JP2010185777A - Flow sensor and flow correction method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a flow sensor which can reduce map data for use in the correction of output and simplify calculation. <P>SOLUTION: A corrected value Va is determined by calculating Va=V-(VT2-VT1)×(T-T1)/(T2-T1), where V is a measured value, Va a corrected value obtained by correcting the measured value V, T a measured temperature (an ambient temperature), T1 a first standard temperature (fixed), T2 a second standard temperature (fixed), VT1 a measured value (fixed) at the first standard temperature, and VT2 a measured value (fixed) at the second standard temperature. Since it is a linear function, calculation can be simplified. Since the measured values V (measurement points) for creating map data are small in number, the map data can be reduced. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a flow rate sensor for detecting a flow rate of gas and a method for correcting the flow rate, and is used, for example, for detection of a gas flow rate in a semiconductor manufacturing apparatus and detection of an intake air amount of an automobile engine.

2. Description of the Related Art Conventionally, as this type of flow sensor, one that detects a gas flow rate based on a heat distribution of a heating resistor disposed in a gas flow path is known. Since the output characteristics of the flow rate sensor are non-linear (curved) characteristics, the correction accuracy cannot be increased by the method of linearly correcting the output.
Therefore, as a method for correcting the output of the flow sensor, non-linear correction is performed using a higher-order arithmetic expression of the third or higher order, and the correction coefficient determined after correction is stored in the memory as map data, thereby corresponding to the output characteristics. A device that performs the correction is proposed (for example, Patent Document 1).

JP 2007-71889 A (23rd and 40th paragraphs, FIGS. 1 and 3).

  However, since the conventional flow rate sensor described above uses a higher-order arithmetic expression of the third or higher order, a large amount of map data is required to increase the output correction accuracy. For this reason, there exists a problem that the effort for creating map data increases. Moreover, since the size of the memory for storing the map data is increased, there is a problem that the flow sensor is increased in size. Further, since the output is corrected using a high-dimensional arithmetic expression, there is a problem that the calculation becomes complicated.

  Accordingly, the present invention has been made to solve the above-described problems, and an object thereof is to realize a flow sensor capable of reducing map data used for output correction and simplifying calculation.

  In order to achieve the above object, according to the present invention, the invention according to claim 1 is a sensor unit (20, 31) for detecting a gas flow rate, and an analog voltage signal output from the sensor unit is converted into a digital value. A flow rate sensor (10) including a conversion unit (35) for conversion, wherein a digital output value output from the conversion unit is V, an ambient temperature is T, a constant is A, and the ambient temperature is a first reference temperature. The absolute value of the difference between the output value at the time and the output value when the ambient temperature is a second reference temperature different from the first reference temperature is B, and the difference between the ambient temperature and the first reference temperature is Provided is a correction circuit (42) for calculating Va = V− (B × C / A), where C is an absolute value and Va is a value obtained by correcting the output value according to an error caused by the ambient temperature. It is characterized by that.

  According to a second aspect of the present invention, in the flow rate sensor (10) according to the first aspect, the constant A is an absolute value of a difference between the first and second reference temperatures.

  According to a third aspect of the present invention, in the flow sensor (10) according to the first or second aspect, the converter (35) converts the analog voltage signal into a binary digital value A. A plurality of inverting circuits (36a, 36b) that invert and output an input signal and whose inverting operation time varies depending on the power supply voltage, and one of the inverting circuits performs an inverting operation. It is configured as a starting inversion circuit (36a) that can be controlled from the outside, and a pulse running circuit (36) that runs a pulse signal when the starting inversion circuit starts to operate, and each inversion circuit in the pulse running circuit And a voltage signal input terminal (35a) for applying the voltage signal as a power supply voltage for each inverting circuit, and generating a plurality of reference voltage signals having mutually different voltages. A reference voltage signal generation circuit (32), and an input voltage switch for switching and inputting a voltage signal for A / D conversion and a plurality of reference voltage signals generated by the reference voltage signal generation circuit to the voltage signal input terminal Means (33) and travel position detection means (38) for detecting the travel position of the pulse signal in the pulse travel circuit based on the output signal from each inverting circuit and generating data corresponding to the travel position And a control means (34) for activating the driving operation of the pulse traveling circuit by operating the inversion circuit for starting, and operating the traveling position detecting means when a predetermined time has passed thereafter, and the traveling position detecting means A data output line (L2) for outputting digital data including the data as an A / D conversion result, and the voltage signal for A / D conversion output from the data output line and Of the digital data of the plurality of reference voltage signals, the first difference that is the difference between any two of the digital data and the second difference that is the difference between the two digital data that is different from the combination of the digital data for obtaining the first difference. And a signal processing circuit (41) for calculating a digital value corresponding to the flow rate based on a ratio to the difference.

