Thermal flowmeter, flow control device, thermal flow measurement method, and recording medium

Abstract

The present application provides a thermal flowmeter, a flow control device, a thermal flowmeter measurement method, and a recording medium, in order to provide a thermal flowmeter capable of correcting errors caused by thermal siphoning with better accuracy than in the past, comprising: a sensor flow path (4) for flowing a fluid to be measured; an upstream resistance component (Ru) provided in the sensor flow path (4); a downstream resistance component (Rd) provided in the sensor flow path (4) further downstream than the upstream resistance component (Ru); a sensor output generator (5) that generates a sensor output corresponding to the flow of the fluid to be measured, based on an upstream voltage (Vu) output in accordance with changes in the upstream resistance component (Ru) and a downstream voltage (Vd) output in accordance with changes in the downstream resistance component (Rd); a tilt influence inferrer (6) that infers a tilt influence produced in the sensor output in accordance with the posture of the sensor flow path (4), based on at least the Prandtl number of the fluid to be measured; and a flow calculator (7) that calculates the flow of the fluid to be measured by correcting the tilt influence from the sensor output.

Description

Technical Field

[0001] The present invention relates to a thermal flowmeter that measures the flow rate of a measurement object based on outputs obtained from an upstream resistor element and a downstream resistor element provided in a sensor flow path through which a fluid to be measured flows. Background Art

[0002] For example, in semiconductor manufacturing processes, thermal mass flow controllers (MFCs) are used to supply various gases at a desired constant flow rate. A MFC consists of a block with internal flow paths for gases to flow in a generally predetermined direction; a thermal flow meter and fluid control valve attached to the block; and a control board that controls the fluid control valve.

[0003] The thermal flowmeter includes a sensor flow path that is a roughly U-shaped capillary tube. The sensor flow path branches off from the internal flow path that serves as the main flow path and merges with the internal flow path again. In the sensor flow path, an upstream resistor component and a downstream resistor component are provided in a portion where the internal flow path within the block is oriented in roughly the same direction as the flow direction of the fluid (gas). The voltage applied to each resistor component is controlled so that the temperature of these resistor components is constant. Since the voltage difference applied to each resistor component changes according to the flow rate of the gas flowing in the sensor flow path, the flow rate of the gas can be calculated based on the voltage difference.

[0004] However, if Figure 10 As shown by the arrow in (a), the packaged flow control device, like the mass flow controller, is designed based on the horizontal placement so that the gas flows in a roughly horizontal direction. Figure 10 As shown by the arrow in (b), if the flow control device is placed vertically so that the gas flows vertically, the thermal flowmeter's zero output will shift depending on the installation orientation and the gas pressure due to a phenomenon known as thermosiphoning. In other words, the flow control device should output a value of zero, but convection within the sensor flow path caused by thermosiphoning may cause the output to deviate from zero. This can lead to measurement errors.

[0005] In order to reduce such measurement errors, Patent Document 1 discloses providing a gyro sensor in a flow control device and performing correction of a zero-point output according to the detected posture.

[0006] However, the thermosiphon phenomenon is affected not only by the posture of the flow control device but also by the pressure, thermal conductivity, specific heat and other thermophysical properties of the gas being measured. Therefore, the correction method of Patent Document 1 is difficult to achieve with sufficient accuracy.

[0007] Prior art literature

[0008] Patent Document 1: Japanese Patent Publication No. 2008-506117 Summary of the Invention

[0009] In view of the above-mentioned problems, an object of the present invention is to provide a thermal flow meter capable of correcting errors caused by the thermosiphon phenomenon with higher accuracy than conventional flow meters.

[0010] That is, the thermal flow meter of the present invention is characterized in that it includes: a sensor flow path for the flow of the fluid of the measurement object; an upstream side resistance component, which is arranged in the sensor flow path; a downstream side resistance component, which is arranged in the sensor flow path at a downstream side of the upstream side resistance component; a sensor output generator, which generates a sensor output corresponding to the flow rate of the fluid of the measurement object based on the voltage output from the flow detection circuit including the upstream side resistance component and the downstream side resistance component; a tilt effect inference device, which infers the tilt effect generated in the sensor output according to the posture of the sensor flow path based on at least the Prandtl number of the fluid of the measurement object; and a flow rate calculator, which corrects the tilt effect according to the sensor output and calculates the flow rate of the fluid of the measurement object.

[0011] In addition, the thermal flow measurement method of the present invention is a flow measurement method using a thermal flow meter, the thermal flow meter including: a sensor flow path for the flow of the fluid of the measurement object; an upstream side resistor component, arranged in the sensor flow path; and a downstream side resistor component, arranged in the sensor flow path at a position further downstream than the upstream side resistor component. The thermal flow measurement method is characterized in that it includes: generating a sensor output corresponding to the flow rate of the fluid of the measurement object based on the voltage output from the flow detection circuit including the upstream side resistor component and the downstream side resistor component; inferring the tilt effect generated in the sensor output according to the posture of the sensor flow path based on at least the Prandtl number of the fluid of the measurement object; and correcting the tilt effect based on the sensor output to calculate the flow rate of the fluid of the measurement object.

