JP2001147723A - Method for controlling flow rate - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a flow rate controlling method, capable off executing quick flow rate control free from overshooting by switching the flow rate control system on the way. SOLUTION: In the case of suddenly changing a flow rate from a zero flow rate state up to a prescribed flow rate set value by a flow rate control method for controlling the flow rate of fluid, by controlling a flow rate control valve 10 arranged fluid passage 4 allowing fluid to flow from a flow rate control part 22, the control part 22 outputs an initial, voltage value smaller than flow start driving voltage, by which the fluid starts to flow by a slight voltage value to the flow rate control valve as an initial control input and then immediately shifts the control to speed type PID control, by using the initial input as a reference. Since the flow rate control system is switched on the way, quick flow rate control which is a free of overshooting can be executed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a flow control method for controlling a flow rate of a relatively small fluid such as gas.

[0002]

2. Description of the Related Art Generally, in order to manufacture a semiconductor product or the like, for example, a CVD film formation or an etching operation is repeatedly performed on a semiconductor wafer or the like. For example, a precision gas flow control device is used. This type of gas flow control device mainly includes a sensor unit for detecting a mass flow rate of a trace gas, a valve unit, and a control circuit unit for controlling the valve unit. The sensor section has a sensor in which an electric heating coil is wound around a sensor tube through which a small percentage of the total gas amount passes, and most of the gas flows through the bypass. The control circuit controls the valve opening of the valve on the basis of the value detected by the sensor, so that the gas flow at the set value flows. Further, in order to control the valve opening, since the entire gas flow rate itself is very small, it is necessary to accurately control the valve opening within a stroke range of, for example, about several tens of μm. Since a large change in thrust can be generated within the stroke range, a laminated piezoelectric element is generally used, so that the valve opening of the diaphragm made of a diaphragm is operated based on a set value. It has become.

By the way, as a method of controlling a gas flow rate by controlling a valve opening degree, that is, an operation amount of a valve,
A position PID control method, a speed-type PID control method, and the like are known. In controlling a gas flow rate, particles that cause a defect of a product do not wind up in a semiconductor processing chamber due to a sudden gas flow. Need to be controlled. Therefore, the occurrence of overshoot that causes the generation of particles must be suppressed as much as possible.

[0004]

By the way, the above-mentioned control formula of the position PID control is generally represented by the following equation 1 in a digital system.

[0005]

(Equation 1)

Here, _mn is the operation amount of the n-th valve,
k is a proportional constant, Ts is the sampling time, Ti is the integration time (constant concept has entered time), e _n is the actual amount of operation of the deviation with respect to the set value, Td is determined by differentiation time constant, m ₀ is required Respectively indicate the amount of bias added. Accordingly, Ts / TiΣe _n term is accumulation of the deviation is determined represents the integral term, Td / Ts (e _n -e
The term _n-1 ) represents the derivative term, and the gradient of the deviation is obtained.

According to this integral term, for example, even if there is a slight deviation in the temperature control of the processing furnace requiring strictness, this is preferable because it is accumulated and corrected, but the response is slow as a whole. Also, when the set value is changed, the value of the integral term may act in a direction that prevents the change.Therefore, in actual control, when the set value is changed, the value of the integral term should be set to zero by software. Has become.

Further, since the differential term takes the difference between the deviations, it is not very useful when there is little change in the manipulated variable, but acts in a suppressed manner when the deviation changes abruptly. Then, in the actual flow control device, if the valve is set to the fully closed state at the initial stage, for example, the flow rate 50
When a set value of% is input, a large deviation occurs. Therefore, in order to prevent this, m ₀ is set to, for example, an operation amount of 50%, and the valve is opened.

However, in general, the response speed of the sensor is not so fast. For example, when the set value is sharply reduced, the output of the sensor for which feedback is being applied indicates the previous state for a short time and the deviation is indicated. Becomes zero, then m
There is a problem that the valve opens to 50% due to the advantage of _0, and as a result, gas flows rapidly and overshoot occurs. On the other hand, the control formula of the speed-type PID control is generally expressed in a digital system as shown in the following Expression 2.

