JP2786815B2 - Mass flow meter converter - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mass flow meter converter, and more particularly to a mass flow meter converter having a small detection sensitivity, for example, suitable for measuring the time difference of a straight pipe type Coriolis flow meter and capable of measuring a time difference with high sensitivity. Mass flowmeter converter.

[0002]

2. Description of the Related Art As is well known, a Coriolis flowmeter supports a measurement tube through which a fluid to be measured flows is supported at both ends, and when a central portion of the supported measurement tube is driven alternately in a direction perpendicular to a support line, a measurement is performed. This is a mass flow meter that detects a phase difference signal proportional to a mass flow rate at a symmetric position between a support portion at both ends of the tube and a central portion. Although the phase difference signal is an amount proportional to the mass flow rate, if the drive frequency is fixed, the phase difference signal can be detected as a time difference signal at the observation position of the measuring tube.

If the frequency of the alternating driving of the measuring tube is made equal to the natural frequency of the measuring tube, a constant driving frequency corresponding to the density of the fluid to be measured can be obtained, and the driving can be performed with a small driving energy. Recently, it has become common to drive the measuring tube at a natural frequency, and the phase difference signal is detected as a time difference signal.

FIG. 4 is a view for explaining an example of the time difference measurement of a conventional Coriolis mass flow meter. FIG. 4 (a)
Is a diagram showing a displacement signal of the measurement tube at the detection position of the measurement tube, a trapezoidal wave signal obtained by amplifying and shaping the detected sine wave signal, time on the horizontal axis, voltage on the vertical axis I have. In the drawing, trapezoids A, B, C, D,... And trapezoids A ₁ , B ₁ , C ₁ , D ₁ ,. The displacement signals of the measuring tubes having different phases with the voltages are shown. Each displacement signal is a continuous trapezoidal waveform having the same shape on the time axis, and the reference time indicating the phase difference time is one trapezoidal waveform A, B ,
C, D, for example, the peak value C or D of the hypotenuse CD
The time between the position and the position O crossing the time axis is represented by T.

[0005] In the displacement signals of the trapezoids A, B, C, D and A ₁ , B ₁ , C ₁ , D ₁ having different phases, for example, a phase difference signal will be described on the sides CD, C ₁ and D ₁ . I do.
The quadrilaterals C, C ₁ , D ₁ and C ₁ are parallelograms, and the parallel lines C
The time difference ΔT between D and C ₁ · D ₁ is a phase difference signal,
C ₁ , D · D ₁ have the same length as the time axis O · O _1, and the projection points from the points C ₁ and D ₁ to the time axis are O ₂ ,
If O ₃ , side O ₂ · O is time (T−ΔT), side O · O ₃
Indicates time (T + ΔT).

The time (T-ΔT) is shown in FIG.
+ ΔT) is shown by the pulse shown in FIG. 4D, and the time difference ΔT is each pulse during the time 2M which is twice as long as the time pulse having the time width M of one trapezoidal wave shown in FIG. 4B. The average is obtained by subtracting the width time. That is, (4 (T + ΔT) −4 (T−ΔT)) / 8 = ΔT (1)

FIG. 6 shows a block diagram of a conventional mass flow meter converter for determining the time difference ΔT. The pulse width of the (T + ΔT) pulse and the (T−ΔT) pulse is determined by the number of clocks oscillated from the clock oscillator 21. Each is measured by a counter 24, an operation is measured by a CPU (central processing unit) 25 based on the equation (1), and a digital value proportional to the mass flow rate is output from a terminal 26.