[0012] In this case, the tilt effect estimator estimates the tilt effect based on the Prandtl number, a value determined by the physical properties of the fluid being measured and influenced by the fluid's pressure and thermal conductivity. This allows the posture effect to be estimated, taking these influences into account. Consequently, measurement errors caused by the thermosiphon phenomenon can be corrected with greater accuracy than before. Furthermore, no hardware changes from conventional thermal flowmeters are required; correction of measurement errors caused by the thermosiphon phenomenon can be achieved simply by changing the software.

[0013] In order to correct the influence related to the size of the convection generated in the sensor flow path, and to associate the tilt influence with the voltages applied to the upstream side resistance component and the downstream side resistance component, so that the tilt influence contained in the sensor output can be easily corrected by calculation, the tilt influence estimator is constructed to infer the tilt influence based on the Nusselt number, Grashof number and Prandtl number of the fluid to be measured.

[0014] As a specific method of using the parameters for estimating the tilt effect, it can be mentioned that, when the Nusselt number is Nu, the Grashof number is Gr, the Prandtl number is Pr, the proportionality constant is A, and the exponent is n, the tilt effect estimator is configured based on Nu=A(Gr×Pr) n To calculate the tilt effect.

[0015] Regarding the Nusselt number, in order to obtain it by calculation instead of direct measurement, the tilt effect can be calculated. For example, when the upstream voltage output from the circuit including the upstream resistance component is set to Vu and the downstream voltage output from the circuit including the downstream resistance component is set to Vd, the sensor output generator is configured to output (Vu-Vd) / (Vu+Vd) as the sensor output, the tilt effect is the voltage difference (Vu0-Vd0) generated by convection in the sensor flow path, and the tilt effect inference unit is configured to infer the voltage difference (Vu0-Vd0) based on the value of the Nusselt number Nu, and the value of the Nusselt number Nu is based on the relationship between the Nusselt number Nu and the voltage difference (Vu0-Vd0) and Nu=A(Gr×Pr). n The flow rate calculator is configured to correct the tilt effect by subtracting a voltage difference (Vu0-Vd0) from a numerator (Vu-Vd) output by the sensor.

[0016] In order to accurately calculate the Nusselt number Nu from the Grashof number Gr and the Prandtl number Pr, which is suitable for the sensor flow path used in the thermal flowmeter, it is sufficient to use the exponent n calculated based on experiments. For example, the calculation formula Nu = A(Gr × Pr) can be used with the exponent n being 2. 2 That's it.

[0017] As a specific expression usable in the inclination influence inferrer, in a case where an inner diameter of the sensor flow path is set to L, a resistance value of the upstream-side resistance member or the downstream-side resistance member is set to R, a thermal conductivity of the fluid of the measurement target is set to λ, a constant-pressure molar specific heat of the fluid of the measurement target is set to Cp, a viscosity of the fluid of the measurement target is set to η, a density of the fluid of the measurement target is set to p, a gravitational acceleration is set to g, a volumetric expansion rate of the fluid of the measurement target is set to β, and a temperature difference between the upstream-side resistance member or the downstream-side resistance member and the fluid of the measurement target is set to ΔT, Nu = L x {((Vu0-Vd0) 2 / R) / (L 2 x ΔT)} / λ, Pr = Cpη / λ, Gr = pgl 3 β ΔT / η 2 .

[0018] As a specific structure example of the inclination influence inferrer, there can be cited one including: a pressure acquisition section that acquires a pressure P of the fluid of the measurement target; a temperature difference acquisition section that acquires a temperature difference ΔT between the upstream-side resistance member or the downstream-side resistance member and the fluid of the measurement target; a Nusselt number calculation section that, based on the acquired pressure P and temperature difference ΔT, calculates a Grashof number Gr and a Prandtl number Pr, and substitutes each value into Nu = A(Gr x Pr) n to calculate a value of the Nusselt number Nu; and a zero-point output calculation section that calculates a voltage difference (Vu0-Vd0) from the calculated value of the Nusselt number Nu.

[0019] If it is a flow control device including the thermal flow meter of the present application, a fluid control valve, and a valve controller that controls an opening degree of the fluid control valve based on a deviation between a set flow rate and a flow rate of the fluid of the measurement target output by the thermal flow meter, a corrected flow rate can be obtained based on a measurement error caused by the thermosiphon phenomenon, and accurate flow control can be achieved.

[0020] In order to keep the temperatures of the upstream-side resistance member and the downstream-side resistance member constant, and thereby calculate a flow rate from changes in the upstream-side voltage and the downstream-side voltage at that time, it is sufficient to include: an upstream-side thermostat control circuit having a bridge circuit including the upstream-side resistance member; and a downstream-side thermostat control circuit having a bridge circuit including the downstream-side resistance member. If this is the case, an error in the zero-point output caused by inclination influence can be appropriately corrected by the correction method.

[0021] As another way of the thermal flow meter to which the correction method of the present application can be applied, there can be cited one including: a bridge circuit including the upstream-side resistance member and the downstream-side resistance member; and a constant-current circuit that supplies a constant current to the bridge circuit.