  In invention of Claim 4, the sensor part (20, 31) which detects the flow volume of gas, The conversion part (35) which converts the analog voltage signal output from the said sensor part into a digital output value, In the flow rate correction method for correcting the output value of the flow rate sensor (10) according to the error caused by the ambient temperature, the output value is V, the ambient temperature is T, the constant is A, and the ambient temperature is the first reference. The absolute value of the difference between the output value at the time of temperature and the output value when the ambient temperature is a second reference temperature different from the first reference temperature is B, and the ambient temperature and the first reference temperature are Va = V− (B × C / A) is calculated, where C is the absolute value of the difference and Va is the value obtained by correcting the output value according to the error caused by the ambient temperature. .

  According to a fifth aspect of the present invention, in the flow rate correction method according to the fourth aspect, the constant A is an absolute value of a difference between the first and second reference temperatures.

  In addition, the code | symbol in each said parenthesis shows the correspondence with the specific means as described in embodiment mentioned later.

(Effect of the invention according to claims 1 and 4)
A digital value output from the conversion unit that converts an analog voltage signal output from the sensor unit into a digital value is V, an ambient temperature is T, a constant is A, a digital value when the ambient temperature is the first reference temperature, and The absolute value of the difference from the digital value when the ambient temperature is a second reference temperature different from the first reference temperature is B, the absolute value of the difference between the ambient temperature and the first reference temperature is C, and the digital output from the conversion unit When a value obtained by correcting the value according to an error caused by the ambient temperature is Va, the digital value can be corrected using a one-dimensional arithmetic expression Va = V− (B × C / A).
Therefore, the map data can be reduced and the calculation can be simplified as compared with the conventional method in which the digital value is corrected using a higher-order arithmetic expression of the third or higher order.

(Effects of inventions according to claims 2 and 5)
Since the constant A is an absolute value of the difference between the first and second reference temperatures, the constant A can be easily obtained.

(Operation of the Invention of Claim 3)
In the pulse running circuit, since a plurality of inverting circuits are connected, for example, if the output of the starting inverting circuit is low level, the next inverting circuit output is high level, and the next inverting circuit output is low level. As described above, the output of each inverting circuit is sequentially inverted at a predetermined inverting operation time. In the pulse traveling circuit, the position of the inverting circuit whose output level is the same as that of the subsequent inverting circuit is the traveling position of the pulse signal.

On the other hand, since the voltage signal to be A / D converted is applied to the power supply line of each inverting circuit in the pulse traveling circuit as the power supply voltage of each inverting circuit via the voltage signal input terminal, The inversion operation time of each inversion circuit changes according to the voltage level.
Therefore, the travel time and travel position of the pulse signal in the pulse travel circuit are determined by the voltage signal input to the voltage signal input terminal.

  As described above, in the A / D conversion circuit, the travel position detecting means detects the travel position of the pulse signal in the pulse travel circuit based on the output signal from each inversion circuit, and according to the travel position. Data is generated, and the data output line outputs digital data including data from the traveling position detection means as an A / D conversion result. In addition, the operation timing of the start inversion circuit and the travel position detection means in the pulse travel circuit is controlled by the control means, and after the start of the travel operation of the pulse travel circuit by this control means, the travel position detection means causes the travel of the pulse signal. The time until the position is detected (sampling time) is set to a predetermined time.

(Effect of the invention according to claim 3)
In the A / D conversion circuit according to claim 3, by using a voltage signal to be A / D converted as a power supply voltage of an inverting circuit provided in the pulse traveling circuit, The voltage signal is converted into digital data by detecting the travel position of the pulse signal within the pulse travel circuit while the travel time of the pulse signal is changed by the voltage signal and the pulse travel circuit travels for a predetermined sampling time. To do.
Therefore, it is possible to digitize minute changes in the voltage signal without using an analog amplification circuit that amplifies the voltage signal as in the prior art, so that A / D conversion can be performed normally even when the ambient temperature is high. Can be done.

In addition, since an A / D conversion circuit including a pulse traveling circuit in which a plurality of inverting circuits are connected has a temperature characteristic, the gain and the offset vary depending on the ambient temperature.
However, the signal processing circuit included in the A / D conversion circuit according to claim 3 is any one of the voltage signal for A / D conversion output from the data output line and each digital data of the plurality of reference voltage signals. A digital value corresponding to the flow rate based on a ratio between a first difference that is a difference between two digital data and a second difference that is a difference between two digital data different from a combination of digital data for which the first difference is obtained. Is calculated.

At this time, since the offsets of the A / D conversion circuit included in the voltage signal for A / D conversion and the plurality of reference voltage signals are both the same, the first difference and the second difference are the A / D conversion circuit. The offset is the value that is deleted. Further, the ratio between the first difference and the second difference is calculated, whereby the fluctuation of the gain of the A / D conversion circuit due to the temperature can be removed.
Therefore, a temperature detection circuit is not required, and a highly accurate output value that is not affected by the ambient temperature can be obtained.

