse injection to reach the target flow rate Q. N1 and N2 are natural numbers greater than 2. N1 and N2 may be fixed values fixed in advance, variable values that can be set arbitrarily, or variable values that can be changed arbitrarily.
[0077] Back to Figure 3 , when the CPU 21 determines that the control mode is the first control mode (S301: first control mode), the first stop valve 11 and the second stop valve 13, which are in the closed state, are opened at the same time (S303). The main controller 500 drives the vacuum pump 240 when sending a gas replacement command to the flow control device 1. Therefore, a pressure difference is generated between the gas source 211 side and the chamber 230 side, and the gas remaining in the piping 200, the groove 12, and the chamber 230 flows to the vacuum pump 240 and is ejected. As a result, for example, the gas mixed with the air remaining in the piping 200, the groove 12, or the chamber 230 is replaced by the gas supplied from the gas source 211. At this time, because the flow coefficient (Cv) of the first stop valve 11 and the second stop valve 13 is large, the gas replacement time is short and ends.
[0078] After that, the CPU 21 maintains the opening of the second stop valve 13 and closes the first stop valve 11 (S305). As a result, the pressure of the tank 12 is gradually reduced. The CPU 21, which has closed the first stop valve 11, measures the pressure of the tank 12 using the pressure sensor 14 (S307). The CPU 21 determines whether the pressure measured using the pressure sensor 14 reaches the target pressure P101 of the first control mode attached to the gas replacement instruction (S309).
[0079] If the CPU 21 determines that the target pressure P101 has not been reached (S309: No), the CPU 21 waits for the target pressure P101 to be reached. On the other hand, if the CPU 21 determines that the target pressure P101 has been reached (S309: Yes), the CPU 21 maintains the closing action of the first stop valve 11 and closes the second stop valve 13 (S311). Then, the CPU 21 ends the processing.
[0080] In contrast, the CPU 21 performs the processing of S313 to S321 when it is determined that the control mode is the second control mode (S301: second control mode). The processing of S313 to S321 is the same as that of S303 to S311 described above, so the description thereof is omitted. In addition, the target pressure P102 is the target pressure of the second control mode attached to the gas replacement instruction, and may be the same value as the target pressure P101 of the first control mode, or may be a different value.
[0081] On the other hand, when the CPU 21 determines that the control mode is the third control mode (S301: third control mode), the CPU 21 opens the first stop valve 11 and the second stop valve 13 at the same time as in S303 (S323). After that, the CPU 21 maintains the opening of the first stop valve 11 and closes the second stop valve 13 (S325). As a result, the pressure of the tank 12 rises slowly. The CPU 21, which has closed the second stop valve 13, measures the pressure of the tank 12 using the pressure sensor 14 (S327). The CPU 21 determines whether the pressure measured using the pressure sensor 14 reaches the target pressure P103 of the third control mode attached to the gas replacement instruction (S329).
[0082] If the CPU 21 determines that the target pressure P103 has not been reached (S329: No), the CPU 21 waits for the target pressure P103 to be reached. On the other hand, if the CPU 21 determines that the target pressure P103 has been reached (S329: Yes), the CPU 21 maintains the closing action of the second stop valve 13 and closes the first stop valve 11 (S331). Then, the CPU 21 ends the processing.
[0083] Conventionally, gas replacement is performed by keeping one of the first stop valve 11 and the second stop valve 13 in a closed state at all times. However, in this method, the time required for gas replacement, i.e., the gas replacement time, becomes longer. In contrast, the flow control device 1 of the present embodiment replaces the gas in the piping 200 and the tank 12 by simultaneously opening the first stop valve 11 and the second stop valve 13 when performing gas replacement. In other words, the flow control device 1 of the present embodiment includes a first stop valve 11 connected to a gas source 211, a second stop valve 13 connected to the first stop valve 11, a tank 12 (gas filling container) located between the first stop valve 11 and the second stop valve 13, a pressure sensor 14 for measuring the pressure of the tank 12, and a central processing unit 21 (controller). The CPU 21 calculates the volume flow rate of the gas ejected from the slot 12 based on the post-filling pressure and post-ejection pressure of the slot 12 measured by the pressure sensor 14 while repeating the pulse ejection by opening and closing the first stop valve 11 and the second stop valve 13, and adjusts the volume flow rate of the gas ejected from the second stop valve 13 to a target flow rate by changing the method of repeating the pulse ejection. In such a pulse ejection type flow control device, the CPU 21 performs a gas replacement process to replace the gas in the slot 12 with the gas supplied by the gas source 211 by opening the first stop valve 11 and the second stop valve 13 at the same time. In such a flow control device 1, it is not necessary to adjust the internal pressure of the gas filling volume to a set pressure during gas replacement as in the case of flow control. Therefore, by opening the first stop valve 11 and the second stop valve 13 at the same time and allowing the gas supplied from the gas source 211 to continue to flow to the gas filling volume to perform gas replacement, the gas replacement time can be shortened compared to the case where the gas replacement is performed by a method in which either the first stop valve 11 or the second stop valve 13 is in a normally closed state. In particular, the first stop valve 11 and the second stop valve 13 of the flow control device 1 have a larger value of the fluid coefficient used in the piping than a normal valve. Therefore, the gas replacement time of the tank 12 can be shortened.