[0022] For an existing thermal flow meter, in order to enjoy the same effect as the thermal flow meter of the present invention only by updating the program, it is sufficient to use the following recording medium in which a program for the thermal flow meter is recorded, the program for the thermal flow meter being a program used in the thermal flow meter, the thermal flow meter comprising: a sensor flow path for the flow of a fluid to be measured; an upstream side resistance component arranged in the sensor flow path; and a downstream side resistance component arranged in the sensor flow path at a position further downstream than the upstream side resistance component, the program for the thermal flow meter enabling a computer to perform the functions of the following components: a sensor output generator for generating a sensor output corresponding to the flow rate of the fluid to be measured based on an upstream side voltage output according to a change in the upstream side resistance component and a downstream side voltage output according to a change in the downstream side resistance component; a tilt effect inference device for inferring a tilt effect generated in the sensor output according to the posture of the sensor flow path based on at least the Prandtl number of the fluid to be measured; and a flow rate calculator for correcting the tilt effect based on the sensor output to calculate the flow rate of the fluid to be measured.

[0023] In addition, the thermal flow meter program can be transmitted electronically or recorded on a program recording medium such as a CD, DVD, or flash memory.

[0024] As described above, the thermal flowmeter of the present invention estimates the effects of tilt based on the Prandtl number of the fluid being measured. This allows for consideration of differences in thermal conductivity and pressure effects on the fluid that are affected by tilt. Consequently, flow measurement errors caused by the thermosiphon phenomenon can be corrected with greater accuracy than ever before. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic perspective view of a mass flow controller including a thermal flow meter according to one embodiment of the present invention.

[0026] Figure 2 is a schematic diagram of a mass flow controller according to the same embodiment.

[0027] Figure 3 This is a schematic diagram showing a sensing mechanism of the thermal flow meter according to the same embodiment.

[0028] Figure 4 This is a functional block diagram of a thermal flowmeter and a mass flow controller according to the same embodiment.

[0029] Figure 5 This is a conceptual diagram showing correction of the influence of tilt on sensor output according to the same embodiment.

[0030] Figure 6Measurement results showing the relationship between the Nusselt number, the Grashof number, and the Prandtl number of the same embodiment.

[0031] Figure 7 Measurement results showing the relationship between the Grashof number and the Nusselt number only.

[0032] Figure 8 is an experimental result when correcting the zero point output based on the Nusselt number calculated from the Prandtl number and the Grashof number of the same embodiment.

[0033] Figure 9 is an experimental result when correcting the zero point output based on the Nusselt number calculated from the Grashof number only without using the Prandtl number.

[0034] Figure 10 is a schematic view showing the installation orientation of a mass flow controller.

[0035] BRIEF DESCRIPTION OF DRAWINGS

[0036] 200 flow control device (mass flow controller)

[0037] IN inlet

[0038] 1 block

[0039] V control valve

[0040] C control device

[0041] 2 main flow path

[0042] 3 shunt assembly

[0043] 100 thermal flowmeter

[0044] SP flow detection circuit

[0045] Ru upstream-side resistance assembly

[0046] Rd downstream-side resistance assembly

[0047] 4 sensor flow path

[0048] 5 sensor output generator

[0049] 6 tilt influence inferrer

[0050] 7 flow calculator DETAILED DESCRIPTION

[0051] The thermal flowmeter 100 of the present embodiment and the flow control device 200 including the same are used to supply a plurality of various kinds of gases including a component gas such as SF6 at a set flow rate in a vacuum chamber, for example, in a semiconductor manufacturing process.

[0052] like Figure 1 As shown, the flow control device 200 is generally in the shape of a thin rectangular parallelepiped and is connected to a pipeline for the flow of component gases. Figure 2 As shown, flow control device 200 includes: a block 1 connected to a pipeline for gas flow, with a main flow path 2 forming part of the pipeline as an internal flow path; a thermal flow meter 100 mounted on the component mounting surface of block 1; a control valve V mounted downstream of thermal flow meter 100; and a control device C responsible for controlling at least control valve V. Specifically, flow control device 200 is a so-called mass flow controller, in which the block 1, thermal flow meter 100, control valve V, and control device C, which are components required for flow control, are packaged together.

[0053] Here, the flow control device 200 is as follows Figure 2 As shown, the block 1 is roughly in the shape of a long rectangular parallelepiped, and a main flow path 2 is formed along its long side. The flow control device 200 is designed to be installed in a manner such that the long side direction of the block 1 is aligned with the horizontal direction as a reference posture. In other words, Figure 10 As shown in (b), when the flow control device 200 is placed vertically so that the gas in the main flow path 2 flows vertically, an error due to the thermosiphon phenomenon will occur in the zero-point output of the flow rate output by the thermal flowmeter 100. The thermal flowmeter 100 of this embodiment includes a structure for correcting the error in the zero-point output caused by the thermosiphon phenomenon, that is, the influence of tilt.