It is a perspective view which shows the main structures of the flow sensor which concerns on embodiment of this invention. 3 is a plan view showing a sensor chip 20 with a rear end portion omitted. FIG. It is AA arrow sectional drawing of FIG. 2 is a circuit diagram showing an electrical configuration of a circuit chip 30. FIG. 3 is a timing chart of each signal in an A / D conversion circuit 35. 4 is a diagram illustrating a relationship between an input voltage Vin and a digital value of the input voltage Vin converted by an A / D conversion circuit 35. FIG. It is a figure which shows the error at the time of correct | amending using said correction curve when carrying out A / D conversion of the voltage signal Vin. It is a figure which shows the relationship between the correction value and the flow volume which correct | amended based on the correction curve shown in FIG. It is a figure which shows a detection error when the output of the A / D conversion circuit 35 fluctuates + 2%. It is a figure which shows a detection error when the output of the A / D conversion circuit 35 fluctuates -2%. It is process drawing which shows the procedure of correction | amendment.

  An embodiment according to the present invention will be described with reference to the drawings. In the following embodiments, a flow sensor for detecting an air intake amount of an internal combustion engine such as an automobile engine will be described as a flow sensor according to the present invention.

[Configuration of flow sensor]
(Main composition)
The main configuration of the flow sensor according to this embodiment will be described with reference to the drawings. FIG. 1 is a perspective view showing the main configuration of the flow sensor according to this embodiment.

  The flow sensor 10 includes a housing 11. The housing 11 accommodates the sensor chip 20 and the circuit chip 30. The sensor chip 20 is configured using a semiconductor substrate 28 (FIG. 2), and the circuit chip 30 is mounted on a circuit substrate (for example, a ceramic substrate) 13. The circuit chip 30 is electrically connected to the sensor chip 20 via wiring (not shown) formed on the circuit board 13 and bonding wires 15.

  As described above, the sensor chip 20 and the circuit board 13 are formed of different materials, and are arranged separately and independently to achieve thermal insulation between the two. A membrane (thin wall portion) 26 is formed in the vicinity of the tip of the sensor chip 20. The membrane 26 is provided with a sensing unit 27 (FIGS. 2 and 3) for detecting the air flow rate.

  Openings 12 are formed on the side walls 16, 16 of the housing 11 that face each other. The direction of the white arrow indicated by reference numeral F1 indicates the direction in which air flows. Air that is a detection target of the flow sensor 10 flows in from one opening 12 and flows out from the other opening 12. The sensing unit 27 is disposed in the air flow path.

  The circuit chip 30 includes a sensor control circuit 31 that controls input / output of the sensing unit 27, an A / D conversion circuit 35 (FIG. 4) that converts an analog signal from the sensor control circuit 31 into a digital signal, and the like. Each of these circuits is configured by forming semiconductor elements such as MOS transistors and diodes, resistance elements, and wirings on a semiconductor substrate.

  The circuit chip 30 is electrically connected to the leads 14 via wiring (not shown) formed on the circuit board 13. The lead 14 is electrically connected to an ECU (Electronic Control Unit) mounted on the vehicle via wiring (not shown). The range from the circuit chip 30 to the bonding wire 15 is sealed with a sealing material (not shown) such as an epoxy resin.

(Configuration of sensor chip 20)
The configuration of the sensor chip 20 will be described with reference to the drawings. FIG. 2 is a plan view in which the rear end portion of the sensor chip 20 is omitted. 3 is a cross-sectional view taken along line AA in FIG.

  The sensor chip 20 is configured using a semiconductor substrate 28. As shown in FIG. 3, a cavity 25 is formed on the back surface of the semiconductor substrate 28. The bottom of the cavity 25 is a membrane 26. In the center of the membrane 26, heating resistors 21 and 22 constituting the sensing unit 27 are arranged in parallel. Resistance thermometers 23 and 24 that function as thermometers are arranged on both sides of the heating resistor.

  As shown in FIG. 2, the heating resistor 21 is electrically connected to the wiring layers 29e and 29d, and the heating resistor 22 is electrically connected to the wiring layers 29b and 29c. The resistance temperature detector 23 is electrically connected to the wiring layer 29f, and the resistance temperature detector 24 is electrically connected to the wiring layer 29a. Each of the wiring layers 29a to 29f is electrically connected to the bonding wire 15 (FIG. 1).

  The heating resistors 21 and 22 are electrically connected to the sensor control circuit 31 (FIG. 4) provided in the circuit chip 30 through the wiring layers 29 b to 29 e and the bonding wires 15. Further, the resistance temperature detectors 23 and 24 are electrically connected to the sensor control circuit 31 through the wiring layers 29 f and 29 a and the bonding wires 15.