[0008]

In the conventional mass flowmeter converter described above, the time proportional to the phase difference of the measuring tube is digitally measured as the number of clocks, but the time having a resolution corresponding to the accuracy is measured. In order to measure, the number of clocks proportional to the time difference counted at all flow rates must be a sufficient number of clocks for accuracy. In a curved tube Coriolis flowmeter where the measuring tube is an axisymmetric curved tube and the Coriolis force generated when the curved tube is driven alternately in the direction perpendicular to the symmetry axis generates around the symmetry axis, the moment around the axis is By increasing the value, a sufficiently high-sensitivity time difference signal can be detected, and the time difference signal can be counted by the number of clocks. However, as in a straight tube type Coriolis flowmeter, the measurement tube has a large bending rigidity and, therefore, a natural frequency. If it is large, the time difference signal generated by the Coriolis force is also small, and a clock oscillator that generates the number of clocks of a high oscillation frequency (for example, 100 MHz) suitable for the accuracy has a problem in reliability and stability at present. Moreover, it is difficult to obtain a satisfactory clock oscillator at low cost.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a mass flowmeter converter for simply and accurately measuring a time difference without using a clock oscillator having a high time resolution.

[0010]

In order to solve the above-mentioned problems, the present invention detects a Coriolis force acting on a measuring tube driven alternately around a support point at a constant frequency at a symmetric position of the measuring tube. In the converter of the mass flow meter for detecting the time difference between the paired displacement signals as ΔT and measuring the mass flow rate in proportion to the time difference ΔT, the displacement signal is converted into a positive and negative peak having an absolute value equal to a reference time axis. When a pair of trapezoidal wave signals having a voltage is formed, and a difference between a time when the oblique side of one trapezoidal wave passes the reference time axis and a time until the peak voltage position of the oblique side passes is T. Have a pulse width equal to the lead time (T + ΔT) and the delay time (T−ΔT), respectively, at which the peak value of the corresponding oblique side of the other trapezoidal wave signal has passed, and have a constant peak value (T + ΔT).
A pulse and a pulse (T−ΔT) having a different peak value from the (T + ΔT) pulse are set as input pulses.
A gate circuit that opens and closes so that (N> 1) the (T + ΔT) pulses and the N (T−ΔT) pulses are alternately taken in; a gate circuit that is closed and N (T + △)
In the period of taking in (T), N (T) taken in
+ ΔT) pulses are sequentially charged, and in the next open period of the gate circuit, N (T−ΔT) pulses are captured, and the captured N (T−ΔT) pulses are charged. It has a charge / discharge circuit that sequentially discharges from a voltage and outputs a subtracted voltage, and a voltage holding circuit that holds the subtracted output voltage, and obtains a mass flow rate in proportion to the held voltage. It is assumed that.

[0011]

Each of them has a constant peak value and a pulse width (T + Δ
(T + ΔT) pulse and (T + ΔT) pulse of (T) and a voltage having a different sign, and the pulse width (T−ΔT) of (T−ΔT)
-ΔT) pulses, and N (T
+ ΔT) to charge the battery by switching only to input to the charging / discharging circuit. In the next period, the input pulse is switched from (T + ΔT) pulse to (T−ΔT) pulse and N (T−ΔT) pulses are switched.
T) Only the pulse is input to the charge / discharge circuit, and N (T)
+ (T) pulses to discharge the voltage charged in the charge / discharge circuit with a voltage corresponding to N (T−ΔT) pulses having different signs, thereby reducing the number of N (T + ΔT) pulses to N Is subtracted from the (T-ΔT) pulse voltage to output a difference voltage having a time difference ΔT equivalent to 2N, and the mass flow rate is calculated from the obtained difference voltage.

[0012]

【Example】

Embodiment 1 (corresponding to claim 1) FIG. 1 is a circuit block diagram for explaining Embodiment 1 of a mass flow meter according to the present invention, in which 1 is a switch SW.
_{1 is} an input terminal, 2 is a (T + ΔT) pulse input terminal, 3
Is a (T-ΔT) pulse input terminal, 4, 5, 8 are contact points,
6 is an inverter, 7 is an operational amplifier, 9 is a voltage holding circuit,
10 is an A / D (analog / digital) converter, 10 is a CPU (Central Processing Unit), and 12 is an output terminal.