[0054] The control device C is a so-called computer that includes a CPU, memory, an A / D converter, a D / A converter, and various input / output components. It is configured to execute a program stored in the memory, causing the various devices to operate in coordination. This allows the calculation unit CAL of the thermal flowmeter 100 to function, as well as the valve controller 9 that controls the opening of the control valve. The valve controller 9 performs feedback control on the opening of the control valve V to minimize the deviation between the measured flow rate output from the calculation unit CAL of the thermal flowmeter and the set flow rate set by the user.

[0055] Next, the thermal flowmeter 100 will be described in detail.

[0056] like Figure 2 As shown, the thermal flow meter 100 includes: a roughly U-shaped sensor flow path 4, which branches off from the main flow path 2 of the self-supplied gas flow and merges with the main flow path 2 again at a confluence point further downstream than the branch point; a flow detection circuit SP, which detects the flow rate of the gas; and a diversion component 3 as a resistance body, which is arranged between the branch point and the confluence point in the main flow path 2.

[0057] The flow diversion component 3 diverts the flow at a predetermined diversion ratio between the main flow path 2 and the sensor flow path 4, and includes resistance components such as a bypass component having a constant flow characteristic. The flow diversion component 3 can be formed by inserting multiple thin tubes into an outer tube, or by laminating multiple thin circular plates with multiple through-holes.

[0058] like Figure 2 As shown, the sensor flow path 4 is formed as a generally U-shaped capillary tube made of metal (e.g., stainless steel). A portion of the sensor flow path 4, located parallel to the flow direction of the main flow path 2, i.e., the longitudinal direction of the block 1, contains a portion of a flow detection circuit SP for detecting the gas flow rate. The flow detection circuit SP detects the gas flow rate by utilizing heat transfer caused by the flow of gas diverted to the sensor flow path 4.

[0059] like Figure 3 As shown, the flow detection circuit SP is a heating resistor whose resistance value increases and decreases with changes in temperature. It includes an upstream resistor element Ru, a coil wound around the outer circumference of the capillary tube forming the sensor flow path 4, and a downstream resistor element Rd, a coil wound downstream of the upstream resistor element Ru in the sensor flow path 4. Here, the upstream resistor element and the downstream resistor element serve both as a heater and a temperature sensor.

[0060] Moreover, the flow detection circuit SP is a circuit of constant temperature drive mode, such as Figure 3 As shown, the upstream constant temperature control circuit CTu is formed by a bridge circuit in which the upstream resistance element Ru is used as a part, and the downstream constant temperature control circuit CTd is formed by a bridge circuit in which the downstream resistance element Rd is used as a part.

[0061] The upstream constant temperature control circuit CTu includes: an upstream bridge circuit, which connects in parallel a series resistor group including the upstream resistance element Ru and the temperature setting resistor R1 connected in series with the upstream resistance element Ru, and a series resistor group including two fixed resistors R2 and R3 connected in series; and a feedback control circuit, which includes an operational amplifier, which feeds back the potential of the connection point between the upstream resistance element Ru and the temperature setting resistor R1 and the difference (Vu) in the potential at the connection point of the two fixed resistors to the upstream bridge circuit to maintain the balance of the upstream bridge circuit.

[0062] The downstream constant temperature control circuit CTd, like the upstream constant temperature control circuit CTu, includes: a downstream bridge circuit, which connects in parallel a series resistor group including a downstream resistance component Rd and a temperature setting resistor R1 connected in series with the downstream resistance component Rd, and a series resistor group including two fixed resistors R2 and R3 connected in series; and a feedback control circuit, which includes an operational amplifier, which feeds back the difference (Vd) between the potential of the connection point between the downstream resistance component Rd and the temperature setting resistor R1 and the potential of the connection point between the two fixed resistors to the downstream bridge circuit to maintain the balance of the downstream bridge circuit.

[0063] Here, the upstream resistor element Ru and the downstream resistor element Rd are made of materials with the same temperature coefficient of resistance. Furthermore, the upstream resistor element Ru and the downstream resistor element Rd are feedback-controlled by respective feedback control circuits so as to have the same resistance value as the temperature-setting resistor R1. That is, since the resistance value remains constant, the voltages Vu and Vd can be controlled so that the temperatures of the upstream resistor element Ru and the downstream resistor element Rd are also maintained constant. In this embodiment, Vu and Vd can be defined as the voltages applied to generate heat in the upstream resistor element Ru and the downstream resistor element Rd, namely, the upstream voltage Vu and the downstream voltage Vd.

[0064] like Figure 4 As shown, the thermal flowmeter 100 further includes the arithmetic unit CAL for calculating the gas flow rate based on the upstream voltage Vu and the downstream voltage Vd output from the flow detection circuit SP. The arithmetic unit CAL is configured to use (Vu-Vd) / (Vu+Vd) as the sensor output and to correct for errors in the zero-point output caused by the thermal siphon phenomenon, which occurs when the flow control device 200 is placed vertically, for example.