[0084] In addition, the flow control device 1 changes the original pressure of the gas source 211, and after replacing the gas, adjusts the pressure of the tank 12 to the target pressures P101 to P103 corresponding to the first to third control modes. Therefore, when the process is performed in the first to third control modes, the deviation of the pressure of the tank 12 is reduced, and the responsiveness of the flow control can be expected to be improved.
[0085] In addition, the processing of S303, S313, and S323 is an example of "replacement processing". In addition, the processing of S309 to S311, S319 to S321, and S329 to S331 is an example of "pressure adjustment processing". The target pressures P101 to P103 of the first to third control modes are not limited to being attached to the gas replacement instruction. For example, in the case where the flow control device 1 includes an operating unit, it is also possible to receive the target pressures P101 to P103 via the operating unit.
[0086] Flow Control Processing
[0087] Next, the above-mentioned flow control process will be described. Figure 5 is a flow chart illustrating the process of flow control processing. Figure 5 FIG. 2 shows a process of improving responsiveness in the first control mode.
[0088] When the CPU 21 receives the flow control start instruction from the main controller 500 and starts the flow control process, it first performs the following steps: Figure 5 As shown, the target flow rate Q is received (S501). The main controller 500, the flow control start instruction is accompanied by the target flow rate Q when the gas is supplied to the chamber 230. When the central processing unit 21 receives the flow control start instruction sent by the main controller 500 using the communication IF24, it stores the target flow rate Q attached to the flow control start instruction in the memory 22. In addition, when the flow control device 1 includes an operating unit, the central processing unit 21 may store the target flow rate Q inputted using the operating unit in the memory 22.
[0089] The CPU 21 receiving the target flow rate Q determines whether it is the first pulse injection (S503). In this specification, "the first pulse injection" is defined as the first pulse injection performed after receiving the flow rate control start command.
[0090] When the CPU 21 determines that it is the first pulse injection (S503: Yes), the optimal filling time tx stored in the memory 22 is substituted into the filling time t1 (S505). The filling time t1 is the time to fill the gas into the tank 12 in order to maintain the opening action of the first stop valve 11 in the current pulse injection. Then, the CPU 21 obtains the previous target flow rate Qold stored in the memory 22 (S507). In addition, the processing of S505 and S507 may be in the reverse order.
[0091] Then, the CPU 21 determines whether the target flow rate Q received in S501 is the same as the previous target flow rate Qold obtained in S507 (S509). When the CPU 21 determines that they are the same (S509: Yes), the CPU 21 fills the tank 12 with gas during the filling time t1 (S511). In other words, the first stop valve 11 and the second stop valve 13 are closed when the flow control process starts. Therefore, the CPU 21 causes the first stop valve 11 to open and starts filling the tank 12 with gas. The CPU 21 uses the timing unit 23 to measure the time for maintaining the opening of the first stop valve 11, that is, the first opening maintenance time. The CPU 21 maintains the opening of the first stop valve 11 until the first opening maintenance time reaches the filling time t1 obtained in S503. When the CPU 21 detects that the filling time t1 has passed using the timing unit 23, the CPU 21 causes the first stop valve 11 to close and ends the filling of gas. When the filling of the gas is completed, the CPU 21 resets the timer 23 .
[0092] The CPU 21 that has completed the gas filling determines whether the first stabilization time t2 has passed (S513). The pressure of the tank 12 immediately after the gas filling is unstable due to adiabatic compression. In addition, the response delay of the first stop valve 11 is also considered. Here, the first stabilization time t2 is set from the time when the first stop valve 11 is closed to the time when the second stop valve 13 is opened. In this method, the ejection time of the gas from the second stop valve 13 is constant with the time from the time when the second stop valve 13 is closed to the time when the first stop valve 11 is opened. In addition, the cycle from the time when the first stop valve 11 is opened to the time when the first stop valve 11 is opened next is constant. Therefore, by determining the filling time t1, the first stabilization time t2 is automatically determined. After the first stop valve 11 is closed, the CPU 21 starts measuring time using the timing unit 23. When the time measured by the timer 23 does not reach the first stable time t2, the CPU 21 determines that the first stable time t2 has not elapsed (S513: No), and waits.