  In this embodiment, an SOI (silicon on insulator) substrate is used as the semiconductor substrate 28. The heating resistors 21 and 22, the resistance temperature detectors 23 and 24, and the wiring layers 29 a to 29 f are formed by doping impurities into a semiconductor layer that is an SOI layer by ion implantation and etching it. The semiconductor layer is sealed with BPSG (Boron-doped Phospho-Silicate Glass) (not shown), and the BPSG is covered with a silicon nitride film (not shown). That is, the surface of the sensing unit 27 is protected by the insulating film made of BPSG and the silicon nitride film.

The cavity 25 is formed by etching the back surface of the silicon substrate that is the support substrate. Etching is performed until the insulating film is a buried layer of the SOI substrate (BO _X layer) is exposed. Since the membrane 26 is very thin compared to the silicon substrate, the heat capacity of the membrane 26 is kept low, and the membrane 26 ensures thermal insulation from the silicon substrate. In addition, since the sensing unit 27 is formed on the membrane 26, the sense of the sensing unit 27 is high. The thickness d of the membrane 26 is about 2 μm, for example.

[Operation]
Next, the operation of the flow sensor will be described.
A sensor control circuit 31 provided in the circuit chip 30 measures the environmental temperature based on changes in resistance values of the resistance temperature detectors 23 and 24. The heating resistors 21 and 22 are heated by the current flowing from the sensor control circuit 31.

  And as shown in FIG. 1, the heat of the heating resistors 21 and 22 is taken away by the flow of air. As shown in FIG. 3, when there is no air flow (μ = 0), the temperature distribution around the heating resistors 21 and 22 is symmetrical about the heating resistor, but when air flows (μ > 0), the temperature distribution changes. The sensor control circuit 31 adjusts the magnitude of the current flowing through the heating resistor so that the heating resistors 21 and 22 are always at a constant temperature.

  The change in the current value when the sensor control circuit 31 performs the above adjustment is output as a sensor signal Sf (FIG. 4) having a voltage change corresponding to the change, and is converted into a digital value by the A / D conversion circuit 35. Is output to the ECU. Then, the ECU calculates the air flow rate based on the input digital value. Further, the temperature distribution around the heating resistor varies depending on the direction of air flow. In other words, heat is taken away more intensely on the upstream side than on the downstream side of the air, and more current is required. From the difference between the two, the air flow direction can be detected simultaneously with the air flow rate.

For example, it is assumed that air flows from the direction indicated by F1 in FIG. 3 (forward flow). Here, since the heating resistor 21 located upstream of the air is deprived of heat more than the heating resistor 22 located downstream, the control circuit keeps the temperature (resistance value) of the heating resistor 21 constant. Finally, the energization amount to the heating resistor 21 is increased. On the contrary, since the air heated by the heat generated from the heating resistor 21 located upstream passes over the heating resistor 22 located downstream, the heat radiation amount of the heating resistor 22 located downstream is reduced. Therefore, the energization amount to the heating resistor 22 located downstream is reduced.
Therefore, the flow rate of air and the direction of flow can be detected based on the energization amount to the heating resistors 21 and 22.

[Electrical Configuration of Circuit Chip 30]
Next, the electrical configuration of the circuit chip 30 will be described with reference to the drawings. FIG. 4 is a circuit diagram showing an electrical configuration of the circuit chip 30.

  The circuit chip 30 includes a sensor control circuit 31, a reference voltage signal generation circuit 32, an analog multiplexer 33, a control circuit 34, and an A / D conversion circuit 35. The sensor control circuit 31 outputs an analog sensor signal Sf to the analog multiplexer 33. The sensor signal Sf is a voltage signal corresponding to the energization amount to the heating resistors 21 and 22 of the sensor chip 20.

  The reference voltage signal generation circuit 32 is configured by connecting resistors R1 to R4 in series between the constant voltage power supply Vcc and the ground. Therefore, the reference voltages Vref1, Vref2, and Vref3 are respectively output from the output terminal a between the resistors R1 and R2, the output terminal b between the resistors R2 and R3, and the output terminal c between the resistors R3 and R4, and the sensor signal Sf. Output regardless.

  These three reference voltages are used to perform non-linearity correction of A / D conversion characteristics described later. Since the resistance R2 and the resistance R3 have the same resistance temperature coefficient, the rate of change ΔR of the resistance value with respect to the change in the ambient temperature is the same. Therefore, (Vref2-Vref1) :( Vref3-Vref1) is always a constant ratio regardless of changes in ambient temperature. Moreover, since the resistance values of the resistor R2 and the resistor R3 are also equal, the ratio is 1: 2. The reference voltages Vref1 and Vref3 are set near the lower limit and the upper limit of the fluctuation range of the sensor signal Sf, and the reference voltage Vref2 is set near the middle between the lower limit and the upper limit.