The (T + ΔT) pulse is, for example, a pulse having a constant voltage (−E ₁ ) and a pulse width of (T + ΔT);
The (T−ΔT) pulse has a constant voltage (+ E ₂ ) and (T−ΔT)
T) is a pulse having a pulse width of T). Amplifying and shaping two paired sinusoidal displacement signals having a phase difference at a detection position of a measuring tube (not shown) by Coriolis force,
Assuming that the trapezoidal wave has the same positive and negative voltage peak values with respect to the time axis, the time T is defined as the time from the time when the hypotenuse of the trapezoidal wave having a constant voltage height of one trapezoidal signal passes through the time axis to the hypotenuse. ΔT indicates the time required for the peak value to reach the time axis cross position of the hypotenuse, and ΔT is a pair of the one trapezoidal wave signal and the other trapezoidal wave signal which make a pair proportional to the phase difference generated in the measurement tube by the Coriolis force. Is the time difference between

For example, a (T + ΔT) pulse having a peak value (−E ₁ ) is input to a terminal 2, and a peak value (+
(T-ΔT) pulses E ₂₎ it is input. A contact 4 is connected to the terminal 2 in series, and a contact 5 is connected to the terminal 3 in series. The contact 4 is input to an integrator comprising an input resistor R and a negative feedback capacitor C, which are integral constants RC, and an operational amplifier circuit 7. Note that a predetermined positive reference voltage (not shown) is applied to a positive input terminal of the operational amplifier circuit 7. In addition, the feedback capacitor C, contact 8 which is opened and closed by the switch SW ₂ are connected in parallel.

A voltage holding circuit 9 is connected to the operational amplifier circuit 7, and the voltage holding circuit 9 has an A / D (analog / analog).
A digital) conversion circuit 10 and a CPU 11 are connected.

FIG. 2 is a time chart of a pulse train for explaining the operation of the first embodiment shown in FIG.
(A) is a pulse train of (T + ΔT) pulses having a peak value (−E ₁ ), and FIG. 2 (b) is a pulse (T−Δ) having a peak value (+ E ₂ ).
Pulse train T), FIG. 2 (c) is a chart gate driving voltage of the switch SW _1, Fig. 2 (d) pulse signal waveform of the switch SW _2, FIG. 2 (e) is showing a measuring cycle.

[0017] ON the contacts 4 when the gate drive voltage applied to the switch SW ₁ is High level signal (closed)
Then, the contact 5 is turned off (open). Conversely, when the signal is a Low level signal, the contact 4 is turned off and the contact 5 is turned on via the inverter 6. The pulse width of the gate driving voltage of the switch SW ₁ is intended to determine the inter-closing gate for taking the (T + ΔT) pulse or (T-ΔT) pulses, (T + delta
T) and (T−ΔT) pulses are N (N
> 1) The time is set to be input.

In the period t _{1 to} t ₂ of the high-level signal of the gate drive voltage, the contact 5 is turned off and the contact 4 is turned on, so that the voltage (−E) inputted from the terminal 2 is (T + ΔT).
The pulse is input to an integrator including the resistor R, the feedback capacitor C, and the operational amplifier circuit 7. Note that (T + ΔT)
Immediately before the pulse is input (t ₁ ) (not shown), the electric charge charged in the feedback capacitor C is completely discharged by the contact 8 driven by the switch SW ₂ (FIG. 2D). . It should be noted that the drive of the switch SW ₂ is, CPU1
1 is programmed.

When the flow rate is constant, each of the N (T + Δ
Since the (T) pulse has the same area ((T + ΔT) · (−E)), when the (T + ΔT) pulse of (−E ₁ ) is input, the inverted (T + ΔT) pulse is output from the integrator. Is integrated. That is, the integration circuit becomes a charging circuit, and the output voltage increases in proportion to the number of (T + ΔT) pulses.

[0020] FIG. 3 is a diagram showing the voltage-time characteristics of the charge and discharge voltage at the time period t ₁ ~t _2, t ₂ ~t ₃ of the time chart shown in FIG. 2, the time t ₁ ~t ₂ In the charging period, the points A to B are charged with a voltage proportional to the number of (T + ΔT) pulses according to the straight lines A and B.

Next, during the period t ₂ -t ₃ of the gate drive voltage of the switch SW ₁ of the Low level signal, the contact 4 is turned off and the contact 5 is turned on by the High level signal from the inverter 6. Input from (+ E
₂ ) Only the (T-ΔT) pulse is input to the integrator.
At this time, the integrator becomes a discharge circuit based on the inverted output of the (T−ΔT) pulse, and the integrator charged to the voltage B by the (T + ΔT) pulse reduces the number of N (T−ΔT) pulses to N. The corresponding voltage is subtracted in proportion to time.