[0093] On the other hand, when the time measured by the timer 23 reaches the first stable time t2, the CPU 21 determines that the first stable time t2 has passed (S513: Yes). Then, the CPU 21 acquires the post-filling pressure P1 (S515). In other words, the CPU 21 uses the pressure sensor 14 to measure the pressure of the tank 12 when the first stable time t2 has passed, and temporarily stores the measured pressure in the memory 22 as the post-filling pressure P1.
[0094] The CPU 21, which has obtained the post-filling pressure P1, ejects the gas from the second stop valve 13 during the ejection time t3 (S517). The ejection time t3 is predetermined as described above. After the CPU 21 closes the first stop valve 11, it opens the second stop valve 13 at the moment of obtaining the post-filling pressure P1. The CPU 21 uses the timing unit 23 to measure the time for maintaining the opening action of the second stop valve 13, that is, the second opening action maintenance time. The CPU 21 maintains the opening action of the second stop valve 13 until the second opening action maintenance time becomes the ejection time t3. When the ejection time t3 has passed, the CPU 21 closes the second stop valve 13 to end the ejection of the gas. At the same time when the gas ejection ends, the CPU 21 resets the timing unit 23.
[0095] The CPU 21, which has completed the gas ejection, determines whether the second stabilization time t4 has passed (S519). The pressure of the groove 12 immediately after the gas is ejected is unstable due to adiabatic expansion. In addition, the response delay of the closing action of the second stop valve 13 is also considered. Therefore, the second stabilization time t4 is set in the period from the closing action of the second stop valve 13 to the opening action of the first stop valve 11. The second stabilization time t4 is predetermined as described above. After the second stop valve 13 has been closed, the CPU 21 starts measuring time using the timing unit 23. When the time measured using the timing unit 23 does not reach the second stabilization time t4, the CPU 21 determines that the second stabilization time t4 has not passed (S519: No) and stands by.
[0096] On the other hand, when the time measured by the timing unit 23 reaches the second stable time t4, the CPU 21 determines that the second stable time t4 has passed (S519: Yes). Then, the CPU 21 acquires the post-discharge pressure P2 (S521). In other words, the CPU 21 uses the pressure sensor 14 to measure the pressure of the groove 12 when the second stable time t4 has passed, and temporarily stores the measured pressure in the memory 22 as the post-discharge pressure P2.
[0097] The CPU 21 calculates the volume flow rate QA ejected from the second stop valve 13 by the pulse injection (S523). In other words, the flow control device 1 uses the following equation 1 to calculate the volume flow rate QA of the gas ejected from the second stop valve 13 per unit time.
[0098] Mathematical formula 1
[0099] Volume flow rate = ejection volume × pulse injection times
[0100] The number of pulse injections is determined by the process time and the pulse injection cycle. The injection amount is calculated by the following mathematical formula 2. ΔP is the pressure difference. P is the atmospheric pressure (101.3 kPa). V is the volume of the groove 12 (cc). T is the fluid temperature (°C). T is the temperature of the gas supplied to the chamber 230.
[0101] Mathematical formula 2
[0102] Ejection volume = ΔP / P×V×273 / (273+T)
[0103] ΔP: Pressure difference = (P1-P2) [kPa]
[0104] P: atmospheric pressure = 101.3 [kPa]
[0105] V: Tank volume [cc]
[0106] T: Fluid temperature [℃]
[0107] In the flow rate control device 1, P and V are predetermined values. In addition, for example, when T is set to 20°C, the discharge amount is converted and obtained at 20°C. The pressure difference ΔP is the difference between the post-filling pressure P1 obtained in S515 and the post-discharge pressure P2 obtained in S521. Therefore, the discharge amount changes according to the pressure difference ΔP between the post-filling pressure P1 and the post-discharge pressure P2.
[0108] The CPU 21 substitutes the post-filling pressure P1 and the post-discharge pressure P2 temporarily stored in the memory 22 in S515 and S521 into the above-mentioned mathematical formula 2 to obtain the discharge amount. The CPU 21 substitutes the number of pulse injections in the process into the number of pulse injections in the mathematical formula 1, and multiplies the discharge amount obtained by the mathematical formula 2 to calculate the volume flow rate QA.