  The analog multiplexer 33 uses the sensor signal Sf from the sensor control circuit 31 and the reference voltages Vref1 to Vref3 from the reference voltage signal generation circuit 32 as a voltage signal Vin to the A / D conversion circuit 35 via the voltage signal input terminal 35a. Output. The control circuit 34 outputs a select signal to the analog multiplexer 33. Further, the control circuit 34 outputs the pulse signal PA to the NAND circuit 36 a and outputs the pulse signal PB to the pulse selector 38 and the latch circuit 39. The pulse signal PB is a pulse signal for obtaining a constant sampling period Δt (for example, ˜100 μsec).

  The analog multiplexer 33 outputs the sensor signal Sf and the reference voltages Vref1 to Vref3 to the power supply line L1 via the voltage signal input terminal 35a. The sensor signal Sf and the reference voltages Vref1 to Vref are repeatedly selected and output in a predetermined order based on a select signal supplied from the control circuit 34.

  The A / D conversion circuit 35 basically has the same configuration as the A / D conversion circuit described in Japanese Patent Laid-Open No. 5259907, and converts the analog voltage signal Vin into a binary digital value. . The A / D conversion circuit 35 includes a pulse running circuit 36, a counter 37, a latch circuit 39, a pulse selector 38, an encoder 40, a signal processing circuit 41, a correction arithmetic circuit 42, an EPROM 43, and a data output line. L2.

  The pulse running circuit 36 is formed by connecting, in a ring shape, a NAND circuit 36a whose inversion operation time changes according to the voltage signal Vin and an even number of inverters 36b whose inversion operation time changes according to the voltage signal Vin. . The voltage signal input terminal 35a is connected to the power supply line L1 of the NAND circuit 36a and the inverter 36b in the pulse traveling circuit 36, and applies the voltage signal Vin as the power supply voltage of the NAND circuit 36a and the inverter 36b.

  The counter 37 counts the number of circulations of the pulse signal in the pulse traveling circuit 36 from the number of inversions of the output level of the inverter 36b provided in the preceding stage of the NAND circuit 36a in the pulse traveling circuit 36, and obtains a binary digital value. appear. The latch circuit 39 latches the digital value output from the counter 37. The pulse selector 38 takes in the output of each inverting circuit (that is, the NAND circuit 36a and the inverter 36b) constituting the pulse traveling circuit 36, extracts a pulse signal that is circulating in the pulse traveling circuit 36 from the output level, A signal representing the position is generated.

  The encoder 40 generates a digital value corresponding to the output signal from the pulse selector 38. The signal processing circuit 41 inputs the digital value from the latch circuit 39 as the upper bit and the digital value from the encoder 40 as the lower bit, and subtracts the upper bit data from the lower bit data to thereby generate the pulse signals PA and PB. A binary digital value representing a phase difference is generated.

  Here, among the binary digital values representing the phase difference between the pulse signals PA and PB, the digital values of the sensor signal Sf and the reference voltages Vref1 to Vref3 are DSf and DV1 to DV3, respectively. A difference between any two of DV3 is defined as a first difference ΔD1, and a difference between two digital values different from a combination of digital values for obtaining the first difference is defined as a second difference ΔD2.

  Assuming that the relationship between the voltage Vin of the sensor signal Sf and the digital value DSf is a linear approximation, the straight line can be expressed by the following equation (1).

  DSf = αVin + β (1)

  However, α is a gain (slope), β is an offset (intercept), and both are constants. Therefore, when the difference between any two digital values of the sensor signal Sf and the reference voltages Vref1 to Vref3 is obtained in the same manner, the offset β is deleted. For example, when the difference obtained by subtracting DV1 from DV2 is obtained as the first difference ΔD1, the result is as shown in Expression (2). Become.

ΔD1 = DV2−DV1 = α (Vref2−Vref1) (2)
ΔD2 = DSf−DV1 = α (Sf−Vref1) (3)

  When the ratio R between the first difference ΔD1 and the second difference ΔD2 is obtained, the following equation (4) is obtained.

R = ΔD2 / ΔD1 = (Sf−Vref1) / (Vref2−Vref1) (4)

  That is, the gain α is also deleted. Therefore, if the ratio R is an output value, an output value that is not affected by the ambient temperature can be obtained without requiring a temperature detection circuit.

  The data output line L2 outputs the digital value generated by the signal processing circuit 41 to the correction arithmetic circuit 42. The correction calculation circuit 42 corrects a variation in digital value caused by temperature characteristics (hereinafter referred to as temperature characteristic correction) with respect to the digital value input from the signal processing circuit 41, and a digital value caused by individual differences in the flow rate sensor 10. Correction processing such as correction of variation (hereinafter referred to as individual difference correction).

  The EPROM 43 stores rewritable map data and computer programs used when the correction calculation circuit 42 performs correction processing. Note that the latch circuit 39 and the pulse selector 38 operate in response to the pulse signal PB output from the control circuit 34.