[0010]

(Equation 2)

Here, Δm _n indicates a difference in valve opening degree,
The present valve opening degree _mn is recognized, and an equation of what kind of change is given thereto is given. According to this control, as is clear from the equation, only the change in the valve opening is calculated, and the sum of the output and the previous position is calculated by the output processing unit, so that the internal output calculation value exceeds the output signal range. It has the advantage that there is no problem, that the output calculation value is unlikely to change excessively, and that the problem does not occur due to saturation in the internal calculation of the integral value.However, steps such as when the set value changes rapidly The response speed at the time of response is the position PID described above.
There is a problem that it is considerably inferior to the control case. FIG. 6 shows this state. When the set value changes rapidly from 0 V to 5 V in a stepwise manner in a digital system, for example, the opening degree (operating amount) of the valve slowly rises as shown by a dotted line, Therefore, the actual flow rate does not overshoot, but the flow rate slowly increases toward the set value, and there is a problem that the response speed is poor.

In particular, the actual flow characteristic of the valve section is such that the flow rate change is small when the valve opening is small, and the flow rate change is large when the valve opening is a certain degree or more. There is a problem that the response when the set value changes stepwise from zero flow rate in the fully closed state is considerably poor.

The present invention focuses on the above problems,
It was created to solve this effectively. An object of the present invention is to provide a flow control method capable of performing flow control quickly and without overshoot by switching the flow control method in the middle.

[0014]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention controls the flow rate of the fluid by controlling a flow rate control valve provided in a fluid passage through which the fluid flows from a flow rate control unit. In the flow control method, when the flow rate suddenly changes from a zero flow state to a predetermined flow rate set value,
The flow control unit outputs to the flow control valve a small initial voltage value smaller than the flow-out drive voltage by which the fluid starts flowing out as a small initial operation amount, and thereafter,
The control is immediately shifted to the speed type PID control based on the initial operation amount.

[0015]

Since the present invention is configured as described above, when the flow rate is rapidly changed from a zero flow state to a predetermined flow rate set value, the flow-out drive voltage at which the fluid starts flowing out regardless of the flow rate set value. An initial voltage value smaller than the initial voltage value by a slightly smaller voltage value is immediately applied to the flow control valve, and thereafter, after a short time elapses, for example, until the sensor section is stabilized, the control is shifted to the speed-type PID control. As a result, the drive voltage is suddenly applied in a step at a time until immediately before the flow control valve is opened.
By shifting to control, the flow control valve becomes a speed type PID
Under the control, the valve opening gradually increases toward the steady operation amount, and finally the flow rate corresponding to the flow rate set value flows. In this way, even if the set flow rate value changes rapidly from the zero flow state, it is possible to control the flow rate to quickly change to the desired set flow rate value without causing overshoot or the like. Become.

In this case, when the fluid indicated by the set flow rate is actually flowing due to a change in the environment in which the fluid is used, for example, due to a gas differential pressure in a piping system to which the fluid passage is connected or an aging of the fluid passage. In some cases, the actual operation amount may deviate from the steady operation amount at this time, but it is preferable to always update the operation amount when the deviation becomes excessively large.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a flow control method according to the present invention will be described below in detail with reference to the accompanying drawings. FIG. 1 is a schematic configuration diagram showing a gas mass flow controller for implementing a flow control method according to the present invention, FIG. 2 is a block diagram showing a configuration of a flow controller of the controller shown in FIG. 1, and FIG. FIG. 4 is a graph showing a change in a flow rate set value, a flow rate, and an operation amount when the control method of the present invention is performed. FIG. 5 is a graph showing some characteristic curves extracted from FIG.

First, a gas mass flow controller for implementing a flow control method according to the present invention will be described. As shown in the figure, the gas mass flow control device 2 has a fluid passage 4 formed of, for example, stainless steel or the like, and most of the flow is provided upstream of the fluid passage 4 in the gas fluid flow direction. A flow bypass 6 is provided, and a flow control valve 10 having, for example, a diaphragm 8 as a valve body for controlling the flow rate of the gas fluid is provided downstream.

At both ends of the bypass 6, sensor pipes 14 of the mass flow rate detection sensor section 12 are connected, and a small amount of gas fluid can flow through the sensor pipe 14 as compared with the bypass 4. A pair of electric heating coils 16 for control are wound around the sensor tube 14, and a mass flow rate of the gas fluid is detected by a sensor control circuit 18 connected thereto.

The detected value here is the A / D of two channels.
The data is input to a flow control unit 22 composed of, for example, a microcomputer via a converter 20. A setting signal indicating a flow rate set value input from the outside is also provided by the A / D.
The data is input to the flow rate control unit 22 via the converter 20. The setting signal is usually an analog signal of 0 to 5 V, and is converted into a digital signal by the A / D converter 20 and then input to the flow control unit 22.