Clearly, the pulse width is (T + ΔT)> (T
−ΔT), a voltage V corresponding to N (T + ΔT) −N (T−ΔT) = 2NΔT (2) is output from the integrator (charge / discharge circuit) as shown in FIG.

As shown in FIG. 2 (e), time t ₁ to t ₂
The period of t ₃ is a measurement period during which the analog output voltage V proportional to 2NΔT is output. The output voltage V is held by the voltage holding circuit 9, and the held voltage is digitally converted by an A / D (analog / digital) conversion circuit 10 and read into the CPU 11 to calculate a mass flow rate. However, a predetermined time is required for the calculation, and further, a period for discharging the charge corresponding to the voltage V charged in the integration capacitor C and preparing for the measurement in the next measurement period is required.

The period of time from the time t ₃ when the voltage V is outputted next input pulse number N is inputted t ₃ ~t ₄ discharges the processing time of the CPU11 above, the charge of the feedback capacitor C , And a time equal to the time t _{1 to} t ₂ and the time t _{2 to} t ₃ . Thus, the input pulse number 3 up to the time t ₁ ~t ₄
One measurement cycle corresponds to the time input by N.

The discharge time of the feedback capacitor C is shown in FIG.
It is given by the High level signal of the voltage pulse of the switch SW ₂ shown in (d). High level signal is CP
Start-up by the operation completion signal of U11, switch SW
₁ of 1 measurement cycle is lowered by a signal that is output at the time t ₄ when finished. Time corresponding to the input time of the input pulse 3N next input from the fall time t ₄ is the next measurement cycle.

However, starting at time t ₄ , input pulse 3
The next measurement cycle corresponds to the time N is entered, in the previous measurement cycle up to the time the above-mentioned t ₁ ~t _4, unlike the polarity of the gate driving voltage of the switch SW _1, the time t ₁
Up to t _{4, the} signal was a high level signal, while the next t
_In the measurement cycle from _{4 to} t ₇ (not shown), the signal becomes a Low level signal.

For this reason, in the next measurement cycle, during the first period of time t _{4 to} t ₅ , the battery is charged with a negative voltage by the inverted output of N (T−ΔT) pulses having a constant peak value (+ E ₂ ). is, in the period following time t ₅ ~t _6, a constant wave height (-
E ₁ ) is discharged at a positive voltage by the inverted output of the N (T + ΔT) pulses, so that the output voltage V varies from time t _{1 to} t ₄
The output voltage V is the same as that in the previous measurement cycle up to.

At time t ₃ in the previous measurement cycle and time t ₆ in the next measurement cycle, the voltage V = 2NΔT output from the integrator is held in the voltage holding circuit 9, and the held analog voltage Is input to the A / D converter 10 and output as a digital signal proportional to the analog voltage. The CPU 11 calculates a mass flow rate of the digital signal in proportion to the voltage V, and outputs the mass flow rate from the terminal 12. As described above, the switch SW ₁
The switching of the contacts 4 and 5 driven by the gate drive voltage of (1) performs charge / discharge calculation of N (T + ΔT) and (T−ΔT) pulses, and outputs a mass signal whose sensitivity is increased by 2N times. You.

The output voltage V is an analog voltage.
The signal is output via an OP-Amp (operational amplifier). If the output voltage V is small and is near zero level, the operation of the OP-Amp becomes unstable due to zero drift or the like. For example,
If the peak value (−E ₁ ) of the (T + ΔT) pal and the peak value (+ E ₂ ) of the (T−ΔT) pulse are equal to | E ₁ | = | E ₂ |, the fluid to be measured is flowing. When there is no ΔT = 0, the output voltage V = 0 and the operation becomes unstable due to the drift. Therefore, the region of the output voltage V measured by selecting | E ₁ |> | E ₂ | Can be a stable positive voltage region not including zero voltage.

