JPH0875520A - Coriolis mass flowmeter - Google Patents

Abstract

(57) [Abstract] [Purpose] When measuring the phase difference between the upstream and downstream detection signals, the delay due to AGC amplification can be corrected so that the error due to the difference in both amplitudes is not included, and the measurement accuracy is improved. Let [Structure] An upstream / downstream vibration amplitude difference detection unit 58, an amplitude control unit 59, and a phase difference measurement unit 60 are provided, and AGC amplification is performed so as to prevent a difference between the upstream and downstream vibration amplitudes, and the phase difference is measured. In the mass flowmeter, the arithmetic processing unit 61 detects the gain control voltage, frequency, and ambient temperature of the AGC amplifiers 52a and 52b, and refers to the ROM table 64 according to these values to obtain the propagation delay amount of the AGC amplifier. By subtracting this amount from the output of the phase difference measuring unit 60, the delay amount is corrected and accurate detection is possible.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a Coriolis mass flowmeter for determining the flow rate by detecting the phase difference of the vibration of the pipe due to the mass and velocity of the fluid on the upstream side and the downstream side of the pipe generated by the Coriolis force. Regarding

[0002]

2. Description of the Related Art FIG. 4 shows the operating principle of a Coriolis mass flowmeter. That is, 1 is a U-shaped pipe through which the measurement fluid flows,
A permanent magnet 2 is fixed to the tip portion thereof, and both ends of the U-shaped pipe 1 are fixed to a base 3. Reference numeral 4 denotes an electromagnetic drive coil provided so as to sandwich the U-shaped pipe 1. Reference numeral 5 denotes a support frame for holding the electromagnetic drive coil 4. The frame 5 is firmly fixed to the base 3. U-shaped pipe 1 is a base 3 like a tuning fork
The part of is the node of vibration, and the structure is less likely to lose vibration energy. Reference numerals 11 and 12 are electromagnetic pickups for detecting the displacement of both legs of the U-shaped pipe.
When the U-shaped pipe 1 is vibrated (sin ωt) at its natural frequency by the electromagnetic force acting between the drive coil 4 and the permanent magnet 2 fixed to the U-shaped pipe 1 facing the drive coil 4, it flows in the U-shaped pipe. Coriolis force is generated in the fluid.

FIG. 5 shows how the U-shaped pipe vibrates. The magnitude of this Coriolis force is proportional to the mass of the fluid flowing in the U-shaped pipe and its velocity, and the direction of the force is the direction of movement of the fluid.
It coincides with the direction of the vector product of the angular velocities that excite the U-shaped pipe 1. Further, since the flow directions of the fluid flow on the input side and the output side of the U-shaped pipe 1 are opposite, a twisting torque is generated in the U-shaped pipe 1 by the Coriolis force on both leg sides. This torque changes at the same frequency as the excitation frequency, and its amplitude value is proportional to the mass flow rate of the fluid. FIG. 6 shows a vibration mode generated by this torsion torque.

The mass flow rate can be known by detecting the amplitude of the torque of this torsional vibration with the pickups 11 and 12, but the actual vibration of the U-shaped pipe is Coriolis to the vibration excited by the electromagnetic drive coil 4. The torsional vibration due to the force is superimposed, and the upstream side is represented by sin (ωt−α) and the downstream side is represented by sin (ωt + α). Therefore, the signals e1 and e detected by the pickups 11 and 12 are
2 has a waveform having a phase difference (Δt) as shown in FIG. 7. This phase difference depends on the piping and the excitation frequency,
For example, in the case of U-shaped pipe, the resonance frequency of U-shaped pipe is 80H.
z, a phase difference of about 120 μS occurs at the maximum flow rate,
A resolution of 1% must be compensated for in this 1/20 range. Therefore, a time measurement resolution of 60 nS is required.

There are various methods for this phase measurement, but the simplest method is the method of counting the time difference gate by the reference clock. An example thereof is shown in FIG. That is, the upstream pickup signal 20 and the downstream pickup signal 21 are amplified by the amplifier 22 (amplification factor: B), then binarized by the comparator 23, and the exclusive OR circuit 24 performs an exclusive OR operation of the binary signals. Is performed to obtain a gate pulse 25 having a pulse width corresponding to the time difference between the upstream pickup signal and the downstream pickup signal, and this is measured by the counter 26 by the reference clock 27. The frequency of the reference clock in this case needs to be about 20 MHz or higher.