The flow control valve 10 has, for example, a laminated piezoelectric element 24 which generates a large thrust change within a small stroke range as an actuator for vertically driving the diaphragm 8. Is controlled by the drive signal (drive voltage). Drive unit 2
6, the digital drive signal from the flow controller 22 is D /
The signal is input after being converted into an analog signal by the A converter 28, and the operation amount of the flow control valve 10, that is, the valve opening is controlled so that a flow rate corresponding to the flow rate set value flows.

In this embodiment, the steady-state operation amount of the flow control valve corresponding to a plurality of flow rate setting values is measured in advance and stored in a table. Is obtained based on the above table, and an initial operation amount smaller than the obtained steady operation amount by a predetermined ratio is first output to the flow control valve 10, and thereafter, The flow rate control unit 22 is programmed so as to immediately shift to the speed-type PID control (represented by the aforementioned equation 2) based on the initial operation amount.

In other words, in other words, when the flow rate setting value is changed, first, the valve opening degree (initial operation amount) is slightly smaller than the valve opening degree (operation amount) corresponding to the newly set flow rate setting value. It opens or closes at a stretch, and then immediately shifts to speed-type PID control featuring stable control, and consequently shifts to the set flow rate quickly without overshooting.

The contents of the program are functionally represented, for example, as shown in FIG.
For example, an EEPROM (electrical) that rewritably stores the measurement result of the flow control valve measured in advance as 0.
al erasable programmable R
OM), a sensor stability confirmation unit 34 indicating whether or not the sensor 12 is stable, and a rewrite command unit 33 for notifying the timing of rewriting the table.

The sensor stability confirmation section 32 is a speed type PI
It informs the timing of switching to D control,
When a sudden decrease or increase in the gas flow rate is sensed, the sensor recognizes that the sensor is stable and notifies the arithmetic unit 30 that the sensor has stabilized. The time until the sensor stabilizes (response time) is very short, for example, on the order of μmsec, whereas the response time of the actuator that operates the diaphragm 8 is relatively long, for example, on the order of msec.

When the input set flow rate value is stable (constant), the rewriting command section 33 operates the actual operation amount output as a drive signal and the operation amount in the table of the flow rate set value at that time. A certain amount between the amount, for example, 10%
If the above difference occurs, a rewrite signal for rewriting the flow rate set value in the table is output, and the arithmetic unit 30 receiving the rewrite signal performs a predetermined calculation and rewrites. . All of the above operations are processed by the program. Here, the contents shown in Table 1 stored in the storage unit 32 will be described.

[0027]

[Table 1]

In Table 1, the flow rate set value Q represents a target flow rate input from the outside. For convenience, the maximum flow rate is divided into 10 steps of 10% from Q _{1 to} Q ₁₀ for convenience. It may be further subdivided. The manipulated variable indicates the valve opening of the control valve, and indicates the valve opening required when the flow rate represented by the flow rate set value is stably flowed. Here, generally, the characteristic between the operation amount (valve opening degree) and the flow rate of the flow control valve has a hysteresis characteristic as shown in FIG. 4, and the operation amount is increased and decreased. In some cases, the flow rate is different even at the same valve opening, and the flow rate is different even when the differential pressure is different.
For example, assuming that the valve opening is the same, the larger the differential pressure, the larger the flow rate, and the overall characteristic curve shows a similar shape. Therefore, as shown in Table 1, for the same flow rate set value Q, there is an operation amount mA when the valve opening increases and an operation amount mB when the valve opening decreases, and both of them are measured. Also, when the flow rate is zero, the valve is pressed with a strong force to prevent leakage, so that the drive voltage is, for example, 5
In the region lower than V, the flow rate is zero because the valve is not open, and when the drive voltage becomes larger than approximately 5 V, the valve starts to open and gas starts to flow.

The compensation for the point that the flow rate changes even with the same valve opening due to the difference in the differential pressure will be described later. The measurement of the manipulated variables mA and mB is performed after the gas flow control device 2 is actually installed in, for example, a piping system of a semiconductor manufacturing apparatus. For example, the flow rate is increased or decreased by 10%, and The valve opening (equivalent to the drive voltage of the actuator) is read when is stable, and the operation amounts mA and mB are read.
Is stored in the storage unit 32.