[Operation of A / D Conversion Circuit 35]
FIG. 5 is a timing chart of each signal in the A / D conversion circuit 35. When a pulse signal PA as shown in FIG. 5 is given to the NAND circuit 36a in the pulse running circuit 36, the NAND circuit 36a and each inverter 36b sequentially start an inverting operation at a speed according to the voltage signal Vin, During the input period of the pulse signal PA, the signal circulation operation is continuously performed, and binary digital data indicating the number of pulse signal circulations is given to the counter 37 in real time.

  Thereafter, the latch circuit 39 latches the digital data of the counter 37 every time the pulse signal PB rises (FIG. 5), and outputs it to the signal processing circuit 41. Further, the pulse selector 38 detects the rotation position of the pulse signal in the pulse traveling circuit 36, and the encoder 40 generates digital data corresponding to the rotation position and outputs it to the signal processing circuit 41.

Then, the signal processing circuit 41 generates binary digital data representing the phase difference between the pulse signals PA and PB.
Further, since the sensor signal Sf and the reference voltages Vref1 to Vref3 are sequentially supplied to the A / D conversion circuit 35 as the voltage signal Vin, in the A / D conversion circuit 35, the sensor signal Sf and the reference voltages Vref1 to Vref3 are Through the above processes, the data is sequentially converted into digital data.

  The correction calculation circuit 42 performs temperature characteristic correction and individual difference correction on the digital data input from the signal processing circuit 41. The correction calculation circuit 42 can be configured by a DSP (Digital Signal Processor) because of its high processing speed.

[A / D conversion characteristics correction]
FIG. 6 is a diagram illustrating the relationship between the input voltage Vin and the digital value of the input voltage Vin converted by the A / D conversion circuit 35. As indicated by a broken line in the figure, the relationship between the input voltage Vin and its digital value has non-linear (curved) A / D conversion characteristics.

  Therefore, the non-linearity error is corrected to increase the accuracy of A / D conversion. In this embodiment, correction using a quadratic function equation (also referred to as parabolic correction) is performed on the A / D conversion characteristics. As a specific method of this correction, a method (Japanese Patent No. 3979358) proposed by the applicant of the present application prior to the present application is used.

In this method, the A / D conversion circuit 35 actually performs A / D conversion on the reference voltages Vref1 to Vref3, sets a linear correction formula based on the result, and digitally calculates the sensor signal Sf based on the linear correction formula. The value (A / D conversion result) is corrected. For this reason, it is possible to perform linear correction with high accuracy by an appropriate linear correction formula corresponding to the temperature change at any time, regardless of environmental changes such as temperature changes.
A curve indicated by a solid line in FIG. 6 is a correction curve indicating a result of correction using a quadratic function equation.

FIG. 7 is a diagram illustrating an error when correction is performed using the above correction curve when the voltage signal Vin is A / D converted. The voltage signal Vin was changed to about 2.5 to 3.5 V, and a plurality of measurement points were provided. In the figure, a solid line indicates an error when the ambient temperature is −40 ° C., and a broken line indicates an error when the ambient temperature is 140 ° C. The error when the ambient temperature was −40 ° C. was less than about 2%, and the error when the ambient temperature was 140 ° C. was less than about −2%.
That is, if the above method is used, the voltage signal Vin can be A / D converted with high accuracy.

[Temperature correction of sensor chip 20]
The output of the sensor chip 20 varies depending on the temperature of the sensor chip 20 even when the detection target flow rate is the same. Therefore, in addition to the above-described temperature characteristic correction of the circuit chip 30, the detection accuracy of the flow sensor 10 is further increased by correcting the temperature characteristic of the output of the sensor chip 20.

  FIG. 8 is a diagram showing the relationship between the correction value corrected based on the correction curve shown in FIG. 6 and the flow rate. A curve indicated by a broken line in the drawing is a curve indicating a relationship between a correction value and a flow rate when the sensor chip 20 is at a reference temperature (for example, 20 ° C.). A curve indicated by T2 in the figure is a curve showing the relationship between the correction value and the flow rate when the temperature of the sensor chip 20 is a temperature T2 higher than the reference temperature. A curve indicated by T1 in the figure is a curve showing the relationship between the correction value and the flow rate when the temperature of the sensor chip 20 is a temperature T1 lower than the reference temperature.

  The correction value in each case where the temperature of the sensor chip 20 is higher or lower than the reference temperature is corrected so as to be the correction value at the reference temperature. This correction method is a feature of the present invention. As a result of studies by the present inventors, it has been found that correction can be performed by using a linear function represented by the following equation (5).

  In Expression (5), V is a measured value (output value of the signal processing circuit 41), Va is a corrected value (output value of the correction arithmetic circuit 42) obtained by correcting the measured value V, T is a measured temperature (actual ambient temperature), T1 is a first reference temperature (fixed), T2 is a second reference temperature (fixed), VT1 is a measured value (fixed) at the first reference temperature, and VT2 is a measured value (fixed) at the second reference temperature.