[0065] Here, the outline of the correction function based on the zero-point output of the operator CAL is as follows: Figure 5 As shown. When the flow control device 200 is placed vertically and the gas inlet IN is facing downward, even when there is no outflow or inflow of gas to the flow control device 200, the convection of gas generated in the sensor flow path 4 causes the gas to flow from the upstream side resistor component Ru to the downstream side resistor component Rd. Therefore, when the gas inlet IN is facing downward, a positive error is generated in the sensor output before correction. In addition, the higher the pressure of the gas sealed in the flow control device 200, the greater the error in the zero-point output in the sensor output before correction. Similarly, when the inlet IN of the flow control device 200 is facing upward, the flow from the downstream side resistor component Rd to the upstream side resistor component Ru due to convection is detected, so the sensor output before correction becomes a negative value and an error is generated. In this case, the higher the pressure of the gas sealed in the flow control device 200, the greater the error in the zero-point output.

[0066] The calculation unit CAL is configured to estimate the error in the zero-point output in each state, that is, the influence of the tilt, and correct the tilt influence based on the sensor output before correction, thereby approaching the actual flow rate.

[0067] Below, while referring to Figure 4 , while explaining the detailed structure of the operator CAL.

[0068] The calculation unit CAL functions as at least a sensor output generator 5 , a tilt influence estimator 6 , a flow rate calculator 7 , and a reception unit 8 .

[0069] The sensor output generator 5 is configured to input the upstream voltage Vu and downstream voltage Vd, the voltages applied to the upstream resistor element Ru and the downstream resistor element Rd, and calculate and output (Vu - Vd) / (Vu + Vd) as the sensor output. The voltage difference (Vu - Vd) varies depending on the flow rate of the gas flowing through the sensor flow path 4, while the voltage sum (Vu + Vd) corresponds to an indicator of the temperature of the gas flowing through the sensor flow path 4. By dividing the voltage difference by the voltage sum, the sensor output is corrected for the temperature effect on the flow rate.

[0070] The tilt effect estimator 6 estimates the tilt effect on the sensor output caused by the posture of the sensor flow path 4 based on at least the Prandtl number of the gas being measured. In this embodiment, the tilt effect estimator 6 estimates the tilt effect based not only on the Prandtl number but also on the Nusselt number and Grashof number of the gas. Furthermore, the tilt effect estimator 6 receives inputs of the gas pressure and temperature obtained by various sensors installed within the flow control device 200 or various sensors installed separately in the semiconductor manufacturing process, and outputs the tilt effect based on these values.

[0071] The tilt effect estimator 6 calculates the Nusselt number from the Prandtl number and the Grashof number based on the gas temperature, pressure, and various physical properties. Furthermore, based on the relationship between the Nusselt number and the voltage difference (Vu-Vd), which constitutes part of the sensor output, the voltage difference (Vu0-Vd0) in the no-flow state is estimated as the tilt effect. To achieve this function, the tilt effect estimator 6 includes at least a temperature acquisition unit 61, a pressure acquisition unit 62, a physical property value storage unit 63, a Nusselt number calculation unit 64, and a zero-point output calculation unit 65.

[0072] The temperature acquisition unit 61 acquires, for example, an output signal from a temperature sensor (not shown) provided on the block 1 of the flow control device 200 as the temperature of the gas, and outputs the temperature to the Nusselt number calculation unit 64. Furthermore, the temperature acquisition unit 61 may also acquire information from other temperature sensors provided on the pipeline connected to the flow control device 200.

[0073] The pressure acquisition unit 62 acquires an output signal from a pressure sensor (not shown) that measures the pressure of the gas within the main flow path of the flow control device 200. The pressure sensor may be provided in the flow control device 200 itself to measure the pressure of the gas flowing in the main flow path, or may be provided in the flow path connecting the flow control device 200 to on-off valves (not shown) provided upstream and downstream of the flow control device 200. The pressure acquisition unit 62 outputs the pressure acquired, for example, when the on-off valves are closed and there is no gas flowing into or out of the flow control device 200, to the Nusselt number calculation unit 64 as the pressure of the gas sealed within the flow control device 200.

[0074] When Nusselt number is Nu, Grashof number is Gr, Prandtl number is Pr, proportionality constant is A, and exponent is n, the Nusselt number calculation unit 64 calculates the value based on Nu=A(Gr×Pr) n To calculate the value of Nusselt number Nu. Here, in this embodiment, the proportional constant A = 1 and the exponent n = 2. The reason is that the inventors of this application found that according to Figure 6 In the experimental results shown, when the above values ​​are used, the Nusselt number Nu can be calculated from the product of the Grashof number Gr and the Prandtl number Pr.

[0075] Here, based on Figure 7 The measurement results shown in the figure illustrate the estimation accuracy of the Nusselt number Nu when the Prandtl number Pr is not used. Figure 6 and Figure 7 In order to make comparisons between different graphs, the intervals between the reference lines of each axis should be roughly the same. Figure 7 As shown in FIG. 1 , although there is a correlation between the Nusselt number Nu and the Grashof number Gr, the deviation of the Nusselt number Nu from the Grashof number Gr becomes larger in the region where the Grashof number Gr is small. In other words, even if an approximate straight line is calculated between the Grashof number Gr and the Nusselt number Nu, and the Nusselt number Nu is estimated from the Grashof number Gr based on the approximate straight line, it is still different from the method of calculating (Gr×Pr) as in the present embodiment. 2 Compared with the case of estimating the Nusselt number Nu, the inference accuracy will also be greatly degraded. Figure 6 and Figure 7The comparison results show that the Nusselt number Nu that is suitable for the structure of the thermal flowmeter 100 of this embodiment is calculated using (Gr×Pr) 2 method.

