Valve control device and inference device

Abstract

The present invention provides a valve control device and an inference device, wherein the valve control device can improve the accuracy of pressure regulation control. In the valve control device (1b), a calculation unit (210) fixes the opening (θr) of the valve body (12) to a constant value when introducing gas into the chamber, and infers exhaust information of the vacuum valve (1) based on change information of the pressure measurement value (Pr) when the opening (θr) of the valve body (12) is fixed to a constant value. Furthermore, the pressure regulation control unit (21) controls the opening of the valve body (12) based on the exhaust information to perform pressure regulation control of the chamber.

Description

Technical Field

[0001] The present invention relates to a valve control device and an inference device. Background Art

[0002] In semiconductor processes such as dry etching, the process gas introduced into the chamber is regulated by a flow controller to meet predetermined conditions such as gas type and flow rate Qin. Chamber pressure Pr is one of the most important process conditions, and the valve body opening position is controlled to achieve a predetermined pressure value, thereby maintaining the chamber pressure Pr at the specified pressure value. An automatic pressure regulating valve (also known as an automatic pressure control (APC) valve) that uses a motor to drive the control valve body, such as the valve described in Patent Document 1, is used as this type of valve.

[0003] [Prior art literature]

[0004] [Patent Document]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2018-106718 Summary of the Invention

[0006] [Problems to be solved by the invention]

[0007] The valve described in Patent Document 1 performs pressure regulation based on pre-stored effective exhaust velocity data. However, the type of gas actually flowing in may not necessarily match the gas type indicated by the pre-stored data. This leads to a problem in which the accuracy of pressure regulation decreases.

[0008] [Technical means to solve the problem]

[0009] The first form of the valve control device of the present invention is a valve control device that controls the valve body opening based on the pressure measurement value of the chamber installed with the vacuum valve and the pressure target value of the chamber, and performs pressure regulation of the chamber, including: an opening setting unit that fixes the valve body opening to a certain value when introducing gas into the chamber; and an inference unit that infers exhaust information of the vacuum valve based on change information of the pressure measurement value when the valve body opening has been fixed to a certain value; the valve control device controls the valve body opening based on the exhaust information, and performs pressure regulation control of the chamber.

[0010] The second form of the inference device of the present invention is an inference device for inferring the exhaust information of a vacuum valve installed in a chamber. Based on the change information of the pressure measurement value of the chamber when the valve body opening of the vacuum valve is constant, at least one of the valve exhaust speed, the flow rate of the gas introduced into the chamber, and the gas type information of the gas is inferred as the exhaust information.

[0011] [Effects of the Invention]

[0012] According to the present invention, it is possible to improve the accuracy of voltage regulation control. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a block diagram showing a schematic configuration of a vacuum valve installed in a vacuum processing apparatus.

[0014] Figure 2 This is a functional block diagram of the vacuum valve 1.

[0015] Figure 3 This diagram explains the voltage regulation logic in opening control.

[0016] Figure 4 Graph showing transition of the state (θr, Pr) due to the opening control.

[0017] Figure 5 It is a diagram showing an opening degree change graph and a pressure change graph.

[0018] Figure 6 This is a graph showing the relationship between gas type and effective exhaust velocity.

[0019] Figure 7 This is a graph showing pressure changes when different gas types are used.

[0020] Figure 8 This is a diagram showing a trajectory when the stored effective exhaust velocity value deviates from the effective exhaust velocity of the gas actually flowing.

[0021] Figure 9 This is a diagram showing an example of calculating the effective exhaust speed ratio when reliability is sufficiently ensured.

[0022] Figure 10 This is a diagram showing an example of calculating the effective exhaust speed ratio when reliability is insufficient.

[0023] Figure 11 This is a diagram showing an example of application of the effective exhaust speed ratio to pressure regulation control, and illustrates a case where the effective exhaust speed of the actually flowing gas is smaller than the stored effective exhaust speed.

[0024] Figure 12 This is a diagram showing an example of application of the effective exhaust speed ratio to pressure regulation control, and illustrates a case where the effective exhaust speed of the gas actually flowing is greater than the stored effective exhaust speed.

[0025] Figure 13 This is a flowchart showing a series of processes in the calibration mode.

[0026] Figure 14 It means immediately after Figure 13 Flowchart of the processing after the processing.

[0027] Figure 15 3 is a graph showing the measured opening value θr (line L31 ) and the measured pressure value Pr (line L32 ) during the calibration process.

[0028] [Explanation of Symbols]

[0029] 1: Vacuum valve

[0030] 1a: Valve body

[0031] 1b: Valve control device

[0032] 3: Vacuum chamber

[0033] 12: Valve body

[0034] 21: Voltage regulation control unit

[0035] 210: Operation unit

[0036] Pr: pressure measurement value

[0037] Ps: pressure target value

[0038] Pp: predicted pressure DETAILED DESCRIPTION

[0039] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Figure 1 This is a block diagram showing the schematic structure of a vacuum valve 1 installed in a vacuum processing apparatus. The vacuum valve 1 is a pressure regulating valve and includes a valve body 1a equipped with a valve element 12, and a valve control device 1b that controls the opening and closing of the valve element 12. The valve body 1a is installed in the vacuum chamber 3 of the vacuum processing apparatus, and a vacuum pump 4 is installed on the exhaust side of the valve body 1a. Gases such as process gases are introduced into the vacuum chamber 3 via a flow controller 32. The flow controller 32 is a device that controls the flow rate Qin of the gas introduced into the vacuum chamber 3 and is controlled by the main controller MC of the vacuum processing apparatus. The pressure within the vacuum chamber 3 (chamber pressure) is measured by a vacuum gauge 31, and the pressure measurement value Pr is input into the valve control device 1b.

[0040] The valve body 1a is equipped with a motor 13 that drives the valve body 12 to open and close. The valve body 12 is driven to open and close by the motor 13. The motor 13 is equipped with an encoder 130 for detecting the opening and closing angle of the valve body 12. The detection signal from the encoder 130 is input into the valve control device 1b as the opening signal θr of the valve body 12 (hereinafter referred to as the opening measurement value θr).

[0041] The valve control device 1b that controls the valve body 1a includes a pressure regulation control unit 21 and a motor drive unit 22. In addition to the pressure measurement value Pr and the opening measurement value θr, the valve control unit 1b receives the target pressure value Ps for the vacuum chamber 3 from the main controller MC of the vacuum processing apparatus. The motor drive unit 22 includes an inverter circuit for driving the motor and a motor control unit that controls the inverter circuit. The measured opening value θr from the encoder 130 is received as input. The pressure regulation control unit 21 receives the chamber pressure Pr measured by the vacuum gauge 31 and the target pressure value Ps for the vacuum chamber 3 from the main controller MC of the vacuum processing apparatus. Furthermore, the pressure regulation control unit 21 transmits a determination result D, described below, to the main controller MC.