Then, the respective operation amounts mA and mB
, A predetermined ratio α (1 <0) is added to obtain the corresponding initial manipulated variables mAo and mBo, and stored similarly. Therefore, the initial operation amounts mAo and mBo are obtained by calculation as follows. mAo = α · mA mBo = α · mB Here, α is set to a value such that the flow rate does not overshoot even if the valve opening corresponding to the initial manipulated variable is established at a stretch, and is usually 0.6 to 0. 7 is set.

The pressure difference for forming Table 1 is set to the largest pressure difference expected in the environment where the flow control device is installed. For example, the gas pressure of the processing gas piping system in a semiconductor manufacturing apparatus is usually 3 kgf / cm ^{2 at} the maximum.
Therefore, at the time of preparing Table 1, data is taken under the basis of a differential pressure of 3 kgf / cm ² . Thus, no overshoot occurs even if this flow control device is provided in any piping system having a maximum differential pressure of 3 kgf / cm ² .
Further, the differential pressure of the piping system in which the flow control device is arranged is smaller than, for example, 3 kgf / cm ² , for example, 2 kgf / c.
If it is m ² , the respective operation amounts and initial operation amounts corresponding to the flow rate set values Q shown in Table 1 are all large (see FIG. 4).

Therefore, in order to ensure quick control speed, all the values of the flow rate set values Q in Table 1 are rewritten by the learning function so as to correspond to the values. Specifically, the flow control valve 10 when the flow rate specified by the flow rate set value is flowing stably
The actual manipulated variable (corresponding to the voltage of the drive signal) is monitored, and based on the ratio between the flow set value and the flow set value in Table 1 corresponding to the manipulated variable, the flow set value in Table 1 To be rewritten. The rewritten state is shown in Table 2.

[0033]

[Table 2]

In this case, it is considered that the manipulated variable actually changes microscopically due to the hysteresis characteristic. Therefore, when a difference of the manipulated variable of, for example, 10% or more occurs, the rewrite command unit 34 writes the data. Output a replacement signal. Also, a flow rate set value Q not stored in Table 1, for example, a flow rate set value Q
_When a flow rate set value between ₁ and Q ₂ of, for example, 15% is input, a value corresponding to each of the flow rate set values Q ₁ and Q ₂ from the manipulated variables and the initial manipulated variables by interpolation is determined. To Further, the steady operation amount and the initial operation amount in Tables 1 and 2 respectively correspond to the drive voltage, and this drive voltage is directly applied to the stacked voltage element 24.
In FIG. 1, reference numeral 34 denotes a D / A converter for converting a flow rate output into, for example, an analog signal of 0 to 5 V, and reference numeral 36 denotes, for example, an interface of the RS232C standard for communicating with the outside.

Next, an example of an embodiment performed by using the gas mass flow controller configured as described above will be specifically described. For example, when a gas fluid flows through the fluid passage 4 having a differential pressure of 3 kgf / cm ^2, a part of the gas flows through the sensor tube 14 of the sensor unit 12 and most of the gas flows through the bypass 6. Head to the gas-using system while being controlled.

The flow rate of the gas fluid flowing through the sensor pipe 14 is detected by the constant temperature difference type sensor 12, the mass flow rate flowing through the entire fluid passage 4 is obtained, and is input to the flow rate control unit 22. Then, feedback control is performed so that the detected value becomes the same as the flow rate set value input from the outside, and the valve opening is controlled by driving the laminated piezoelectric element 24 of the flow control valve 10. Become.

Now, a description will be given of a case where the flow rate set value greatly changes in this embodiment, for example, a case where the flow rate set value Q suddenly changes to Q _{3 in} Table 1 from a zero flow state as an example. As shown in FIG.
In the flow rate set value Q ₃ is input as a set signal, calculation unit 30 obtains the steady operation amount mA ₃ at a flow rate increases from the table 1 stored in the storage unit 31, over - the predetermined ratio α to the value The initial operation amount mAO ₃ at the time of increase is obtained by the calculation. If the initial manipulated variable is calculated in advance and tabulated in Table 1, the initial manipulated variable mAO ₃ (<mA ₃ ) may be directly read as a value corresponding to the flow rate set value Q _3. Good. This initial manipulated variable mAO ₃ is immediately transmitted from the flow control unit 22 to the flow control valve 10 as a drive signal, and the diaphragm 8 is opened up to the initial manipulated variable mAO ₃ at a stretch.