[0109] Then, the CPU 21 calculates the next filling time t1new by PID (Proportional Integral Derivative) operation using the volume flow rate QA calculated in S521 (S525).
[0110] For PID calculation processing, refer to Figure 6 Provide explanation. Figure 6This is a functional block diagram of the controller 15. The controller 15 includes a volume flow calculation circuit 53 that calculates the volume flow QA using the post-filling pressure P1 and the post-discharge pressure P2. The volume flow calculation circuit 53 outputs the calculated volume flow QA to the deviation calculation unit 51. The deviation calculation unit 51 inputs the target flow Q received in S501, compares it with the volume flow QA input from the volume flow calculation circuit 53, and calculates the control deviation composed of the difference between the target flow Q and the volume flow QA. The deviation calculation unit 51 outputs the calculated control deviation to the PID operation circuit 52. The PID operation circuit 52 uses the control deviation input from the deviation calculation unit 51 to calculate the next filling time t1new in the following mathematical formula 3. Kp is a proportional constant. Ki is an integral constant. Kd is a differential constant. e(t) is the deviation this time. e(t-1) is the deviation of the previous time.
[0111] Mathematical formula 3
[0112]
[0113] Kp: proportional constant
[0114] Ki: integration constant
[0115] Kd: differential constant
[0116] e(t): Deviation of this time
[0117] e(t-1): the previous deviation
[0118] The amount of gas filled into the groove 12 is adjusted by the opening and closing action of the first stop valve 11. Therefore, when the ejection time t3 and the second stabilization time t4 are constant, the volume flow rate QA of the gas ejected from the second stop valve 13 can be adjusted by changing the filling time t1. Therefore, by calculating the next filling time t1new, the method of the next pulse injection is changed from the current pulse injection method, and the volume flow rate QA can be made closer to the target flow rate Q.
[0119] Back to Figure 5The CPU 21 that calculates the next filling time t1new in S525 overwrites the previous target flow rate Qold with the target flow rate Q used in the pulse injection, and stores it in the memory 22 (S527). Then, the CPU 21 determines whether the target flow rate Q has been changed (S529). When the CPU 21 determines that the target flow rate Q has not been changed (S529: No), it determines whether the flow control end instruction has been received (S531). In other words, when the CPU 21 receives the flow control end instruction sent by the main controller 500 using the communication IF24, it determines that the flow control end instruction has been received, and when it has not been received, it determines that the flow control end instruction has not been received. In addition, when the flow control device 1 includes an operating unit, the CPU 21 may receive the flow control end instruction via the operating unit. In addition, the CPU 21 receives the registration of the process time, and after receiving the flow control start instruction, when the process time has passed, it may automatically determine that the flow control end instruction has been received.
[0120] When the CPU 21 determines that it has not received the flow control command (S531: No), it returns to the process of S503 to determine whether it is the first pulse injection. At this point in time, since it is not the first pulse injection (S503: No), the CPU 21 substitutes the next filling time t1new calculated in S525 into the filling time t1 (S537). As a result, the filling time t1 of this pulse injection is changed from the filling time t1 of the previous pulse injection, and the pulse injection method is changed. Then, the CPU 21 proceeds to S507. Since the processing after S507 is the same as above, the description is omitted.
[0121] The central processing unit 21 improves the accuracy of adjusting the volume flow rate to the target flow rate by repeatedly executing the processes from S503 to S531 and S537.
[0122] When the CPU 21 determines that the flow control end instruction has been received (S531: Yes), the filling time t1 used in the last pulse injection is stored as the optimal filling time tx in the non-volatile memory of the memory 22 (S533). In other words, the optimal filling time tx stored in the previous process is overwritten on the filling time t1 used in the last pulse injection in this process. Then, the CPU 21 ends the processing.
[0123] Thus, when the CPU 21 receives the flow control command for the next process and executes the flow control process, the first pulse injection of the next process is performed using the filling time t1 used in the last pulse injection of the previous process. Therefore, in the next pulse injection, the possibility that the flow rate of the gas ejected from the second shutoff valve 13 will be adjusted to the target flow rate Q from the first pulse injection becomes high. Therefore, the response time of the flow control device 1 from receiving the flow control start command to adjusting the flow rate of the gas to the target flow rate Q can be shortened.