[0042] The valve control device 1b includes, for example, a microcomputer or other processing unit having a central processing unit (CPU), memory (ROM, RAM), and peripheral circuits. The functions of the voltage regulator control unit 21 and the motor control unit of the motor driver 22 are implemented using software programs stored in the ROM. Alternatively, a digital arithmetic unit such as a field programmable gate array (FPGA) and its peripheral circuits may be included in place of the microcomputer.

[0043] Figure 2 This is a functional block diagram of the vacuum valve 1. The pressure regulation control unit 21 includes a calculation unit 210, a feedforward controller 220, a feedback controller 230, and a storage unit 240. Storage unit 240 stores parameters required for pressure regulation control (e.g., data related to the effective exhaust velocity Se, described later). The opening measurement value θr detected by the encoder 130 is input to the motor drive unit 22 and the calculation unit 210.

[0044] In this embodiment, similar to the invention of Patent Document 1, coarse adjustment using open control is performed until the pressure reaches a near target value. Then, fine adjustment is performed after switching to closed control to approach the target pressure value. In the pressure regulation control unit 21, the calculation unit 210 and the feedforward controller 220 correspond to the open control unit, while the subtractor 229 that generates the deviation ε (=Pr - Ps) and the feedback controller 230 correspond to the closed control unit.

[0045] The calculation unit 210 receives inputs such as the pressure measurement value Pr, the target pressure value Ps, and the measured opening value θr. Based on the pressure measurement value Pr, the target pressure value Ps, and the measured opening value θr, the calculation unit 210 calculates the target opening estimated value θse, the predicted pressure Pp, and exhaust information about the valve body 1a (details will be described later). Generally speaking, even if the valve body opening is fixed at a certain value, it takes a certain amount of time for the chamber pressure to reach the pressure equilibrium value corresponding to the valve body opening. Furthermore, the predicted pressure Pp is the estimated pressure value after Δt seconds have passed since the time the pressure measurement value Pr was measured (for example, a time longer than the control cycle, such as t = 0.4 seconds). The method for calculating the predicted pressure Pp will be described later.

[0046] Feedforward controller 220 outputs opening setting θ1 based on target opening estimated value θse. Feedback controller 230 also outputs opening setting θ2 based on deviation ε = Pr - Ps. Adder 225 adds the output opening setting θ1 and opening setting θ2, and the result of the addition is input to motor driver 22 as opening setting θset. Motor driver 22 drives motor 13 based on opening setting θset and measured opening value θr input from encoder 130.

[0047] If the measured opening value θr eventually becomes the target opening estimated value θse through the opening control, then Figure 2 The opening setting θ1 output by the feedforward controller 220 is fixed to the above value, and the control is switched from open control to closed control. If the control is switched to closed control, the pressure target value Ps is input. Figure 2 In subtractor 229, feedback controller 230 outputs the opening setting θ2 based on the deviation ε = Pr - Ps. Typically, feedback controller 230 includes a proportional gain and an integral gain (so-called proportional integral (PI) gain). The motor drive unit 22 controls the opening based on the opening setting θset = θ1 (fixed) + θ2. Unless the valve body 12 is operating at high speed, the measured opening value θr is θr = θ1 + θ2.

[0048] Before the closing control starts, the pressure measurement value Pr is input to the subtractor 229 instead of the pressure target value Ps. Therefore, during the opening control, the pressure deviation ε = 0 is input to the feedback controller 230, and the opening setting θ2 = 0 is output from the feedback controller 230.

[0049] (How to calculate the predicted pressure Pp)

[0050] As an example of a method for calculating the predicted pressure Pp, there is a method described in paragraph 0024 of Japanese Patent Application Laid-Open No. 2018-106718. Here, only the key points are described. Regarding the pressure measurement value Pr of the vacuum chamber 3, the exhaust equation represented by the following equation (1) holds true.

[0051] V×(dPr / dt)+Se×Pr=Qin…(1)

[0052] In equation (1), V is the chamber volume including the vacuum chamber 3, Se is the effective exhaust speed of the exhaust system including the conductance of the vacuum valve 1, and Qin is the flow rate of the gas introduced into the vacuum chamber 3. Information on the effective exhaust speed Se is given as the correlation Se(θ) between the opening θ of the vacuum valve 1 and the effective exhaust speed Se.

[0053] In pressure control, the effective exhaust speed Se estimated based on equation (1) can be used. Furthermore, the exhaust equation (1) can be used to calculate the predicted pressure. The general solution of equation (1) is expressed as equation (2).

[0054] [Mathematical formula 1]

[0055]

[0056] As a method for calculating the predicted pressure Pp t seconds after the present time using equation (2), for example, the following discretized equations (3) and (4) are used. Using equations (3) and (4), the predicted pressure Pp t seconds after the present time is calculated by calculating the recursive formula for each Δt period from the present time to t seconds after the present time. For example, if k is set to 1 to 99 and t seconds after the present time is set to 0.4 seconds, Δt = 4 msec.

[0057] P(after Δt) = Cp(now) × P(now)

[0058] + Cq(now)×{Qin(now)+A×Δt}…(3)

[0059] P((k+1)×Δt)=Cp(k)×P(k×Δt)

[0060] + Cq(k)×{Qin(now)+A×k×Δt}…(4)

[0061] in,

[0062] Cp(k)=exp{(-Se(k×Δt) / V)×Δt}

[0063] Cq(k)=(1 / V)×{1 / (-Se(k×Δt) / V)}×(Cp(k)-1)

[0064] To calculate the predicted pressure Pp t seconds from now using Equations (3) and (4), the estimated flow rate from now until t seconds from now and the effective exhaust speed Se from now until t seconds from now are required. For example, in Equation (4), {Qin(now) + A × k × Δt} represents the estimated flow rate k × Δt seconds from now. Here, it is assumed that the flow rate changes as A × k × Δt, and A is a constant.

[0065] The judgment of using predicted pressure Pp is based on Figure 3 The voltage regulation logic shown in is used. Figure 3 The coordinate system shown in FIG is the θr-Pr coordinate system, with the state (θr, Pr) as its coordinate points, and the state (θse, Ps) as its origin. θse and Ps represent the estimated target opening value and the target pressure value. In this θr-Pr coordinate system, the opening θ of the control valve body 12 is determined in either the opening or closing direction based on whether the state (θr, Pr) is in the first through fourth quadrants and the magnitude relationship between the predicted pressure Pp and the target pressure value Ps.

[0066] When the state (θr, Pr) before the opening change is in the second or fourth quadrant, the opening θ is adjusted toward the target estimated opening value θse. As a result, the measured pressure value Pr changes toward the target pressure value Ps. That is, the pressure decreases in the second quadrant, while the pressure increases in the fourth quadrant. If the starting point for control is either the second or fourth quadrant, the fourth quadrant is selected in the case of an increasing pressure rising from the starting pressure Pstt, and the second quadrant is selected in the case of a decreasing pressure falling from the starting pressure Pstt.