As a result, the gas suddenly flows out, and this gas flow is detected by the sensor 12 having a high response sensitivity. Then, the sensor stability confirmation section 32 (see FIG. 2) informs the arithmetic section 30 that the sensor is stabilized. . Then, the operation unit 30
Switches the control method to the speed-type PID control as shown in the above equation 2 at the point A. As described above, this control method recognizes the current valve opening and stably and slowly adjusts the valve opening toward the set value based on the current valve opening. , Ie, the initial manipulated variable mAO ₃ stably and slowly rises. During this time, the gas flow rate is gradually increased rapidly towards the flow setpoint Q _3, the initial operating quantity MAO _3, which are set smaller than the steady operation amount mA ₃ as overshoot does not occur over without shoot occurs, it is possible to increase the gas flow rapidly to flow setpoint Q _3. Accordingly, since no overshoot occurs, for example, in a semiconductor manufacturing apparatus, it is possible to prevent the particles from being wound up due to a sudden gas entrainment.

This operation is the same not only when the flow rate set value increases stepwise from zero flow rate, but also when the flow rate set value increases stepwise from a state in which the gas fluid flows to some extent. Conversely, the same operation is performed when the flow rate set value decreases stepwise from a state in which the gas fluid is flowing to some extent. However, in this case, since the gas flow rate decreases, a value is selected or calculated from the column of the steady operation amount mB at the time of decrease and the initial operation amount mBO at the time of decrease. In this case, when the flow rate set value Q is very low, for example, at a set flow rate of about 0.5% of the maximum flow rate, the voltage value of the initial manipulated variable mAO at this time depends on the flow rate of the flow control valve. The drive signal (drive voltage) immediately before may be slightly smaller than the drive signal (drive voltage) to be started. This will be described with reference to FIG.
For example, assuming that the steady operation amount at a set flow rate of 0.5% of the maximum flow rate is 7 volts and α = 0.6, the initial operation amount mAO is 4.2 volts (= 7V × 0.6), The valve begins to open and has a voltage below which gas flows out, ie, less than 5 volts.

When a flow rate setting value Q not stored in Table 1 is input (in practice, it is considered that there are many cases of this kind), this flow rate setting value Q is stored in Table 1. The corresponding steady-state operation amount and initial operation amount are obtained from the steady-state operation amount corresponding to the flow rate set value on both sides of the position located at the predetermined position and the initial operation amount at that time by an interpolation method such as proportional division. The time Δt from the time t to the point A is very small and depends on the response of the sensor.
It is set to about c.

Further, in the above operation, in preparing Table 1, actual measurement was carried out under the condition of a differential pressure of 3 kgf / cm ² , and the piping system on which the flow control valve was mounted was also operated at a differential pressure of 3 kgf / cm ^2.
Since it is the case of the piping system of ² , the use of the values shown in Table 1 as it is causes almost no problem. However, due to the nature of the gas used, it is used in an environment where the differential pressure is lower than 3 kgf / cm ^2. May also occur.

In this case, as described above with reference to FIG. 4, if the differential pressure is different, the flow characteristics are different even at the same valve opening, so that Table 1 needs to be corrected. As shown in FIG. 5, Table 1 measured at a differential pressure of 3 kgf / cm ² (corresponding to Table 1) corresponds to, for example, when a flow rate control device is used under a differential pressure of 0.5 kgf / cm ^2. 0.5k differential pressure
Although it is necessary to create a flow rate characteristic of gf / cm ² , in the present embodiment, a learning function is provided, and automatically
In Table 2 corresponding to a differential pressure of 0.5 kgf / cm ² .

More specifically, for example, when the gas flow rate is increased, an initial operation amount is initially selected or calculated based on Table 1, and the valve opening is controlled by the initial operation amount. However, in a state where the flow rate of the flow rate set value Ql flows stably, the actual manipulated variable mAk at that time differs from the value of the steady-state manipulated variable mAl in the case of increase in Table 1. Therefore,
The flow rate set value Qk in Table 1 corresponding to the actual manipulated variable mAk at this time is obtained, and the flow rate set value Q ′ in Table 1 is rewritten by the Ql and Qk to obtain the flow rate set value Q ′ shown in Table 2. create. This calculation formula is given as follows. Q '= Q · Q1 / Qk

In the actual control, as shown in FIG. 2, the rewriting command section 33 constantly monitors the actual operation amount indicated by the flow rate set value and the drive signal. When the difference between the actual manipulated variable and the steady-state manipulated variable in Table 1 corresponding to the flow rate set value at that time exceeds a predetermined range, for example, 10% or more, a rewrite signal is output. Rewriting operation is performed,
A new table as shown in Table 2 is created and stored. Thereafter, the above-described control is performed based on Table 2.