[0124] in addition, Figure 2 The flow control process shown in S50, Figure 5 The illustrated process is an example of a "flow control process". Figure 5 The process of S533 is an example of the “optimal filling time storage process.” The process of S505 is an example of the “filling time control process.”
[0125] In addition, as described above, in the case of a gas used to gasify a solid material such as TMA, the evaporation amount of the solid material changes as the solid material decreases. In addition, the initial pressure of the gas source 211 changes. In other words, the use conditions of the gas change in each process and in the middle of the process. In this case, the main controller 500 changes the target flow rate Q according to the change in the use conditions of the gas in order to stabilize the molecular weight of the gas supplied to the chamber 230.
[0126] When the central processor 21 changes the target flow rate Q, there is a concern that the gas cannot be adjusted to the target flow rate after the change when the filling time t1 calculated in the pulse injection performed before the change is used. Here, the central processor 21 makes a correction to change the filling time t1 according to the change in the target flow rate Q. In this case, the central processor 21 adjusts the filling time t1 using a correction pattern stored in the memory 22 in the learning process described later, thereby shortening the correction time.
[0127] In other words, when the CPU 21 receives the flow control command, for example, if the target flow rate Q in the current process is different from the target flow rate Qold in the previous process (S509: No), the CPU 21 corrects the filling time t1 (S539).
[0128] For another example, when the CPU 21 changes the target flow rate Q to the target flow rate Qnew in the middle of this process (S529: Yes), the changed target flow rate Qnew is substituted into the target flow rate Q (S535). After that, the CPU 21 performs the above-mentioned processes of S503 to S509 and S537. In the process of S509, the CPU 21 determines that the target flow rate Q is different from the previous target flow rate Qold (S509: No), and corrects the filling time t1 (S539).
[0129] Correction reference for filling time t1 Fig. 7A , Figure 7B , Figure 7C The memory 22 stores a correction pattern generated in the learning process described later. When correcting the filling time t1, the CPU 21 reads, for example, the correction pattern generated in the learning process described later from the memory 22. Fig. 7A The correction graph G1 shown. The correction graph G1 is a graph showing the relationship between the volume flow rate and the filling time. The central processing unit 21 obtains the filling time associated with the volume flow rate corresponding to the target flow rate Q in the read correction graph G1, and changes the filling time t1 to the obtained filling time. Thus, when the target flow rate Q is less than the previous target flow rate Qold, the filling time t1 becomes shorter according to the deviation. On the other hand, when the target flow rate Q is more than the previous target flow rate Qold, the filling time t1 becomes longer according to the deviation.
[0130] By correcting the filling time t1 in this way, the number of pulse injections required to change the gas flow rate from the target flow rate Q before the change to the target flow rate Qnew after the change is reduced compared to the above. Figure 7B As shown in the case where no correction is made, Figure 7C In other words, the CPU 21 changes the volume flow rate of the gas ejected from the second shutoff valve 13 with good responsiveness according to the change of the target flow rate Q. The process in step S539 is an example of a "correction process".
[0131] Learning Processing
[0132] Next, the above-mentioned learning process will be described. Figure 8 is a flowchart illustrating the steps of the learning process. Figure 8 The illustrated process is an example of a "learning process".
[0133] The central controller 21 first substitutes the temporary filling time ty into the filling time t1 (S401). The temporary filling time ty may be, for example, the optimal filling time tx stored in the memory 22. In addition, the temporary filling time ty may be attached to the learning instruction sent by the main controller 500. Alternatively, the temporary filling time ty may be specified in the control program 31.
[0134] The central controller 21 combines the processing of S403 to S415 with the above-mentioned processing based on the filling time t1 obtained in S401. Figure 5 The processes of S511 to S523 shown in the flow control process are performed in the same manner. The description of the processes of S403 to S415 is omitted here.
[0135] exist Figure 8 The central controller 21 that calculates the volume flow QA in the process of S415 shown in the figure determines whether the volume flow QA is stable (S417). In other words, the central controller 21 determines whether the volume flow is stable based on the volume flow QA calculated in the process of S415 and the volume flow QA previously calculated in the learning process. For example, when the rate of change of the volume flow QA calculated this time relative to the volume flow QA calculated previously exceeds a predetermined threshold, the central controller 21 determines that the volume flow QA is unstable (S417: No). In this case, the central controller 21 repeats the process after S403. Thus, in the learning process, the volume flow QA corresponding to the filling time t1 is accumulated and stored in the memory 22.