[0067] On the other hand, when the state (θr, Pr) before the opening change is in the first or third quadrant, the direction of the opening adjustment is set according to the magnitude relationship between the predicted pressure Pp and the target pressure Ps. In the first quadrant, if the predicted pressure Pp is greater than the target pressure Ps (e.g., Pp>Ps), the opening is adjusted in the direction of increasing the opening to reduce the chamber pressure (indicated by the rightward arrow), or the opening value is maintained unchanged, as indicated by circle 50. Conversely, if the relationship is Pp<Ps), the opening is adjusted in the direction of decreasing the opening to increase the chamber pressure (indicated by the leftward arrow). In the third quadrant, if Pp>Ps, the opening is adjusted in the direction of increasing the opening to reduce the chamber pressure (indicated by the rightward arrow). Conversely, if the relationship is Pp<Ps, the opening is adjusted in the direction of decreasing the opening to increase the chamber pressure (indicated by the leftward arrow), or the opening value is maintained unchanged, as indicated by circle 50.

[0068] (Control example of opening control)

[0069] Reference Figure 4 、 Figure 5 The change in the state (θr, Pr) during the open control is described. Figure 4 This diagram shows an example of the transition of the state (θr, Pr) under open control when the state (θstt, Pstt) at the starting point of control is in the fourth quadrant, as at point B1. Furthermore, the state at point B2 is represented as (θse0, Pra), and the state at point B3 is represented as (θse0, Prb). Figure 5 The upper graph is a pressure transition graph showing the relationship between the pressure measurement value Pr (solid line) and the predicted pressure Pp (dashed line), where the vertical axis represents pressure P and the horizontal axis represents time t. Figure 5 The graph below is an opening transition graph showing a temporal change in the opening measurement value θr, where the vertical axis represents the opening θ and the horizontal axis represents time t.

[0070] like Figure 5 As shown in the opening change graph, if the opening is changed from θstt at point B1 to θse0 at point B2 by opening control, the pressure measurement value starts to rise from Pstt at point B1. After the opening is changed, the opening is temporarily fixed at θ = θse0. Figure 5 The pressure change graph shows the pressure change after t = t0 when the opening θ is changed to θse0 (time t0) and then continues in the state of θ = θse0. When the opening is changed from θstt to θse0, the pressure measurement value Pr starts to rise from Pstt as shown by line L1A.

[0071] After a sufficient amount of time has passed, the pressure measurement value Pr converges to the pressure equilibrium value (pressure value in equilibrium) Pe(θse0) at the opening θse0. Furthermore, the opening θse0 is set smaller than the target opening θs corresponding to the pressure target value Ps, and the pressure equilibrium value Pe(θse0) to which line L1A converges is higher than the pressure target value Ps.

[0072] in this way, Figure 5 Line L1A in the pressure change graph shows the change in the measured pressure value Pr after the opening is fixed at θse0 at t = t0. Line L1B shows the predicted pressure Pp relative to line L1A. The predicted pressure Pp is the pressure value predicted Δt seconds (e.g., 0.4 seconds) after the measured pressure value Pr is measured. Assuming the measured pressure value Pr Δt seconds later is correctly predicted, line L1B is the line offset by -Δt from line L1A. If line L1A converges to the pressure equilibrium value Pe(θse0), line L1B, which shows the predicted pressure Pp, also converges to Pe(θse0).

[0073] Figure 4Point B2 (θse0, Pra) represents the time point when the opening is fixed at θse0 and rises to P = Pra. Figure 5 The predicted pressure Pp line L1B in the pressure change diagram of t = ta is Ppa Ps. Figure 3 The logic of the valve body 12 is that the opening degree is maintained at the position of point B2. Figure 3 In the third quadrant, in the case of Pp Ps, the opening is adjusted in the direction of reducing the opening (the direction indicated by the left arrow), or the opening value is maintained unchanged. However, since it is set to θse0<θs, the opening θse0 is treated as the lower limit value of the opening control, and the logic for Pp Ps is "maintaining the opening value unchanged".

[0074] If θ=θse0 is maintained and the pressure measurement value Pr further increases, the trajectory representing the state (θr, Pr) moves upward from point B2 and reaches point B3 at t=tb where Pr=Prb. As line L1A rises, line L1B of the predicted pressure Pp also rises. If the pressure measurement value Pr reaches the pressure measurement value Prb at point B3, the predicted pressure Pp becomes Pp>Ps. As a result, according to Figure 3 The logic of the opening θ is to change in the direction of increasing the value, and the trajectory of the state (θr, Pr) moves from point B3 to point B4 near the target opening estimated value θse.

[0075] As described above, the predicted pressure Pp is calculated based on the exhaust characteristic data stored in the storage unit 240. The exhaust characteristic data is valve conductance data related to a standard gas such as argon gas stored in advance in the storage unit 240, or valve conductance data related to a standard gas such as argon gas stored in advance in the storage unit 240. Figure 1 In this way, the initial calibration data related to the effective exhaust speed obtained by flowing the calibration gas in the state where the vacuum valve 1 is installed in the user's vacuum device. However, the gas type introduced into the vacuum chamber 3 is not necessarily the same as the gas type of the exhaust characteristic data stored in the storage unit 240. If the gas type is different, the calculation result of the predicted pressure Pp based on formula (2) will be different. Figure 3 The opening control of the logic will have adverse effects.

[0076] Furthermore, pressure regulation control is performed based on the exhaust equation (Equation (1), which includes the effective exhaust velocity Se as one of the exhaust characteristic data. Therefore, if the gas type for the effective exhaust velocity Se stored in the storage unit 240 differs from the type of gas actually flowing, the accuracy of pressure regulation control will decrease.

[0077] (Relationship between gas type and predicted pressure Pp)

[0078] As described above, the effective exhaust speed of the exhaust system including the conductance of the vacuum valve 1 is given as the correlation Se(θ) between the opening θ of the vacuum valve 1 and the effective exhaust speed Se. Figure 6 This graph shows the relationship between gas type and effective pumping speed. The vertical axis represents the effective pumping speed Se, and the horizontal axis represents the opening θ. Line L11 represents the effective pumping speed Se(θ) for argon gas, line L12 represents the effective pumping speed Se(θ) for helium gas, and line L13 represents the effective pumping speed Se(θ) for xenon gas. While the effective pumping speed Se(θ) varies with the opening θ, the relationship between the effective pumping speed Se(θ) by gas type is always L12 > L11 > L13, regardless of the opening θ.

[0079] Generally speaking, for each gas type, the effective exhaust speed Se(θ) of the exhaust system is largely determined by the exhaust speed Sp of the vacuum pump 4, the valve conductance C(θ) of the vacuum valve 1, and the structure of the vacuum chamber 3. Except in the region where the opening θ is large, the valve conductance C of the vacuum valve 1 dominates the effective exhaust speed Se(θ). In the low opening range (θ = 0% to approximately 20%) during pressure regulation, the effective exhaust speed Se is largely determined by the valve conductance C(θ). Furthermore, the effective exhaust speed Se depends not only on the opening θ but also on the gas flow rate Qin. However, the influence of the gas flow rate Qin is smaller than that of the opening θ. Therefore, in the description of this embodiment, the effective exhaust speed Se is treated solely as a function of the opening θ.