By the way, when the U-shaped pipe is used in an actual plant, there is a problem that the pressure loss is large because it is bent and it is difficult to clean the pipe. Therefore, a straight pipe type Coriolis flowmeter using a straight pipe has also been proposed. FIG. 9 shows an example of a straight pipe type Coriolis flowmeter. In FIG. 9, 1
Reference numeral 5 denotes a straight pipe through which the measurement fluid flows, the permanent magnet 2 is fixed to the central portion thereof, and both ends of the straight pipe 15 are fixed to the base 3. Reference numeral 4 denotes an electromagnetic drive coil provided so as to sandwich the straight pipe 15 and reference numeral 5 denotes a support frame for holding the electromagnetic drive coil 4, which frame is firmly fixed to the base 3.
In the straight pipe system, the pipe through which the fluid passes has a high rigidity and is more difficult to bend than the U-shaped pipe, so that there is a problem in that the time difference is small.

For example, the resonance frequency of a straight pipe is about 1 KHz, and a phase difference of about 2 μS occurs at the maximum flow rate.
It is necessary to measure with a resolution of 1% in 20 ranges. Therefore, a time measurement resolution of 1 nS is required. In addition, the measurement by the counter requires a reference clock of 1 GHz and cannot be manufactured in practice. Even if it is possible, if a comparator is used to obtain the time difference signal from the pickup signal, the dead zone of the input signal is Jitter occurs due to a problem (a halfway level where the output of the comparator is not "1" or "0" is called a dead zone, and how quickly the input signal crosses this dead zone has a great effect). It is doubtful whether it will be obtained.

For this reason, conventionally, the structure shown in FIG. 10 is used for the measurement, and the upstream pickup signal 20 and the downstream pickup signal 21 are subtracted, that is, sin (ωt + α) −sin (ωt−α) = 2cosω
The calculation of t * sinα is performed by the differencer (subtractor) 28, and sinα
A weak phase signal (having a period of 1 mS and a phase α of 0.1 nS) is obtained, which is highly amplified (amplification factor: C) by an amplifier 29 to generate an exciting current s of an electromagnetic drive coil.
The phase of inωt is advanced by 90 ° in the frequency multiplier 31 and c
Get osωt. When the cos ωt is a positive value, C * sin α * cos ωt is output, and when the cos ωt is a negative value, B * (− cos ωt * sin α) is output.
As described above, the timing of code control is not obtained from the waveform of the code control target, but other signals are used in order to reduce the influence of malfunction due to noise or the like.

C * sinα * obtained in this way
Although there are various methods for measuring cosωt, if the amount of time is measured using, for example, a microcomputer (also referred to as a microcomputer), C * sinα * cosωt is initially set as indicated by reference numeral 44 in FIG. Is charged in the capacitor for several cycles, and then the SW is switched so that the constant current circuit 33 discharges it with a constant current. If the time taken to cross the value is measured, the value C * sinα is converted into a pulse width by the comparator 34, and the phase difference can be obtained by measuring this pulse width by the microcomputer. Note that, for sin α, α is very small, so in FIG.
Is approximated by.

FIG. 11 shows the waveforms of the respective parts when the amplitudes of the detection signals on the upstream side and the downstream side are the same and the code control signal is synchronized with cosωt. It should be noted that each of these signals is represented by a mathematical expression as follows. V _U = A * sin (ωt−α) upstream detection signal 20 V _D = A * sin (ωt + α) downstream detection signal 21 V _V = 2A * sin α * cosωt downstream / upstream differential signal 41 Vc = | C * 2A * sinα * cosωt | After the code control process, the downstream / upstream side differential signal 43 Vi = ∫ | C * 2A * sinα * cosωt | dt Integrated signal 44