Further, in order to make the region of the output voltage V to be measured a negative voltage region not including the zero voltage, a (T-ΔT) pulse wave such that the output voltage V becomes negative at zero flow rate. It is necessary to select an n value that multiplies the high value by n> 1. The (+ nE ₂ ) indicated by the dotted line in FIG. 1 is the (T−ΔT) pulse of the voltage (+ nE ₂ ) selected in this manner, and the measured voltage of (−V ₁ ) indicated by the dotted line in FIG. Can be a negative voltage range.

According to the converter for the mass flow meter according to the first embodiment described above, 2N times that is obtained by subtracting N (T + ΔT) pulses and N (T−ΔT) pulses having substantially equal areas. Since the obtained time difference 2NΔT signal is obtained, N
By selecting the value of, high-sensitivity and high-accuracy mass flow rate can be calculated without having a special clock oscillator.

[0032]

As is apparent from the above description, the present invention has the following effects. (1) Effect of the first aspect: In order to obtain the time difference ΔT proportional to the Coriolis force, each of them has a constant peak value (T + Δ
T) pulse and (T−ΔT) pulse with the opposite voltage (T−
ΔT) Since N pulses having a pulse width of pulse are charged / discharged to obtain an output voltage of 2N · ΔT, a high-oscillation that matches the accuracy in a mass flow meter with low sensitivity like a straight tube type Coriolis flow meter. Even if a special clock pulse oscillator for oscillating a clock of a frequency is not prepared, a stable and highly sensitive mass flow signal can be detected by inexpensive means.

[Brief description of the drawings]

FIG. 1 is a circuit block diagram for explaining Embodiment 1 of a mass flow meter according to the present invention.

FIG. 2 is a pulse train chart for explaining the operation of the first embodiment shown in FIG.

Is a diagram showing the voltage characteristics of the charge and discharge voltage in FIG. 3 period _{t 1 ~t 2, t 2 ~t} 3 of FIG.

FIG. 4 is a diagram for explaining an example of a time difference measurement of a conventional Coriolis mass flow meter.

FIG. 5 is a block diagram showing a conventional mass flow meter converter for calculating a time difference ΔT.

[Explanation of symbols]

1 ... input terminal of the switch SW _1, 2 ... positive pulse input terminal, 3 ... negative pulse input terminal, 4, 5, 8 ... contact, 6 ...
Inverter, 7 operational amplifier, 9 voltage holding amplifier circuit,
10 ... A / D (analog / digital) converter, 10 ...
CPU (Central Processing Unit), 12 output terminals.

Continuation of front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G01F 1/84

Claims (1)

(57) [Claims] 1. A Coriolis force acting on a measuring tube driven alternately around a supporting point at a constant frequency is detected as a time difference ΔT between a pair of displacement signals detected at symmetric positions of the measuring tube. In a converter of a mass flow meter for measuring a mass flow rate proportional to a time difference ΔT, the displacement signal is a pair of trapezoidal wave signals having positive and negative peak voltages having the same absolute value with respect to a reference time axis. When the difference between the time when the hypotenuse passes through the reference time axis and the time until the peak voltage position of the hypotenuse passes is T, the peak value of the corresponding trapezoid of the other trapezoidal wave signal has passed. (T + ΔT) having a pulse width equal to the advance time (T + ΔT) and the delay time (T−ΔT), respectively, having a constant peak value
A pulse and a pulse (T−ΔT) having a different peak value from the (T + ΔT) pulse are set as input pulses.
A gate circuit that opens and closes so that (N> 1) the (T + ΔT) pulses and the N (T−ΔT) pulses are alternately taken in; a gate circuit that is closed and N (T + △)
In the period of taking in (T), N (T) taken in
+ .DELTA.T) pulses are sequentially charged, and in the next open period of the gate circuit, N (T-.DELTA.T) pulses are fetched, and the fetched N (T-.DELTA.T) pulses are charged. A charge / discharge circuit that sequentially discharges from the voltage and outputs a subtracted voltage; and a voltage holding circuit that holds the subtracted output voltage, wherein a mass flow rate is obtained in proportion to the held voltage. And mass flow meter converter.