[0080] Figure 7 The pressure changes when the gas flow rate Qin is constant and the gas type is different. Line L1A is Figure 5 The effective exhaust speed in this case is Se1, which is the same as line L1A shown in FIG. The effective exhaust speed Se2 of line L2A satisfies Se2 > Se1, and the effective exhaust speed Se3 of line L3A satisfies Se1 > Se3. The pressure equilibrium value Pe2 (θse0) of line L2A for the effective exhaust speed Se2 (> Se1) is lower than the pressure equilibrium value Pe (θse0) of line L1A. Conversely, the pressure equilibrium value Pe3 (θse0) of line L3A for the effective exhaust speed Se3 (< Se1) is higher than the pressure equilibrium value Pe (θse0). Furthermore, the dashed line L1B represents the predicted pressure Pp of line L1A.

[0081] Here, consider a case where the effective exhaust velocity of the gas actually flowing is Se1, and the effective exhaust velocity stored in storage unit 240 is Se2. The pressure measurement value at t = t10 is Pr0. Since the effective exhaust velocity of the gas actually flowing is Se1, the predicted pressure in this case should be predicted based on line L1B, which represents the predicted pressure of line L1A.

[0082] However, the effective exhaust speed stored in storage unit 240 is Se2. The predicted pressure estimated based on Se2 is below line LIB, which represents the original predicted pressure based on Se1, and becomes Pr4 at t=t10. Specifically, if the effective exhaust speed Se2 stored in storage unit 240 has an error relative to the actual effective exhaust speed Se1 of the flowing gas, such that Se2 > Se1, the predicted pressure line L1A corresponding to the pressure measurement value Pr will be represented by line L1C, which is below line L1B. Conversely, if the effective exhaust speed Se3 (< Se1) is stored in storage unit 240, the predicted pressure line L1A corresponding to the pressure measurement value will be represented by line L1D, which is above line L1B.

[0083] exist Figure 7 In the example, the timing when the pressure measurement value Pr represented by line L1A becomes Pr>Ps is compared with the timing when the lines L1B, L1C, and L1D exceed the target pressure value Ps. The timing when the predicted pressure represented by line L1B exceeds the target pressure value Ps is Δt seconds earlier than the timing when Pr>Ps. On the other hand, if the timing when the lines L1C and L1D exceed the target pressure value Ps is Δt1 seconds earlier and Δt2 seconds earlier, respectively, then according to Figure 7 It can be seen that Δt1 < Δt < Δt2. When using line L1C, the predicted time becomes shorter than Δt, and the predicted timing becomes later than when using line L1B. Conversely, when using line L1D, the predicted time becomes longer than Δt, and the timing of Pr > Ps can be predicted earlier than when using line L1B.

[0084] In this way, when the value of the effective exhaust speed stored in the storage unit 240 deviates from the effective exhaust speed of the gas actually flowing, Figure 4 The trajectory of the state (θr, Pr) shown in the figure changes to Figure 8 The trajectory shown. The trajectory represented by the dotted line indicates that when the effective exhaust speed Se2 (>Se1) is stored in the storage unit 240, the timing at which the pressure Pp exceeds the target pressure value Ps, that is, the timing at which the opening θ is reversed in the increasing direction, is later than in the case of Se1. Therefore, the possibility of overshooting to a pressure that is excessively large compared to the pressure target value Ps becomes higher. Conversely, when the effective exhaust speed Se3 (<Se1) is stored in the storage unit 240, as shown by the dotted line, the timing at which the opening θ is reversed in the decreasing direction becomes earlier than in the case of Se1. Therefore, the closing control is started from a state significantly away from the coordinate origin (θse, Ps), and the pressure regulation time becomes longer.

[0085] (Estimation of the effective exhaust speed during pressure regulation)

[0086] As described above, if there is an error between the effective exhaust velocity pre-stored in storage unit 240 and the actual effective exhaust velocity of the gas currently flowing, problems such as overshoot or prolonged pressure regulation can arise. To address these issues, this embodiment estimates the effective exhaust velocity during pressure regulation to determine the effective exhaust velocity corresponding to the actual gas currently flowing. The thus-obtained effective exhaust velocity is used for pressure prediction calculations and pressure regulation control, thereby improving pressure regulation response.

[0087] Semiconductor processes using a pressure-regulating vacuum valve 1 include numerous pressure-regulating operations in which conditions such as the type of gas introduced into the chamber, the gas flow rate Qin, and the target pressure value Ps are changed at predetermined intervals. In each pressure-regulating operation, the gas flow rate Qin is initially controlled by a flow controller 32 to a predetermined value. Simultaneously, the valve opening of the vacuum valve 1 is adjusted to control the effective pumping speed Se, thereby converging the measured pressure value (chamber pressure) Pr toward the target pressure value Ps.

[0088] Generally speaking, flow control by flow controller 32 completes convergence earlier than pressure control by vacuum valve 1. Furthermore, during pressure control, the flow rate value Qin also converges to a roughly constant value, Qin0, at the time the effective exhaust velocity is estimated. The gas type can also be considered to have been replaced with a changed gas type.

[0089] Here, if the opening θ is fixed during a period near the pressure control convergence completion timing, the fixed opening θ is set. Figure 5 During the period (t t0) when the opening θse0 is constant, the gas type, flow rate, and opening are constant, so the effective exhaust speed Se can be considered to be a constant value Se0. In this case, at any time point t1 and time point t2 during the period when the opening θ is constant, the following equations (5) and (6) hold true. Pr1 and Pr2 are the pressure measurement values ​​at t = t1 and t2, respectively, and dPr1 / dt and dPr2 / dt are the pressure change rates over time of the pressure measurement values ​​at t = t1 and t2. V is the chamber volume, which is determined through initial calibration and stored in the storage unit 240.

[0090] Qin0=V×dPr1 / dt+Se0×Pr1…(5)

[0091] Qin0=V×dPr2 / dt+Se0×Pr2…(6)

[0092] According to equations (5) and (6), the effective exhaust speed Se0 can be expressed by the following equation (7).

[0093] Se0=-V×(dPr2 / dt-dPr1 / dt) / (Pr2-Pr1)…(7)

[0094] By using formula (7), the effective exhaust speed Se0 can be inferred based on the pressure measurement value Pr1, pressure measurement value Pr2 and pressure change rate dPr1 / dt, pressure change rate dPr2 / dt measured at time t1 and time t2, and the chamber volume V stored in the storage unit 240.

[0095] (Effective exhaust velocity ratio a_gs)

[0096] Typically, the pressure change rates dPr1 / dt and dPr2 / dt cannot be measured directly, so the difference between the pressure measurements is used instead. The numerator on the right side of equation (7) = (dPr2 / dt - dPr1 / dt) represents the difference in the pressure change rates. Therefore, it is susceptible to pressure measurement noise, and if left unchanged, reliability is low. Therefore, in this embodiment, the following processing is implemented.