[0076] The calculation method of the Nusselt number Nu in this embodiment will be described in further detail. When the inner diameter of the sensor flow path 4 is L, the resistance value of the upstream resistor element Ru or the downstream resistor element Rd is R, the thermal conductivity of the fluid being measured is λ, the constant-pressure Mohr heat of the fluid being measured is Cp, the viscosity of the fluid being measured is η, the density of the fluid being measured is ρ, the gravitational acceleration is g, the volumetric expansion coefficient of the fluid being measured is β, and the temperature difference between the upstream resistor element Ru or the downstream resistor element Rd and the fluid being measured is ΔT, the Grashof number Gr and the Prandtl number Pr are respectively as follows.

[0077] Pr=Cpη / λ

[0078] Gr=ρgL 3 βΔT / η 2

[0079] The Nusselt number calculation unit 64 reads information such as the constant-pressure molar specific heat (Cp), the volume expansion coefficient (β), and the density (ρ) stored in the physical property value storage unit 63 based on information such as the gas type received from the user by the reception unit 8 or the pressure and temperature acquired by the pressure acquisition unit 62 and the temperature acquisition unit 61. The Nusselt number calculation unit 64 then substitutes the read physical property values ​​and the acquired pressure and temperature into the calculation formulas for the Grashof number Gr and the Prandtl number Pr to calculate the respective values. Finally, the Nusselt number calculation unit 64 calculates the square of the product of the Grashof number Gr and the Prandtl number Pr as the Nusselt number Nu. The calculated Nusselt number Nu is output to the zero-point output calculation unit 65.

[0080] The zero-point output calculation unit 65 calculates the tilt effect based on the relationship between the voltage difference (Vu0-Vd0) in the no-flow state and the Nusselt number Nu. Specifically, when the inner diameter of the sensor flow path 4 is L, the resistance value of the upstream resistor Ru or the downstream resistor Rd is R, the thermal conductivity of the fluid to be measured is λ, and the temperature difference between the upstream resistor Ru or the downstream resistor Rd and the fluid to be measured is ΔT, the ...

[0081] Nu=L×{((Vu0-Vd0) 2 / R) / (L 2 ×ΔT)} / λ

[0082] The zero point output calculating section 65 calculates the voltage difference (Vu0-Vd0). Here, the sign of (Vu0-Vd0) is positive when the inlet of the gas in the flow rate control device 200 is on the lower side and is negative when the inlet of the gas is on the upper side.

[0083] The flow rate calculator 7 corrects the inclination influence inferred by the inclination influence inferring section 6 with respect to the sensor output (Vu-Vd) / (Vu+Vd) before correction output from the sensor output generator 5, and calculates the flow rate of the gas based on the corrected sensor output. That is, the flow rate calculator 7 corrects the offset of the zero point output by subtracting (Vu0-Vd0) as the inclination influence from the voltage difference (Vu-Vd) before correction, and calculates the flow rate by substituting the corrected sensor output {(Vu-Vd)-(Vu0-Vd0)} / (Vu+Vd) into a prescribed flow rate calculation function. More specifically, if the flow rate is set to F and the flow rate calculation function is set to Sens(X), the flow rate is converted to F = Sens({(Vu-Vd)-(Vu0-Vd0)} / (Vu+Vd)).

[0084] According to the thermal flowmeter 100 and the flow rate control device 200 thus configured, as shown in the graph of Figure 8 , the inclination influence exhibited by the sensor output is corrected well in precision, and an accurate flow rate is obtained. Here, Figure 8 , the measured results shown in Figure 8 show two types of measurement results when the flow inlet IN is upward or downward for a plurality of gas types. As shown in (a) of Figure 8 , depending on the gas type, the higher the seal pressure in the flow rate control device 200, the more significantly the offset amount of the zero point output increases, but, as shown in (b) of , by using the correction method of the present embodiment, the offset amount of the zero point output can be greatly reduced regardless of the gas type and the seal pressure. It is considered that the reason for this is that the inclination influence inferring section 6 infers the inclination influence based on a value affected by the pressure or the thermal conductivity of the fluid, that is, the Prandtl number, and thus not only the influence of the magnitude of the convection caused by the thermal siphon phenomenon itself on the zero point output but also the influence of the difference in the ease of heat transfer caused by the gas type on the zero point output can be corrected.