  Va = V- (VT2-VT1) * (T-T1) / (T2-T1) (5)

In the above formula (5), (VT2−VT1) is a flow rate proportional coefficient proportional to the flow rate, and (T−T1) / (T2−T1) is a temperature proportional coefficient proportional to the temperature.
That is, the temperature characteristic correction of the sensor chip 20 is performed using the linear function shown in the above equation (5). At this time, the first and second reference temperatures can be arbitrarily selected.

For example, the first reference temperature can be 20 ° C. and the second reference temperature can be 90 ° C. In this case, the measured value at 20 ° C. for the predetermined flow rate is the measured value VT1 at the first reference temperature, and the measured value at 90 ° C. is the measured value VT2 at the second reference temperature.
In this case, the temperature proportionality coefficient is (T−T1) / (T2−T1) = (25−20) / (90−20) = 5/70.

  9 and 10 are diagrams illustrating detection errors of the flow sensor 10. The output value of the flow sensor 10 when the ambient temperature of the flow sensor 10 is −30 ° C., −20 ° C., 0 ° C., 20 ° C., 50 ° C., 80 ° C., and 100 ° C. is expressed by the above equation (5). And the error between the value and the actual flow rate was calculated. The first reference temperature T1 is 20 ° C., the second reference temperature T2 is 100 ° C., and the number of flow rate measurement points is 10 for each ambient temperature. The range of the flow rate is 0 to 250 [g / s]. 9 shows the detection error when the output of the A / D conversion circuit 35 fluctuates by + 2%, and FIG. 10 shows the detection error when the output of the A / D conversion circuit 35 fluctuates by −2%. Indicates.

As shown in FIGS. 9 and 10, the error of the flow sensor 10 is caused even when the output of the A / D conversion circuit 35 fluctuates ± 2% and the temperature changes to −30 to 100 ° C. It can be seen that it is less than 0.6%.
In other words, by correcting the A / D conversion characteristics and the temperature characteristics in the circuit chip 30 and further correcting the temperature characteristics of the sensor chip 20, the flow rate detection accuracy can be further improved. .

  As described above, since the temperature characteristic correction of the sensor chip 20 can be performed using a linear function, the calculation is simpler than the conventional one using a high-dimensional function, so the processing time for performing the correction Can be shortened.

In addition, since map data for correction is created using the above linear function, the flow rate measurement point is reduced to 10 for each temperature, and correction is performed using a conventional high-dimensional function. The amount of data can be greatly reduced compared to the map data.
Therefore, since the memory area for storing the map data can be greatly reduced, the flow sensor 10 can be reduced in size.
Note that if correction using the conventional high-dimensional function is performed with the same accuracy as the method of this embodiment, at least 100 measurement points are required for each temperature, and the map data is Become enormous.

[Individual difference correction]
Since there are individual differences in the flow sensor 10, there is an offset between the output characteristics between the flow sensors. Therefore, a reference flow sensor (master sensor) is determined, and the offset amount between the output characteristics of the flow sensor and the output characteristics of other flow sensors is adjusted.

[Correction procedure]
FIG. 11 is a process diagram showing a correction procedure. After correcting the circuit chip 30 (steps 1 to 3), the sensor chip 20 is corrected (steps 4 to 6). First, the ambient temperature is set to a reference temperature (for example, 20 ° C.), the flow rate is changed, and digital values from the data output line L2 at a plurality of measurement points (for example, 10 measurement points) are stored in the memory ( Step 1). This measurement is performed a plurality of times, and the average value is calculated (step 2).

  Next, the relationship between the average value obtained in step 2 and the voltage signal Vin is obtained, and the parabola is corrected (step 3). Next, the data subjected to the parabola correction in the step 3 is corrected using the above-described linear function, and the temperature characteristic correction of the sensor chip 20 is performed (step 4). Next, the offset with respect to the output characteristic of the reference flow rate sensor is corrected, and the solid difference of the sensor chip 20 is corrected (step 5).

  Then, output adjustment such as offset and gain for connecting the A / D conversion circuit 35 and the ECU is performed (step 6). This is because the input range of the ECU differs depending on the vehicle. Further, the relationship between the voltage signal Vin and the output value, which have been corrected and adjusted by executing the above steps 1 to 6, is converted into map data and stored in the EPROM 43.

  The flow sensor 10 actually attached to the vehicle selects an output value with reference to the map data stored in the EPROM 43 for the voltage signal Vin, and outputs it to the ECU. The ECU controls the amount of air taken into the internal combustion engine based on the output value input from the flow sensor 10.