[0097] The storage unit 240 stores data related to the effective exhaust speed and other factors. For example, the correlation Se(θ) between the effective exhaust speed and the valve opening degree obtained through initial calibration, or the valve conductance C(θ) stored in the storage unit 240 before shipment, is stored. As described above, the effective exhaust speed is considered to be approximately equal to the valve conductance within the opening range where pressure regulation control is performed. Therefore, the data stored in the storage unit 240 (Se(θ), C(θ)) is represented by Ser(θ), which is called the reference exhaust speed. This reference exhaust speed Ser(θ) is the effective exhaust speed associated with a known gas type.

[0098] The effective exhaust speed Se0 calculated by equation (7) is the effective exhaust speed at a fixed opening θse0. The effective exhaust speed Se0, where the gas type is unknown, can be expressed as shown in equation (8) below using the reference exhaust speed Ser(θ) where the gas type is known. In equation (8), the coefficient a_gs is referred to as the effective exhaust speed ratio. Using equation (7), the effective exhaust speed ratio a_gs is expressed as shown in equation (9) below.

[0099] Se0=a_gs×Ser(θ)…(8)

[0100] a_gs=-(V / Ser(θ))×(dPr2 / dt-dPr1 / dt) / (Pr2-Pr1)…(9)

[0101] (Predicted pressure Pp)

[0102] The predicted pressure Pp after the opening θ is fixed at θse0 is calculated based on the effective exhaust speed S(θse0) = Se0. As described above, the flow rate Qin can also be considered to have converged to the specified flow rate value Qin0, so the constant A in equations (3) and (4) becomes A = 0. In addition, the effective exhaust speed Se (k × Δt) included in Cp(k) and Cq(k) in equation (4) depends on the planned value of the opening, but Figure 5 Since the opening θ at t = t0 is fixed, the effective exhaust speed Se0 at the opening θse0 can be used as Se(k × Δt). Alternatively, the effective exhaust speed ratio a_gs and the known reference exhaust speed Ser(θ) can be used, adopting Se0 = a_gs × Ser(θ). Furthermore, according to equation (5), Qin(now) = Qin0 = V × dPr1 / dt + Se0 × Pr1. By applying these equations to equations (3) and (4), the predicted pressure Pp after Δt seconds can be calculated.

[0103] (Reliability Determination of Effective Exhaust Speed ​​Se0 and Effective Exhaust Speed ​​Ratio a_gs)

[0104] As described above, within the range of openings θ = 0% to approximately 20% for pressure regulation, the effective pumping rate is largely determined by the valve conductance. In particular, at low openings (θ < approximately 5%), the valve conductance is inversely proportional to the gas molecular weight M, √M. For example, assuming the effective pumping rate when discharging argon (Ar) gas with a molecular weight of M = 40 is the reference pumping rate Ser(θ), the effective pumping rate ratio a_gs associated with hydrogen (H2) gas with a molecular weight of M = 2 is approximately 4.5, and the effective pumping rate ratio a_gs associated with xenon (Xe) gas with a molecular weight of M = 131 is approximately 0.55. Therefore, the effective pumping rate ratio a_gs associated with gas types with molecular weights ranging from M = 2 to M = 131 is approximately between 0.55 and 4.5. In addition, in the high opening range, the exhaust speed of the vacuum pump is dominant. When a turbomolecular pump is used, the effective exhaust speed ratio a_gs in the high opening range is within the range of a_gs=0.55 to 4.5.

[0105] Therefore, if the effective exhaust speed ratio a_gs calculated by equation (9) is within the range of 0.55 to 4.5, it can be determined that the calculated effective exhaust speed ratio a_gs is highly reliable. If the calculated effective exhaust speed ratio a_gs is within the range of 0.55 to 4.5, the effective exhaust speed ratio a_gs is adopted. If it is outside the range, the effective exhaust speed ratio a_gs is not adopted, thereby ensuring the reliability of the effective exhaust speed ratio a_gs.

[0106] Furthermore, the effective exhaust speed Se0 calculated by equation (7) is the effective exhaust speed when the opening is constant. Therefore, it is ideal that the effective exhaust speed ratio a_gs calculated by equation (9) is also constant. Therefore, it is more preferable to determine that the calculated effective exhaust speed ratio a_gs is reliable and adopt it if the calculated effective exhaust speed ratio a_gs remains within the allowable range (0.55 to 4.5) for a predetermined time, or remains within each of a plurality of divided intervals of the allowable range for a predetermined time.

[0107] Furthermore, the (dPr2 / dt - dPr1 / dt) portion of the effective exhaust velocity ratio a_gs calculated using equation (9) is susceptible to noise in the pressure measurement value Pr. Therefore, to reduce the influence of this noise, performing a moving average of the pressure change rate (the difference in pressure measurement values) and, further, performing a moving average of the effective exhaust velocity ratio a_gs itself before performing dwell determination can further improve reliability. However, under conditions of extremely fast pressure response, sufficient dwell time cannot be guaranteed. Therefore, a trade-off between adopting a valid ratio and ensuring reliability is necessary, and this consideration should be taken into account when appropriately determining the determination threshold.

[0108] Figure 9 、 Figure 10 Is to say Figure 5 The diagram shown is a calculation example of the effective exhaust speed ratio a_gs in a typical response relationship of an example in which the pressure rises from the starting pressure Pstt. Figure 9 This indicates that the reliability of the effective exhaust speed ratio a_gs is sufficiently ensured. Figure 10 Indicates a situation where reliability is insufficient. Figure 9 In the figure, the calculated effective exhaust velocity ratio a_gs stays within the allowable range (0.55 to 4.5) for a sufficient time, and it is estimated that a_gs=3.

[0109] On the other hand, Figure 10 In this case, the calculated effective exhaust speed ratio a_gs fluctuates greatly, entering or leaving the allowable range (0.55 to 4.5), and reliability cannot be ensured. In this case, pressure regulation control is performed based on the reference exhaust speed Ser(θ) associated with the reference gas type stored in the storage unit 240, that is, the effective exhaust speed ratio a_gs is set to a_gs = 1.

[0110] In addition, the estimation calculation of the effective exhaust speed Se0 requires a fixed opening condition, so Figure 9 、 Figure 10 After the time t0, the a_gs signal is output. Figure 10Examples where the estimated value cannot be determined, such as the example of , are likely to occur when the pressure response is extremely fast or the signal noise (SN) of the pressure signal (pressure measurement value Pr) is poor.

[0111] (Application example of a_gs)

[0112] Figure 11 、 Figure 12 1 is a diagram showing an example of applying the estimated effective exhaust speed ratio a_gs to pressure regulation control. The effective exhaust speed ratio a_gs determined as described above is used in the calculation of the predicted pressure Pp, thereby correcting the predicted pressure Pp.

[0113] Figure 11 Yes Figure 7 The graph of lines L1A, L1B, L2A, and L1C illustrates a case where the effective exhaust speed associated with the actual flowing gas is Se1, while the effective exhaust speed stored in storage unit 240 is Se2 (>Se1). For example, this corresponds to a case where the actual flowing gas is Xe, but the reference exhaust speed Ser(θ) (=Se2) stored in storage unit 240 is the effective exhaust speed associated with Ar. t = t20 is the time at which the effective exhaust speed ratio a_gs is determined.