[0136] For example, when the rate of change of the volume flow rate QA calculated this time relative to the previously calculated volume flow rate QA is below a predetermined threshold, the central controller 21 determines that the volume flow rate QA is stable (S417: Yes). In this case, the central controller 21 outputs a completion signal (S419). In other words, the central controller 21 outputs a completion signal indicating completion of type acquisition to the memory 22. In the memory 22, the number of types that must be acquired in the learning process is stored ("3" in this embodiment). The memory 22 to which the completion signal has been input reduces the number of types by 1. Thus, the central controller 21 can determine how many times the above process needs to be repeated. In addition, if the number of types is multiple, it is not limited to "3".
[0137] The central controller 21 stores the filling time t1 and the volume flow rate QA in the memory 22 (S421). In other words, the central controller 21 associates the filling time t1 used when acquiring the type with the volume flow rate QA calculated before sending the completion signal, for example, Fig. 7A As shown, it is stored in the memory 22 as type A1.
[0138] return Figure 8 , the central controller 21 determines whether the predetermined number, ie, three types, have been stored (S423). In other words, the central controller 21 determines whether the number of types that must be acquired in the learning process stored in the memory 22 is "0".
[0139] When the number of types to be acquired in the learning process is not "0", the central controller 21 determines that three types are not stored (S423: No). In this case, the central controller 21 receives a change in the filling time t1 (S427). The filling time t1 may be changed according to a change condition predetermined in the control program 31 or according to an instruction from the main controller 500. If the flow rate regulating device 1 includes an operation unit, the filling time t1 may be changed according to a change instruction input by the user via the operation unit.
[0140] Then, the central controller 21 returns to the process of S403. The central controller 21 performs the above-mentioned process after S403 according to the changed filling time t1. Fig. 7A As shown, type A2 different from type A1 is stored in the memory 22 .
[0141] return Figure 8 , repeat the above process, and when the number of types that must be acquired in the learning process is "0", the central controller 21 determines that three types have been stored (S423: Yes). In this case, the central controller 21 generates a correction pattern and stores it in the memory 22 (S425). In other words, the central controller 21, as Fig. 7A As shown, based on the three types A1 to A3 stored in the memory 22, a correction graph G1 showing the relationship between the filling time and the volume flow rate is created and stored in the memory 22. Then, the central processing unit 21 ends the learning process.
[0142] Thus, the flow control device 1 obtains a correction pattern G1 suitable for the characteristics of the device itself, which can be used in Figure 5 In S539 of the flow control processing shown.
[0143] As described above, the flow control device 1 of this embodiment uses the optimal filling time tx stored in the previous process to open and close the first stop valve 11 during the initial pulse injection when the process is executed. As a result, in the initial pulse injection, the gas suitable for adjusting the volume flow rate QA of the gas ejected from the second stop valve 13 to the target flow rate Q is filled into the groove 12, and the volume flow rate QA of the gas ejected from the second stop valve 13 is adjusted to the target flow rate Q, or the possibility of being adjusted to a value close to the target flow rate Q becomes high. Therefore, according to the pulse injection type flow control device 1 of this embodiment, the response time from the start of the process to the stabilization of the gas at the target flow rate is shortened, and the responsiveness can be improved.
[0144] In addition, the present invention is not limited to the above-mentioned embodiment, and various applications are possible. In the chamber 230, a film may be formed on a wafer by a method other than ALD. In addition, the flow rate adjustment device 1 is not limited to a semiconductor manufacturing device, and can also be applied to other devices.
[0145] For example, in Figure 5 The optimal filling time tx stored in S533 may not be the filling time t1 in the last pulse injection of the previous process. For example, in each pulse injection of the previous process, the filling time t1, the volume flow rate, and the pressure of the slot 12 are stored in association with each other, and the filling time t1 associated with the pressure of the slot 12 closest to the target pressure of the next process may be used as the optimal filling time tx. However, the last pulse injection is considered to be closest to the gas usage condition of the first pulse injection in the next process. Here, as in the above manner, by storing the filling time t1 measured in the last pulse injection that is likely to adjust the volume flow rate QA to the target flow rate Q as the optimal filling time tx, the processing load and memory load when storing the optimal filling time tx can be reduced.
[0146] For example, omitting Figure 8 The learning process shown and Figure 5 However, as in the above-mentioned manner, when the target flow rate Q is changed from the target flow rate Qold used in the previous process or is changed during the execution of the flow control process, the gas flow rate can be quickly adjusted to the changed target flow rate by correcting the optimal filling time according to the difference between the target flow rate and the changed target flow rate based on the correction graph G1 showing the relationship between the filling time and the volume flow rate learned in advance.