<Other embodiments>
(1) In the above-described embodiment, (T2−T1) of the temperature proportionality coefficient (T−T1) / (T2−T1) in the equation (5) is the difference between the first reference temperature T1 and the second reference temperature T2. Although it was an absolute value, it can be replaced with an arbitrary constant A. Even when the flow rate detection value is corrected using this equation, the same effects as those of the above-described embodiment can be obtained.

(2) In the above-described embodiment, the flow sensor according to the present invention has been described as detecting the amount of air taken into the internal combustion engine. However, the present invention is a flow rate for detecting the flow rate of a gas used in a semiconductor manufacturing process. It can also be applied to sensors.

<Reference example>
(1) When the measurement value is V, the correction value obtained by correcting the measurement value V is Va, and the measurement value at 20 ° C. is Vb, the correction value Va may be obtained using the equation Va = V−Vb. it can. Also in this case, since the calculation is simpler than that using a conventional high-dimensional function, the processing time for performing correction can be shortened.

(2) The correction value Va can also be obtained using the equation Va = (V−Vb) / Vb. Also in this case, since the calculation is simpler than that using a conventional high-dimensional function, the processing time for performing correction can be shortened.

(3) Furthermore, when the measured value at 80 ° C. is Ve, the correction value Va can be obtained using the equation Va = (V−Vb) / (Ve−Vb). Also in this case, since the calculation is simpler than that using a conventional high-dimensional function, the processing time for performing correction can be shortened.

10 .... Flow sensor, 20 .... Sensor chip, 21,22 ... Heating resistor,
23, 24 ... Resistance temperature detector, 26 ... Membrane, 27 ... Sensing part,
28..Semiconductor substrate, 30..Circuit chip, 32..Reference voltage signal generation circuit,
35..A / D conversion circuit, 36..Pulse travel circuit, 42..Correction arithmetic circuit.

Claims (5)

A sensor unit for detecting the flow rate of gas;
In a flow sensor comprising a conversion unit that converts an analog voltage signal output from the sensor unit into a digital value,
The digital output value output from the conversion unit is V, the ambient temperature is T, the constant is A, the output value when the ambient temperature is the first reference temperature, and the ambient temperature is different from the first reference temperature. The absolute value of the difference from the output value at the second reference temperature is B, the absolute value of the difference between the ambient temperature and the first reference temperature is C, and the output value is in accordance with an error caused by the ambient temperature. When the corrected value is Va,
Va = V− (B × C / A)
A flow sensor comprising a correction circuit for calculating   The flow rate sensor according to claim 1, wherein the constant A is an absolute value of a difference between the first and second reference temperatures. The converter is
An A / D conversion circuit for converting the analog voltage signal into a binary digital value, wherein a plurality of inverting circuits that invert the input signal and output the inverting operation time depending on the power supply voltage are connected. In addition, one of the inverting circuits is configured as a starting inverting circuit that can control the inverting operation from the outside, and a pulse traveling circuit that travels a pulse signal when the starting inverting circuit starts operating,
A voltage signal input terminal connected to a power supply line of each inverting circuit in the pulse traveling circuit and applying the voltage signal as a power supply voltage of each inverting circuit;
A reference voltage signal generating circuit for generating a plurality of reference voltage signals having different voltages from each other;
Input voltage switching means for switching and inputting a voltage signal for A / D conversion and a plurality of reference voltage signals generated by the reference voltage signal generation circuit to the voltage signal input terminal;
Travel position detection means for detecting a travel position of the pulse signal in the pulse travel circuit based on an output signal from each inverting circuit and generating data corresponding to the travel position;
Control means for operating the starting inversion circuit to start the running operation of the pulse running circuit, and then operating the running position detecting means when a predetermined time has elapsed;
A data output line for outputting digital data including data from the running position detecting means as an A / D conversion result;
A first difference that is a difference between any two of the digital data of the A / D conversion voltage signal and the plurality of reference voltage signals output from the data output line, and the first difference. A signal processing circuit that calculates a digital value corresponding to the flow rate based on a ratio of a second difference that is a difference between two digital data different from a combination of digital data to be obtained;
The flow sensor according to claim 1, further comprising: According to an error caused by the ambient temperature, the digital output value of the flow sensor includes a sensor unit that detects a gas flow rate and a conversion unit that converts an analog voltage signal output from the sensor unit into a digital value. In the flow correction method to correct,
The output value is V, the ambient temperature is T, the constant is A, the output value when the ambient temperature is the first reference temperature, and the ambient temperature is the second reference temperature different from the first reference temperature. The absolute value of the difference from the output value is B, the absolute value of the difference between the ambient temperature and the first reference temperature is C, and the output value is corrected according to the error caused by the ambient temperature, Va. In case,
Va = V− (B × C / A)
A flow rate correction method characterized by calculating   The flow rate correction method according to claim 4, wherein the constant A is an absolute value of a difference between the first and second reference temperatures.

Images