[0114] The measured pressure value Pr is represented by line L1A for the effective exhaust speed Se1. Before the effective exhaust speed ratio a_gs is determined, that is, at t < t20, the predicted pressure Pp is calculated based on the reference exhaust speed Ser(θ) = Se2 stored in storage unit 240. Specifically, the effective exhaust speed ratio a_gs is a_gs = 1. At this time, the predicted pressure Pp is given by line L1C (Pp), which applies the predicted pressure for the effective exhaust speed Se2 to the measured pressure value Pr (line L1A), rather than by line L1B (Pp'), which represents the predicted pressure relative to line L1A. Consequently, an error occurs in the calculated predicted pressure Pp, resulting in a smaller value than the predicted pressure Pp' obtained using line L1B (Pp').

[0115] Based on the estimated effective exhaust speed Se0, at t = t20, the effective exhaust speed ratio s is determined as a_gs (= Se0 / Ser(θ) = Se1 / Se2 < 1). Once the effective exhaust speed ratio a_gs is determined at t = t20, the predicted pressure is calculated based on the determined effective exhaust speed ratio a_gs, namely the effective exhaust speed Se1 (= a_gs × Se2), and thus corrected to the error-free Pp' indicated by line L1B (Pp'). At t20, the predicted pressure Pp' is used to determine whether it exceeds the determination threshold (= Ps).

[0116] Figure 12 Yes Figure 7 The graph of lines L1A, L1B, L3A, and L1D illustrates a case where the effective exhaust speed associated with the actual flowing gas is Se1, and the effective exhaust speed stored in storage unit 240 is Se3 (<Se1). For example, this corresponds to a case where the actual flowing gas is He, but the reference exhaust speed Ser(θ) (=Se3) stored in storage unit 240 is the effective exhaust speed associated with Ar. t = t20 is the time at which the effective exhaust speed ratio a_gs is determined.

[0117] The measured pressure value Pr is represented by line L1A for the effective exhaust speed Se1. Before the effective exhaust speed ratio a_gs is determined, that is, at t < t20, the predicted pressure is calculated based on the reference exhaust speed Ser(θ) = Se3 stored in storage unit 240. Specifically, the effective exhaust speed ratio a_gs is a_gs = 1. At this time, the predicted pressure is given by line L1D (Pp), which applies the predicted pressure Pp for the effective exhaust speed Se3 to the measured pressure value Pr (line L1A), rather than by line L1B (Pp'), which represents the predicted pressure Pp' relative to line L1A. Consequently, an error occurs in the calculated predicted pressure Pp, resulting in a larger value than when line L1B (Pp') is used.

[0118] Based on the estimated effective exhaust speed Se0, at t = t20, the effective exhaust speed ratio s is determined as a_gs (= Se0 / Ser(θ) = Se1 / Se3 > 1). Once the effective exhaust speed ratio a_gs is determined at t = t20, the predicted pressure is calculated based on the determined effective exhaust speed ratio a_gs, namely the effective exhaust speed Se1 (= a_gs × Se3), and thus corrected to the error-free Pp' indicated by line L1B (Pp'). At t t20, the predicted pressure Pp' is used to determine whether it exceeds the determination threshold (= Ps).

[0119] Furthermore, the faster the pressure response condition, the closer the pressure measurement value Pr is to the target pressure value Ps when it is met. Therefore, there are cases where the a_gs value is determined after the predicted pressure Pp exceeds the judgment threshold (e.g., the target pressure value Ps), leaving too little time for correction. In such cases, the predicted pressure Pp, which uses the reference gas data (reference exhaust speed Ser(θ)) stored in storage unit 240, is used—that is, the predicted pressure Pp with a_gs = 1 in Se0 = a_gs × Ser(θ)—to continue the determination of whether the judgment threshold has been exceeded.

[0120] The above describes the case where the target pressure value Ps is greater than the starting pressure Pstt. Conversely, the case where the target pressure value Ps is smaller than the starting pressure Pstt can also be inferred and applied to pressure regulation, thereby improving the accuracy of pressure regulation control.

[0121] (Application of Se0 and a_gs to Initial Correction Processing)

[0122] In the above embodiment, when the opening θ is fixed at θse0 during pressure regulation, the effective exhaust speed Se0 or the effective exhaust speed ratio a_gs is calculated and used to correct the predicted pressure or applied to pressure regulation control, thereby improving the pressure regulation response. The following describes the application of the effective exhaust speed Se0 or the effective exhaust speed ratio a_gs to the initial correction process.

[0123] Initial calibration processing is the process of calibrating the reference gas data (reference pumping speed Ser(θ)) stored in the storage unit 240 to data suitable for the actual vacuum system. When a calibration command is input to the pressure regulation control unit 21 from an external device (e.g., the main controller of the vacuum processing apparatus), the pressure regulation control unit 21 switches from normal pressure regulation mode to calibration mode and performs a series of calibration processes. For example, a user operates the operating unit of the main controller of the vacuum processing apparatus according to the calibration manual to transmit the command to the valve control unit 1b, causing the pressure regulation control unit 21 to perform the calibration process.

[0124] Figure 13 、 Figure 14 This is a flowchart showing a series of processes in the calibration mode. Figure 15 3 is a graph showing the opening measurement value θr (line L31 ) detected by the encoder 130 and the pressure measurement value Pr (line L32 ) measured by the vacuum gauge 31 during the calibration process. Figure 13 、 Figure 14 The flowchart is executed by the voltage regulation control unit 21 and is started when a correction processing instruction is input to the voltage regulation control unit 21 from an external device.

[0125] exist Figure 13In step S10, it is determined whether the chamber volume V of the vacuum chamber 3 is stored in the storage unit 240. When the chamber volume V of the vacuum chamber 3 is not stored in the storage unit 240, the process proceeds to step S12, and an instruction to request the sending of the chamber volume V is sent to the external device. In the following, the external device is set as the main controller MC of the vacuum device for explanation. The main controller MC displays the request instruction for the existence of the chamber volume V on the display device of the main controller MC, etc., and urges the user to input the chamber volume V. In step S14, it is determined whether the chamber volume V has been received from the main controller MC. If it has been received (yes), the chamber volume V is stored in the storage unit 240 in step S16, and then the process proceeds to step S20.

[0126] If it is determined in step S10 that the chamber volume V is stored in the storage unit 240, or if the chamber volume V is stored in step S16, the opening degree is set to θ=100% in step S20. Figure 2 The calculation unit 210 outputs the opening instruction θ = 100%. Here, θ = 100% is set, but it can also be different from 100%. Figure 15 If θ=100% is set at t=t1, the pressure measurement value Pr (line L32) decreases.