[0085] Figure 7 Here, as a comparative example, as shown in Figure 9 , the Nusselt number Nu is calculated using only the Grashof number Gr without using the Prandtl number Pr, and the result after correcting the zero point output is shown in Figure 8 . In addition, with respect to the graphs of Figure 9 , in order to be able to compare each axis, the intervals of the reference lines are made uniform in substantially the same unit amount. With Figure 9Compared with the zero-point output without any correction shown in (a), Figure 9 As shown in (b), even when only the Grashof number Gr is used, the zero point output can be corrected to some extent. Figure 8 (b) and Figure 9 As can be seen from the comparison with (b) of FIG1 , it can be confirmed that when the Prandtl number Pr is also used as in this embodiment, the calibration accuracy is improved, especially in the region where the sealed pressure is high. Thus, the thermal flowmeter 100 of this embodiment can calibrate the zero-point output with high accuracy regardless of the type of gas or the pressure of the sealed gas.

[0086] In addition, the tilt effect estimator 6 can calculate the magnitude of the tilt effect based on the pressure and temperature of the gas and the various physical properties of the gas. If information related to the installation direction of the flow control device 200 is set, the positive or negative value displayed as the zero point output can also be determined. Therefore, there is no need to use additional sensors such as gyroscope sensors that are not usually used in the flow control device 200.

[0087] That is, the measurement error caused by the thermosiphon phenomenon can be accurately corrected only by changing the software without changing the hardware from a normal thermal flow meter.

[0088] Other embodiments will be described.

[0089] The configuration of the tilt effect estimator is not limited to that described in the aforementioned embodiment. Specifically, the tilt effect estimator can estimate the tilt effect based on at least the Prandtl number of the fluid. For example, the tilt effect estimator can estimate the tilt effect based on the relationship between the Prandtl number and the voltage difference (Vu0 - Vd0) representing the error in the zero-point output. Alternatively, the tilt effect estimator can be configured to estimate the tilt effect based on the relationship between the Nusselt number and the Prandtl number, rather than the Grashof number.

[0090] The method of expressing the influence of tilt is not limited to the voltage difference (Vu0-Vd0). For example, it is also possible to calculate the zero-point outputs Vu0 and Vd0 separately so that the upstream voltage Vu and the downstream voltage Vd can be corrected separately. In this case, the temperature index Vd+Vu can also be corrected.

[0091] The user pre-sets the installation orientation of the flow control device or the type of fluid (gas) being measured through the receiving unit, but the flow control device can also automatically obtain this information. For example, the flow control device can also include a gyro sensor to obtain the orientation of the fluid inlet or the posture of the sensor flow path, thereby automatically setting the positive or negative value of the tilt effect. In addition, the correction amount for the tilt effect can be changed according to the tilt angle. In addition, since the thermal conductivity of the flowing fluid can be inferred based on the temperature index Vu+Vd, it can also be used to identify the type of fluid based on this value and obtain other required physical properties.

[0092] The calibration method of the thermal flowmeter of the present invention is not limited to a constant temperature drive method and can also be applied to a constant current drive method or other methods. For example, a thermal flowmeter using a constant current drive method can be used as long as the flow detection circuit includes a bridge circuit including the upstream resistor element and the downstream resistor element, and a constant current circuit that supplies a constant current to the bridge circuit.

[0093] Furthermore, various modifications and combinations of the embodiments are possible without departing from the spirit of the present invention.

[0094] Industrial Applicability

[0095] According to the present invention, a thermal flowmeter can be provided that takes into account the difference in thermal conductivity of the fluid and the influence of pressure with respect to the influence of tilt, and can correct the flow measurement error caused by the thermosiphon phenomenon with higher accuracy than before.

Claims

1. A thermal flow meter, characterized in that include: A sensor flow path for the flow of the fluid to be measured; an upstream resistance component, disposed in the sensor flow path; a downstream resistor component disposed in the sensor flow path at a downstream side relative to the upstream resistor component; a sensor output generator that generates a sensor output corresponding to a flow rate of a fluid to be measured based on a voltage output from a flow rate detection circuit including the upstream-side resistor element and the downstream-side resistor element; a tilt effect estimator for estimating a tilt effect on the sensor output caused by the posture of the sensor flow path based on a relationship between a Nusselt number calculated from a Prandtl number of a fluid to be measured and a voltage difference caused by convection in the sensor flow path, a relationship between the Prandtl number of the fluid to be measured and the voltage difference, or a relationship between the Nusselt number of the fluid to be measured and the Prandtl number; as well as The flow rate calculator corrects the tilt effect based on the sensor output to calculate the flow rate of the fluid to be measured.

2. The thermal flow meter according to claim 1, characterized in that The tilt influence estimator is configured to estimate the tilt influence based on the Nusselt number, the Grashof number, and the Prandtl number of the fluid to be measured.

3. The thermal flow meter according to claim 2, characterized in that When Nusselt number is Nu, Grashof number is Gr, Prandtl number is Pr, proportionality constant is A, and exponent is n, the tilt influence estimator is configured based on Nu=A(Gr×Pr) n To calculate the tilt effect.