However, the detection of the phase difference as described above can be applied only when the amplitudes of the detection signals on the upstream side and the downstream side are exactly the same, and there is a problem that an error occurs if there is a difference in the amplitudes. This is shown below by a mathematical formula. Here, the meaning of each symbol is as follows. ω: Resonance frequency of vibration tube α: Phase difference generated by mass flow rate A: Amplitude of downstream detection signal B: Amplitude of upstream detection signal C: Amplification factor

(1) Output of the subtractor V _V V _V = A * sin (ωt + α) -B * sin (ωt-α) = 2A * sin α * cosωt- (BA) * sin (ωt-α) ( 2) The sign control signal V _S drive coil inflow current sin ωt is frequency-multiplied and co
A code control signal V _S that is sωt is created. If cosωt ≧ 0: V _S = + 1, if cosωt <0: V _S = −1 (3) Output of code controller Vc Vc∝V _S * C * {2A * sin α * cosωt− (B
-A) * sin (ωt-α)} (4) Output Vi of the integrating circuit However, the integration range is T1: 0 and T2: 4π.

As described above, the output Vi of the integrator has a difference due to different amplitudes (B ≠ A), as shown in the second term of the above equation, and this appears as an error of the integrated value. become. Figure 12 shows the vibration frequency of the vibration tube at 1KH.
An error represented by the following equation is shown, where z is the occurrence time difference is 2 μS. Where ∫ (B = A) is the integrated value when the amplitudes are equal, ∫
(B ≠ A) shows the integrated values when the amplitudes are different.

According to FIG. 12, when the amplitude difference is 1% (upstream / downstream amplitude ratio 101%), the error is 0.
It turns out that it will be 5%. This difference in amplitude is caused by the difference in initial sensitivity of the pickup coil, the difference in sensitivity aging characteristics, temperature dependence, etc. Therefore, this difference in amplitude is constantly monitored and corrected so that the amplitudes are the same. There is a need.

As this method, for example, as shown in FIG.
AGC (Automati) with variable amplification factor or gain
It can be reduced to a certain extent by providing a c Gain Control amplifier 35. However, a phase delay occurs in this AGC amplifier, and this phase delay is, for example, as shown in FIG.
As shown in Fig. 4, since it changes according to the change of the amplification factor, it is difficult to obtain the target performance of 0.01%.

[0016]

As described above in detail, the straight pipe type Coriolis flowmeter which has a small pressure loss and is easy to clean has a high rigidity of the pipe through which the fluid passes, and is more upstream than the U-shaped pipe. There is a problem that the time difference on the downstream side becomes small.
Therefore, a weak phase signal was obtained by subtracting the upstream pickup signal and the downstream pickup signal with a difference device (subtractor). In this method, the upstream pickup signal and the downstream pickup signal should be matched. If you try to use the AGC amplifier, the AGC amplifier causes a phase delay,
There is a problem that an error caused by this occurs and the measurement accuracy is reduced. Therefore, an object of the present invention is to improve the measurement accuracy by making it possible to match the pickup signals on the upstream side and the pickup signal on the downstream side even if the AGC amplifier is used, and to eliminate the error due to the phase delay.

[0017]

In order to solve such a problem, in the invention of claim 1, a fluid is caused to flow in a vibrating pipe, and the pipe is twisted by Coriolis force generated by the flow and angular vibration of the pipe. In a Coriolis mass flowmeter which vibrates and obtains a mass flow rate from the phase difference or time difference of the vibrations, a pair of detectors for detecting the asymmetric flexural vibration of the pipe, and an output side of at least one of the pair of detectors. A variable amplification factor amplifier connected to the amplifier for amplifying its output, an amplitude difference detection means for detecting an amplitude difference of the amplifier output, and an amplification rate of the amplifier controlled according to the output from the amplitude difference detection means. Amplitude control means for making the output amplitudes of the amplifiers coincide with each other, and phase / time difference measuring means for obtaining the phase difference or time difference between the amplifier output signals having the same amplitude. It is characterized in that the output from the phase and time difference measuring means are provided with correction means for correcting in accordance with the transmission delay of the amplifier.