[0127] In step S30, an instruction is sent to the main controller MC to allow the specified gas type to flow only at a specified flow rate Qin0. The user inputs a command to the main controller MC to allow the gas type specified in the calibration process manual to flow only at a specified flow rate Qin0. In step S40, it is detected whether gas has flowed in based on the increase in the pressure measurement value Pr. If gas inflow is detected in step S40, the process proceeds to step S50, and the opening θ is reduced from θ = 100% to the specified opening θtest. Figure 15 In the process, gas inflow is detected at t=t3, and the opening θ is set to θ=θtest. The opening θtest is equivalent to Figure 5 The fixed opening θse0.

[0128] If the processing of step S50 is completed, the process proceeds to Figure 14 Step S60. In step S60, the gas flow rate Qin0 is calculated based on equations (5) and (6), and the gas type is estimated based on the effective exhaust speed ratio a_gs calculated using equation (9). The flow rate Qin0 is calculated using equation (10) by multiplying both sides of equation (5) by Pr2 and by Pr1.

[0129] Qin0=V×(Pr2×dPr1 / dt-Pr1×dPr2 / dt) / (Pr2-Pr1)…(10)

[0130] Alternatively, it can be calculated and obtained from the equations (6) and (8) according to Qin0=V×dPr2 / dt+a_gs×Ser(θtest)×Pr2.

[0131] Furthermore, when the reference exhaust speed Ser(θ) stored in the storage unit 240 in equation (9) is the valve conductance C(θ), the effective exhaust speed ratio a_gs is calculated using the following equation (11). When the opening θtest is small, the influence of the valve conductance on the effective exhaust speed becomes dominant, and the effective exhaust speed Se becomes a value close to the valve conductance C.

[0132] a_gs=-(V / C(θtest))×(dPr2 / dt-dPr1 / dt) / (Pr2-Pr1)…(11)

[0133] In addition, the effective exhaust speed ratio a_gs calculated by formula (11) uses the valve conductance C as a reference. As the valve conductance C stored in the storage unit 240 in the initial state, the conductance related to Ar gas is generally used. In the following, the conductance when Ar gas flows is used for explanation. Therefore, the gas types recorded in the manual record the same gas types as the gas types of the valve conductance C. If the effective exhaust speed ratio a_gs calculated by formula (11) is approximately 1, it can be determined that Ar gas has been introduced as instructed. In addition, the calculation error of the effective exhaust speed ratio a_gs can be taken into account and an allowable range Δ such as 1±Δ can be set for the judgment standard. If it is 1-Δa_gs 1+Δ, it is determined that Ar gas has been introduced.

[0134] In step S70, it is determined whether the gas flow Qin0 and gas type estimated in step S60 meet the conditions described in the manual. If the conditions are met, the process proceeds to step S80. On the other hand, if the conditions are not met, that is, if at least one of the gas flow Qin0 and gas type is different from the conditions described in the manual, the process proceeds to step S72, and a command D to confirm the gas introduction conditions is sent to the main controller MC, and the process returns to step S80. Figure 13 Step S40.

[0135] On the other hand, when the process proceeds from step S70 to step S80, a process is performed to obtain the pressure measurement values ​​Pr at a plurality of opening degrees θ(i). In addition, i=1 to N (positive integers). In the correction process, the opening degree instruction θ(i) for correction is output from the operation unit 210. Figure 15 In the example shown in , the opening θ is changed from 100% to 0% at t=t11. The pressure measurement value Pr rises due to the change in the opening, and when the pressure measurement value Pr stabilizes and becomes approximately constant, the pressure measurement value Pr (1) is obtained. Similarly, at Figure 15At time t12, time t13, ..., and time t1N, the opening command θ(i) is output in the order of θ(2), θ(3), ..., and θ(N). For each opening command θ(i), a pressure measurement value Pr(2), a pressure measurement value Pr(3), ..., and a pressure measurement value Pr(N) are obtained. That is, the pressure measurement value Pr(θ(i)) can be obtained for N openings θ(i).

[0136] In step S90, the effective exhaust speed Se(θ(i)) is calculated by the following formula (12) using the gas flow Qin0 and the obtained pressure measurement value Pr(θ(i)). Formula (11) is strictly a formula that is valid under equilibrium conditions. Figure 15 The time intervals of t11, t12, t13, ..., t1N are set large enough to roughly satisfy the equilibrium state.

[0137] Se(θ)=Qin0 / P(θ)…(12)

[0138] In addition, in the calculation of the effective exhaust speed in steps S80 and S90, the effective exhaust speed Se(θ) is calculated based on the pressure measurement value Pr after the opening degree change by waiting until it substantially converges, and the effective exhaust speed Se(θ) is calculated using equation (12) based on the pressure measurement value Pr. However, the effective exhaust speed Se(θ) can also be calculated using equation (7) by obtaining the pressure change rate. In this case, it is not necessary to wait until the pressure measurement value Pr substantially converges, thereby shortening the time required for the calibration process.

[0139] In step S100, the effective exhaust speed Se(θ(i)) obtained by the correction process is stored in the storage unit 240 of the pressure regulation control unit 21. If the effective exhaust speed Se is already stored in the storage unit 240, the effective exhaust speed Se(θ(i)) obtained by the correction process is used to correct the effective exhaust speed Se or directly overwrite the stored effective exhaust speed Se.

[0140] As described above, steps S70 and S72 are provided to confirm whether the user has flowed only the specified flow rate Qin0 of the gas type specified in the manual. This prevents calibration processing using an incorrect gas type or flow rate. If the gas type or flow rate does not comply with the manual, pressure regulation accuracy may decrease. However, performing the calibration process as described above prevents this decrease in pressure regulation accuracy.

[0141] In addition, Figure 13In the flowchart shown in FIG, the processing of steps S10 to S16 is provided, and when the chamber volume V is not stored in the storage unit 240, the user is prompted to input the chamber volume V. However, the chamber volume may be estimated and calculated using the pressure-increasing method and stored in the storage unit 240. In the pressure-increasing method, the chamber volume V is calculated using the following equation (13) based on the pressure change rate dP / dt when the gas introduction amount Qin is fixed, with the valve body opening set to 0% and the valve body pressed to achieve a completely sealed state (effective exhaust speed = 0).

[0142] V=Qin / (dP / dt)…(13)

[0143] Alternatively, the calibration gas type is not limited to Ar gas; for example, the gas type can be selected from Ar gas and nitrogen gas. In this case, the valve conductance C stored in storage unit 240 is the value when Ar gas flows. Therefore, the effective pumping speed ratio a_gs = 1 for Ar gas and the effective pumping speed ratio a_gs = 1.2 for nitrogen gas, using Ar gas as a reference, are initially stored in storage unit 240. Furthermore, if the user selects Ar gas as the gas type to be introduced, a_gs = 1.2 for nitrogen gas is selected based on the selection information as the effective pumping speed ratio a_gs used for gas type determination. Specifically, if the effective pumping speed ratio a_gs calculated in step S60 is approximately 1.2, it can be determined that the introduction condition is satisfied if nitrogen gas, as declared by the user, is being introduced.

[0144] It will be understood by those skilled in the art that the exemplary embodiments are specific examples of the following forms.