[0147] For example, instead of Figure 3In the processing of S303, S313, and S323, the first stop valve 11 and the second stop valve 13 are alternately opened and closed to replace the gas. However, as in the present embodiment, when the gas replacement is performed, by opening the first stop valve 11 and the second stop valve 13 at the same time, the gas supplied from the gas source 211 flows to the groove 12 and is easily replaced with the gas remaining in the groove 12. As a result, compared with the case where the first stop valve 11 and the second stop valve 13 are alternately opened and closed to perform the gas replacement, the gas can be replaced faster.
[0148] For example, omitting Figure 3 The processing of S305 to S311, S315 to S321, and S325 to S331 may also be performed. For example, the first stop valve 11 and the second stop valve 13 that are in the opening action are closed at the same time, and the gas replacement is ended without adjusting the pressure. However, as in the above-mentioned method, after the first stop valve 11 and the second stop valve 13 are opened at the same time and the gas in the groove 12 is replaced, the pressure of the groove 12 is adjusted to the target pressure of the groove 12 in the case of performing the flow control processing according to the time point when the first stop valve 11 and the second stop valve 13 are closed. In the process after the gas replacement, since the pressure of the groove 12 is controlled to the target pressure from the initial pulse injection, the delay in the response time due to the pressure deviation of the groove 12 can be suppressed.
[0149] For example, Figure 2 As shown, it is also possible not to perform the processing according to the setting of the mode. For example, the controller 15 may perform the gas replacement processing, the flow control processing, and the learning processing separately. However, as described above, by performing the flow control processing, the gas replacement processing, and the learning processing according to the settings of the flow control mode, the gas replacement mode, and the learning mode, each processing can be performed at any time point, and the usability is good.
[0150] In addition, the processing of each flowchart described in the above embodiment may be changed in order within a range without contradiction. In addition, the storage medium storing the control program 31 is also a novel and useful invention.
[0151] Description of Reference Numerals
[0152] 1-pulse jet flow regulating device; 11-first stop valve; 12-groove; 13-second stop valve; 14-pressure sensor; 15-controller.
Claims
1. A pulse jet flow control device, characterized in that: include: A first stop valve connected to a gas source; a second stop valve connected to the first stop valve; a gas-filled container located between the first stop valve and the second stop valve; as well as a pressure sensor to measure the pressure of the gas-filled container; The adjusting device performs a process of controlling the gas to a target flow rate multiple times; The first stop valve, the second stop valve and the pressure sensor are communicatively connected to a controller that controls the operation of the pulse jet flow rate adjustment device; The controller performs the following steps in each process: a flow control process, based on the post-filling pressure measured by the pressure sensor after the gas filling container is filled with gas by opening and closing the first stop valve while repeating the pulse injection by opening and closing the second stop valve after the first stop valve is opened and closed, and the post-injection pressure measured by the pressure sensor after the gas is ejected from the gas filling container by opening and closing the second stop valve, and adjusting the volume flow rate to a target flow rate by changing a method of repeating the pulse injection; In addition, the controller: performing an optimal filling time storage process, wherein in the flow rate control process, when the pulse injection in which the flow rate has been adjusted to the target flow rate is performed, the first shutoff valve is opened and closed, and a filling time of the gas filled into the gas filling container is stored as an optimal filling time; as well as A filling time control process is executed, and in the flow control process of the process to be performed next to the process in which the optimal filling time storage process has been performed, when an initial pulse injection is performed, the first shutoff valve is opened and closed using the optimal filling time stored in the optimal filling time storage process.
2. The pulse jet flow rate adjustment device according to claim 1, wherein: In the optimal filling time storage process, the filling time of the last pulse injection is stored as the optimal filling time.
3. The pulse jet flow rate adjustment device according to claim 1 or 2, wherein: The controller performs a replacement process to replace the gas remaining in the gas-filled container by simultaneously opening the first shutoff valve and the second shutoff valve.
4. The pulse jet flow rate adjustment device according to claim 3, wherein: The controller performs a pressure adjustment process to close the first and second stop valves that are in an opening operation so that the pressure of the gas-filled container is adjusted to a target pressure of the gas-filled container when the flow rate control process is performed after the replacement process.