By providing a learning function in this way and performing a table rewriting operation when necessary, it can be applied not only to a piping system having any differential pressure, but also to a gas cylinder. It is possible to cope with a change in the differential pressure due to a change in the original pressure. In the above-described embodiment, the switching of the control is performed by the operation of the sensor stability confirmation unit 32. However, this is merely an example, and the speed type PID control is performed immediately after the output of the initial operation amount. Any switching function may be provided as long as it can be shifted to.

In the method described above, if the flow rate set value Q is very small as described above, for example, the flow rate set value Q
Is about 0.5 to 1% of the maximum flow rate, the drive voltage corresponding to the initial manipulated variable mAO when the predetermined ratio α (for example, 0.6 to 0.7) is added is: There is a case where the voltage value becomes lower than the outflow drive voltage at which the fluid starts to flow out. The initial manipulated variable at this time is, of course, zero flow rate. This will be described with reference to FIG. FIG. 6 is a graph showing an example of the relationship between the drive voltage and the flow rate (corresponding to the valve opening). The drive voltage differs from the characteristics shown in FIG. 4 but has similar characteristics. As shown in FIG. 6, the relationship between the drive voltage and the flow rate is not linear, and when the voltage is gradually increased from the fully closed state (0 V), even if the voltage is increased, the flow rate does not change and the state of zero is maintained. There is an area shown. In FIG. 6, almost no gas flows up to 50V. Incidentally, in the case shown in FIG. 4, almost no gas flows up to 5V. The reason is that when the drive voltage is increased, a force is generated in a direction to open the valve body (diaphragm 10), but the force does not immediately displace, and the force (surface pressure) by which the valve body presses the valve seat is reduced. Used to do. The voltage rise after the surface pressure becomes zero appears as a displacement.

For the valve actuator of the mass flow controller, a solenoid, thermal expansion, or the like is used in addition to the above-described laminated piezoelectric element. Is designed to apply surface pressure. In addition, such a design and adjustment should be made in order to absorb the mechanical displacement of the valve and the displacement due to the difference in thermal expansion of the parts due to the change in the ambient temperature, and to prevent leakage with a slight displacement. is necessary. Thus, in the case shown in FIG.
0 V is the flow-out drive voltage. Under such characteristics, for example, when the maximum flow rate is 2000 cc / min and the flow rate set value is 10 cc / min which is 0.5% of the maximum flow rate value, for example, the drive voltage for the fixed operation amount mA at that time Becomes approximately 75 V. Therefore, if α = 0.6, the initial operation amount mAO is 45 V (= 75 V ×
0.6). The drive voltage 45V is slightly lower than the flow-out drive voltage 50V, and gas does not flow out at this time. After the initial operation amount is set to the drive voltage 45V, the speed-type PID control is performed as described above. The process shifts immediately, and gas starts flowing when the drive voltage exceeds approximately 50V.

As described above, when the drive voltage (initial voltage value) of the initial manipulated variable starts flowing and is slightly lower or lower than the drive voltage, for example, 50 V, the speed type PID is immediately thereafter.
Since the control is shifted to the control, the occurrence of overshoot can be reliably eliminated, and the desired flow rate can be promptly and stably set. Therefore, in this case, when the flow rate is suddenly changed from the zero flow state to the predetermined flow rate set value, regardless of the value of the flow rate set value Q at that time, a voltage smaller than the above-mentioned flow-out drive voltage, for example, approximately 50 V value,
For example, a voltage smaller by 5 V is output as a drive voltage to the flow control valve 10 as an initial voltage value (initial operation amount), for example, 45 V. Then, immediately thereafter, the process proceeds to the high-speed PID control in the same manner as described above. Therefore, in the case of this embodiment, the flow rate is reduced from zero to the maximum flow rate (flow rate 100%).
Even when the flow rate is rapidly changed, the driving voltage of 45 V is output as the initial voltage value.