[0145] [1] A valve control device in one form controls the valve body opening based on the pressure measurement value of a chamber in which a vacuum valve is installed and the pressure target value of the chamber, thereby regulating the pressure of the chamber. The valve control device comprises: an opening setting unit for fixing the valve body opening to a certain value when introducing gas into the chamber; and an inference unit for inferring exhaust information of the vacuum valve based on information on changes in the pressure measurement value when the valve body opening has been fixed to a certain value; the valve control device controls the valve body opening based on the exhaust information, thereby regulating the pressure of the chamber.

[0146] For example, Figure 2The calculation unit 210 functions as an estimation unit and an opening setting unit. Based on the pressure measurement value change information when the valve body opening is fixed at a constant value θse0, namely, the pressure change rate dPr1 / dt and the pressure change rate dPr2 / dt, the calculation unit 210 estimates the effective exhaust speed Se0 calculated by equation (7), the effective exhaust speed ratio a_gs calculated by equation (9), and the flow rate Qin0 calculated by equation (10) as exhaust information. This exhaust information corresponds to the type of gas actually flowing. Therefore, by calculating the predicted pressure Pp or performing pressure regulation control based on the estimated exhaust information, the accuracy of the pressure regulation control can be improved.

[0147] [2] The valve control device described in [1] further includes a predicted pressure calculation unit, which calculates a pressure prediction value relative to the pressure measurement value after a specified time based on the exhaust information, and increases the valve body opening if the pressure prediction value exceeds an upper limit threshold, or decreases the valve body opening if the pressure prediction value exceeds a lower limit threshold.

[0148] like Figure 11 As shown in Figure 2, the estimated effective exhaust speed becomes the effective exhaust speed Se1 corresponding to the gas type actually flowing. The pressure prediction value (predicted pressure) calculated using this effective exhaust speed Se1 becomes the error-free predicted pressure Pp' shown by line L1B. As a result, the timing of reversing the valve body opening θ toward the increase direction can be prevented from being delayed or advanced, and overshoot in the pressure response or, conversely, a prolonged pressure regulation time can be prevented. Furthermore, the estimated effective exhaust speed (Se0) and the effective exhaust speed ratio a_gs have the relationship shown in equation (8). Therefore, the effective exhaust speed ratio a_gs, which serves as gas type information, can be estimated as exhaust information, and the pressure prediction value can be calculated based on the estimated effective exhaust speed ratio a_gs and equation (8).

[0149] [3] In the valve control device described in [1], the exhaust information includes information on the type of gas introduced into the chamber, and the inference unit infers the gas type information based on a reference exhaust speed associated with a reference gas and the change information. An effective exhaust speed ratio a_gs, which is one of the exhaust information, is a value corresponding to the type of gas introduced into the vacuum chamber 3, as shown in equation (8), and is inferred based on a reference exhaust speed Ser associated with a reference gas (e.g., Ar gas), pressure Pr1, pressure Pr2, and pressure change rates dPr1 / dt and dPr2 / dt, as shown in equation (9).

[0150] [4] In the valve control device described in [1], the exhaust information includes information related to at least one of the flow rate and gas type of the gas introduced into the chamber, and the valve control device includes: a judgment unit that judges whether the inferred exhaust information is consistent with the reference flow rate and reference gas type; and an output unit that outputs the judgment result of the judgment unit.

[0151] like Figure 13 、 Figure 14 As shown in the calibration process shown in FIG, the flow rate and gas type of the gas introduced into the vacuum chamber 3 are estimated as exhaust information (step S60). Based on the estimation result, it is determined whether the estimated gas flow rate and gas type are consistent with the gas flow rate (reference flow rate) and gas type (reference gas type) specified in the manual (step S70), and the determination result is output (step S72). The user can use this determination result (condition confirmation command D) to recognize that the introduction conditions (gas flow rate and gas type) of the introduced gas are different from the reference introduction conditions (reference flow rate and reference gas type), and can perform calibration processing under the correct gas introduction conditions.

[0152] [5] One embodiment of an inference device is an inference device for inferring exhaust information of a vacuum valve installed in a chamber. Based on information on changes in the pressure measurement value of the chamber when the valve body of the vacuum valve is at a constant opening, at least one of the valve exhaust speed, the flow rate of the gas introduced into the chamber, and the gas type information of the gas is inferred as the exhaust information. The inference result is used for valve control, thereby enabling calculation of the predicted pressure Pp or pressure regulation control based on the inferred exhaust information, thereby achieving improved accuracy of the pressure regulation control. In addition, the inference device can be configured independently of the valve control device or as a structure incorporated into the valve control device.

[0153] While various embodiments and modifications have been described above, the present invention is not limited to these. Other configurations that are conceivable within the technical scope of the present invention are also encompassed within the scope of the present invention. For example, in the above embodiment, the pressure regulating control unit 21 of the valve control device 1b is used to perform the calibration process. However, a separate calibration device may be provided separately from the valve control device 1b, and the calibration process may be performed based on commands from the calibration device.

Claims

1. A valve control device for controlling the opening of a valve body based on a pressure measurement value of a chamber in which a vacuum valve is installed and a pressure target value of the chamber to regulate the pressure of the chamber, wherein: include: an opening setting portion for fixing the valve body opening to a constant value when gas is introduced into the chamber; as well as an estimating unit that, in a state in which the valve body opening is fixed at a constant level, estimates exhaust information of the vacuum valve based on a ratio of a calculated amount calculated based on a difference between pressure measurement values ​​at a first time and a second time and a pressure change rate over time of the pressure measurement values ​​at the different times, and a difference between the pressure measurement values ​​at the first time and the second time; The valve control device controls the valve body opening based on the exhaust information to perform pressure regulation control of the chamber.

2. The valve control device according to claim 1, wherein: The system further includes a predicted pressure calculation unit that calculates a predicted pressure value after a predetermined time has passed relative to the pressure measurement value based on the exhaust information. If the pressure prediction value exceeds an upper limit threshold, the valve body opening is increased, or if the pressure prediction value exceeds a lower limit threshold, the valve body opening is decreased.

3. The valve control device according to claim 1, wherein The exhaust information includes gas type information of the gas introduced into the chamber. The estimating unit estimates the gas type information based on a reference exhaust speed related to a reference gas and the ratio.

4. The valve control device according to claim 1, wherein The exhaust information includes information related to at least one of a flow rate and a type of gas introduced into the chamber, and the valve control device includes: a determination unit for determining whether the estimated exhaust information is consistent with a reference flow rate and a reference gas type; as well as The output unit outputs the determination result of the determination unit.

5. An inference device, wherein: This is an inference device that estimates the exhaust information of the vacuum valve installed in the chamber. In a state where the valve body opening of the vacuum valve is fixed, at least one of the valve exhaust speed, the flow rate of the gas introduced into the chamber, and the gas type information of the gas is inferred as the exhaust information based on the ratio of a calculated amount to a difference, wherein the calculated amount is an amount calculated based on the difference between the pressure measurement values ​​at the first moment and the second moment and the pressure change rate of the pressure measurement values ​​at the different moments over time, and the difference is the difference between the pressure measurement values ​​at the first moment and the second moment.

Images