4. The thermal flow meter according to claim 3, characterized in that When the upstream voltage output from the circuit including the upstream resistance component is set to Vu, the downstream voltage output from the circuit including the downstream resistance component is set to Vd, the zero point output of the upstream voltage is set to Vu0, and the zero point output of the downstream voltage is set to Vd0, the sensor output generator is configured to output (Vu-Vd) / (Vu+Vd) as the sensor output, The tilt effect is the voltage difference (Vu0-Vd0) generated by convection in the sensor flow path, and the tilt effect estimator is configured to estimate the voltage difference (Vu0-Vd0) based on the value of the Nusselt number Nu, and the value of the Nusselt number Nu is based on the relationship between the Nusselt number Nu and the voltage difference (Vu0-Vd0) and Nu=A(Gr×Pr) n Calculate, The flow rate calculator is configured to correct the tilt effect by subtracting a voltage difference (Vu0-Vd0) from a numerator (Vu-Vd) output by the sensor.

5. The thermal flow meter according to claim 4, characterized in that The exponent n is 2.

6. The thermal flow meter according to claim 4, characterized in that When the inner diameter of the sensor flow path is L, the resistance value of the upstream resistance element or the downstream resistance element is R, the thermal conductivity of the fluid to be measured is λ, the constant-pressure Mohr heat of the fluid to be measured is Cp, the viscosity of the fluid to be measured is η, the density of the fluid to be measured is ρ, the gravitational acceleration is g, the volume expansion coefficient of the fluid to be measured is β, and the temperature difference between the upstream resistance element or the downstream resistance element and the fluid to be measured is ΔT, Nu=L×{((Vu0-Vd0) 2 / R) / (L 2 ×ΔT)} / λ Pr=Cpη / λ Gr=ρgL 3 bΔT / h 2 。 7. The thermal flow meter according to any one of claims 4 to 6, characterized in that: The tilt impact inferrer includes: A pressure acquisition unit that acquires a pressure P of a fluid to be measured; a temperature difference acquiring unit for acquiring a temperature difference ΔT between the upstream resistance element or the downstream resistance element and a fluid to be measured; The Nusselt number calculation unit calculates the Grashof number Gr and the Prandtl number Pr based on the obtained pressure P and temperature difference ΔT, and substitutes each value into Nu = A (Gr × Pr) n to calculate the value of the Nusselt number Nu; and The zero-point output calculation unit calculates the voltage difference (Vu0-Vd0) based on the calculated Nusselt number Nu.

8. The thermal flow meter according to any one of claims 1 to 6, characterized in that: The flow detection circuit comprises: an upstream-side constant temperature control circuit having a bridge circuit including the upstream-side resistance component; and The downstream constant temperature control circuit has a bridge circuit including the downstream resistance component.

9. The thermal flow meter according to any one of claims 1 to 6, characterized in that: The flow detection circuit comprises: a bridge circuit comprising the upstream-side resistance component and the downstream-side resistance component; and The constant current circuit supplies a constant current to the bridge circuit.

10. A flow control device, characterized in that include: The thermal flow meter according to any one of claims 1 to 9; Fluid control valves; as well as The valve controller controls the opening of the fluid control valve based on a deviation between a set flow rate and a flow rate of the fluid to be measured outputted by the thermal flow meter.

11. A thermal flow measurement method using a thermal flow meter, the thermal flow meter comprising: A sensor flow path for the flow of a fluid to be measured; an upstream resistor component disposed in the sensor flow path; and a downstream resistance element provided in the sensor flow path at a position downstream of the upstream resistance element, The thermal flow measurement method is characterized by comprising: generating a sensor output corresponding to the flow rate of the fluid to be measured based on a voltage output from a flow rate detection circuit including the upstream-side resistor element and the downstream-side resistor element; inferring an effect of tilt on the sensor output caused by the posture of the sensor flow path based on a relationship between a Nusselt number calculated from a Prandtl number of a fluid to be measured and a voltage difference caused by convection in the sensor flow path, a relationship between the Prandtl number of the fluid to be measured and the voltage difference, or a relationship between the Nusselt number of the fluid to be measured and the Prandtl number; and The influence of the tilt is corrected based on the sensor output, and the flow rate of the fluid to be measured is calculated.

12. A recording medium, characterized in that A program for a thermal flowmeter is recorded. The thermal flowmeter includes: a sensor flow path through which a fluid to be measured flows; an upstream resistor element disposed in the sensor flow path; and a downstream resistor element disposed in the sensor flow path downstream of the upstream resistor element. The program for the thermal flowmeter causes a computer to function as the following components: a sensor output generator that generates a sensor output corresponding to a flow rate of a fluid to be measured based on a voltage output from a flow rate detection circuit including the upstream-side resistor element and the downstream-side resistor element; a tilt effect estimator that estimates a tilt effect on the sensor output caused by the posture of the sensor flow path based on a relationship between a Nusselt number calculated from a Prandtl number of a fluid to be measured and a voltage difference caused by convection in the sensor flow path, a relationship between the Prandtl number of the fluid to be measured and the voltage difference, or a relationship between the Nusselt number of the fluid to be measured and the Prandtl number; and The flow rate calculator corrects the tilt effect based on the sensor output to calculate the flow rate of the fluid to be measured.

Images