In this embodiment, the control speed is slightly lower than the method described in Table 1, but the occurrence of overshoot is more reliably suppressed than in the method shown in Table 1. It is possible to quickly and stably set a desired flow rate. In this case, when obtaining the initial voltage value, as described above, for example, the maximum flow rate of 0.5
% May be obtained as a drive voltage corresponding to a predetermined flow rate set value with a small flow rate such as a flow rate of, for example, a voltage value smaller than 75 V by a predetermined ratio, for example, 45 V (= 75 V × 0.6), Alternatively, a flow driving voltage, for example, a predetermined ratio of 50 V, for example, 45% (= 50
(V × 0.9), or may be obtained as an initial voltage value by a predetermined amount, for example, 5 V lower than the flow-out drive voltage, for example, 50 V. These calculations and the like are, of course, performed by the flow controller 22.

In any case, since the characteristics of the flow control valve 10 are unique to each control valve, the flow-out drive voltage is obtained in advance at the factory shipment stage or the like. Further, in consideration of the quickness of control, the difference between the above-mentioned flow-out drive voltage and the initial voltage value is preferably set as small as possible within a range where the initial voltage value does not exceed the flow-out drive voltage. Is preferably in the range of 5 to 10 V, for example, in the range of 10 to 20% with respect to the flow-out drive voltage. Also, in this case, the flow-out drive voltage may change due to the aging of the flow control valve 10 or the like, so that
Of course, this is constantly monitored and updated. Further, when changing from a state in which a gas having a predetermined flow rate is already flowing to another flow rate, it goes without saying that the method described in Table 1 is used.

The numerical example shown in FIG. 6 is merely an example for facilitating understanding of the contents, and it is a matter of course that the present invention is not limited to this numerical example. FIG. 7 is a graph showing the relationship between the gas flow rate, the drive voltage, and the set value when controlled by the above method. As shown in the graph of FIG. 7, it can be seen that the gas flow rate stably reaches the predetermined flow rate in about 2 seconds without overshooting.

[0052]

As described above, according to the flow control method of the present invention, the following excellent functions and effects can be exhibited. When the flow rate is rapidly changed from a zero flow state to a predetermined flow rate set value, a desired flow rate can be quickly set without causing overshoot almost certainly. Therefore, when applied to a gas supply system of a semiconductor manufacturing apparatus, it is possible to prevent the particles from being rolled up due to overshoot, thereby contributing to an improvement in yield.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram showing a gas mass flow controller for implementing a flow control method according to the present invention.

FIG. 2 is a block diagram illustrating a configuration of a flow control unit of the control device illustrated in FIG. 1;

FIG. 3 is a graph showing changes in a flow rate set value, a flow rate, and an operation amount when the control method of the present invention is performed.

FIG. 4 is a graph showing a relationship between an operation amount of a flow control valve and a flow rate when a differential pressure is changed.

FIG. 5 is a graph showing some characteristic curves extracted from FIG. 4;

FIG. 6 is a graph showing an example of a relationship between a drive voltage and a flow rate (corresponding to a valve opening).

FIG. 7 is a graph showing a relationship between a gas flow rate, a driving voltage, and a set value.

FIG. 8 is a graph showing changes in a flow rate set value, a flow rate, and an operation amount when speed type PID control is performed from the beginning.

[Explanation of symbols]

 2 Gas mass flow control device 4 Fluid passage 6 Bypass 8 Diaphragm 10 Flow control valve 12 Mass flow detection sensor 14 Sensor tube 22 Flow control unit 24 Stacked piezoelectric element 26 Piezoelectric element drive unit 30 Operation unit 31 Storage unit 32 Sensor stability check unit 33 Rewriting command section

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) // F17D 3/01 H01L 21/302 B

Claims (3)

[Claims] 1. A flow control method for controlling a flow rate of a fluid by controlling a flow rate control valve provided in a fluid passage through which a fluid flows from a flow rate control unit, from a zero flow state to a predetermined flow rate set value. When rapidly changing the flow rate, the flow rate control unit outputs an initial voltage value smaller than the flow-out drive voltage at which the fluid starts flowing out by a small voltage value to the flow rate control valve as a small initial manipulated variable, and immediately thereafter. A flow control method characterized by shifting to speed type PID control based on the initial operation amount. 2. The flow rate according to claim 1, wherein the initial voltage value is a voltage value smaller by a predetermined ratio than a drive voltage of a steady operation amount corresponding to a predetermined flow rate set value with a small flow rate. Control method. 3. The flow control method according to claim 1, wherein the initial voltage value is a voltage value smaller by a predetermined amount than the outflow drive voltage.