5. The pulse jet flow rate adjustment device according to claim 1 or 2, wherein: The controller: performing a learning process, in which the first stop valve is opened and closed to fill the gas into the gas filling container, the filling time is variable, and pulse injection is repeated while the second stop valve is opened and closed to fill the gas into the gas filling container, and the filling time is calculated for each volume flow rate of the gas ejected from the second stop valve, so as to learn the relationship between the filling time and the volume flow rate; In the flow control process, a correction process is performed. When the target flow is changed from the target flow used in the previous process, or when the target flow is changed during the execution of the flow control process, the optimal filling time is corrected according to the difference between the target flow and the changed target flow based on the relationship between the filling time and the volume flow learned in the learning process.
6. The pulse jet flow rate adjustment device according to claim 3, characterized in that: The controller: performing a learning process, in which the second stop valve is opened and closed to discharge the gas from the gas filling container at a constant time, the first stop valve is opened and closed to discharge the gas into the gas filling container at a variable time, pulse injection is repeated, the filling time is calculated for each volume flow rate of the gas discharged from the second stop valve, and the relationship between the filling time and the volume flow rate is learned; During the flow control process, a correction process is performed to correct the optimal filling time according to a difference between the target flow rate and the changed target flow rate based on the relationship between the filling time and the volume flow rate learned in the learning process, when the target flow rate is changed from the target flow rate used in the previous process, or when the target flow rate is changed during the execution of the flow control process; In addition, the controller: executing a mode setting process to set a flow control mode for executing the flow control process, a gas replacement mode for executing the replacement process, and a learning mode for executing the learning process; The mode processing set in the mode setting processing is executed.
7. A pulse jet flow rate adjustment method, characterized in that: Used in a pulse jet flow control device, wherein the pulse jet flow control device comprises a first shut-off valve connected to a gas source, a second shut-off valve connected to the first shut-off valve, a gas-filled container located between the first shut-off valve and the second shut-off valve, and a pressure sensor for measuring the pressure of the gas-filled container, wherein the pulse jet flow control device uses a controller communicatively connected to the first shut-off valve, the second shut-off valve, and the pressure sensor to control the action, The process of controlling the gas to the target flow rate is performed multiple times; During each process: a flow control step, based on the post-filling pressure measured by the pressure sensor after the gas filling container is filled with gas by opening and closing the first stop valve while repeating the pulse injection by opening and closing the second stop valve after the first stop valve is opened and closed, and the post-injection pressure measured by the pressure sensor after the gas is ejected from the gas filling container by opening and closing the second stop valve, and adjusting the volume flow rate to a target flow rate by changing the way of repeating the pulse injection; In addition, the following activities are carried out: an optimal filling time storing step of, in the flow control step, opening and closing the first shutoff valve when the pulse injection in which the volume flow rate has been adjusted to the target flow rate is performed, and storing the filling time at which the gas filling container is filled with the gas as the optimal filling time; and A filling time control step, in the flow control step of the process to be performed next to the process in which the optimal filling time storage step has been performed, when an initial pulse injection is performed, the first shutoff valve is opened and closed using the optimal filling time stored in the optimal filling time storage step.
8. A program, characterized in that is built into a controller for controlling the operation of a pulse jet type flow control device, wherein the pulse jet type flow control device includes a first shut-off valve connected to a gas source, a second shut-off valve connected to the first shut-off valve, a gas-filled container located between the first shut-off valve and the second shut-off valve, and a pressure sensor for measuring the pressure of the gas-filled container; The flow control device, when performing a plurality of processes of controlling the gas to a target flow rate, executes in each process: a flow control process, based on the post-filling pressure measured by the pressure sensor after the gas filling container is filled with gas by opening and closing the first stop valve while repeating the pulse injection by opening and closing the second stop valve after the first stop valve is opened and closed, and the post-injection pressure measured by the pressure sensor after the gas is ejected from the gas filling container by opening and closing the second stop valve, and adjusting the volume flow rate to a target flow rate by changing a method of repeating the pulse injection; In addition, in the controller: performing an optimal filling time storage process, in which, when pulse injection in which the volume flow rate is adjusted to the target flow rate is performed in the flow control process, the first shutoff valve is opened and closed, and a filling time at which the gas is filled into the gas filling container is stored as an optimal filling time; and A filling time control process is executed. In the flow control process of the process to be performed next to the process in which the optimal filling time storage process has been performed, when an initial pulse injection is performed, the optimal filling time stored in the optimal filling time storage process is used to cause the first shutoff valve to perform a switching action.
Citation Information
Patent Citations
Flow controller and test method therefor, and flow control method
CN101551675A
Method and device for controlled dosing of a gas with fluctuating supply pressure
CN102834164A
Gas control device with composite function
CN110571456A
Apparatus and method of pulse type flow rate adjustment
TW552490B