In the invention of claim 1, the amplifier is
When provided for both of the pair of detectors, the amplification factor of one amplifier can be made variable according to the output from the amplitude difference detection means, and the amplification factor of the other amplifier can be fixed.
Alternatively, the correction means includes a voltage detection means for detecting an amplification factor control voltage of the amplifier, a frequency detection means for detecting an operating frequency of the amplifier, a temperature detection means for detecting an ambient temperature of the amplifier, and the amplifier. Gain control voltage of
Storage means for storing in advance the relationship between at least one of the frequency and the ambient temperature and the amount of transmission delay of the amplifier, and the stored contents are read out in accordance with the output from each of the detection means, and the phase is used by using the value. It can be composed of an arithmetic processing means for correcting the output value of the time difference measuring means.

[0019]

By providing the correction circuit for correcting the phase delay due to the gain of the AGC amplifier that equalizes the amplitude difference between the pickup signals on the upstream side and the downstream side, it is possible to measure the phase difference or the time with high accuracy and less error. . Further, by considering the installation mode of the AGC amplifier, the configuration method of the memory provided in the phase delay correction circuit, etc., the configuration can be simplified, the cost can be reduced, and the flexibility can be provided according to the usage mode. To

[0020]

1 is a block diagram showing an embodiment of the present invention. The following speed information is obtained by the electromagnetic pickups 11 and 12 provided on the upstream side and the downstream side of the vibrating pipe 15. Note that the meaning of each symbol is the same as above. (20) V _U = B * sin (ωt−α): upstream detection signal (21) V _D = A * sin (ωt + α): downstream detection signal

The outputs of the electromagnetic pickups 11 and 12 are guided to low pass filters (LPF) 50a and 50b,
A high frequency band component is removed from the resonance frequency of the vibration system to improve the S / N ratio of the signal. Next, the buffers 51a and 51b
Then the impedance is converted and sent to the next stage. The output signal of the buffer is expressed by the following equation. Note that βa and βb are L
The phase delay caused by the PFs 50a and 50b and the buffers 51a and 51b is shown. (70a) V _U = B * sin (ωt-α-βa) (70b) V _D = A * sin (ωt + α-βb)

The buffer outputs 70a and 70b are guided to amplifiers (AGC amplifiers) 52a and 52b with variable amplification factors. Here, the amplifier 52b amplifies with a fixed amplification factor, and the amplifier 52a amplifies with a variable amplification factor. The reason why the variable amplification factor amplifier is provided on each of the upstream side and the downstream side is that there is always a phase delay even when the amplification factor is 1, so that it is provided both upstream and downstream in order to compensate for this. At this time, the amplification factor on the fixed amplification side is set in consideration of the dynamic range of the AGC amplifier so that the amplitude of the electromagnetic pickup signal after AGC amplification is ½ of the maximum amplitude that can be output by the AGC amplifier.

On the variable amplification side, gain control is performed as follows so that the same amplitude as the output of the variable amplification factor amplifier 52a can be obtained. First, in the upstream / downstream vibration difference detection unit 58, the peak detectors 53a and 53b detect the peak voltages of the output signals from the variable amplification factor amplifiers 52a and 52b, and the subtraction circuit 54 determines the difference between these values. The difference is integrated by the integrating circuit 55. Amplification factor setting circuit 56
a and 56b are variable amplification factor amplifiers 52a and 52b, the electromagnetic pickup 11,
The amplitude of the 12 signal after AGC amplification is set to a voltage corresponding to the amplification factor D such that the amplitude of the 12 signals is 1/2 of the maximum outputtable amplitude of the variable amplification factor amplifiers 52a and 52b.

The voltage from the amplification factor setting circuit 56b is given to the variable amplification factor amplifier 52b, and the signal from the electromagnetic pickup 12 is amplified at a constant amplification factor. The voltage from the amplification factor setting circuit 56a is added by the adding circuit 57 to the integrating circuit 55.
The output is added to the variable gain amplifier 52a and is applied to the variable gain amplifier 52a to change the gain. Variable amplification factor amplifier 52
The outputs of a and 52b are as follows. (71a) V _U = D * A / B * B * sin (ωt-α-βa-γa) (71b) V _D = D * A * sin (ωt + α-βb-γb)

The meanings of the signals shown at 71a and 71b are as follows. D * A / B: Upstream side Variable amplification side amplification rate D: Downstream side Fixed amplification side amplification rate γa: Upstream side Variable amplification side phase delay amount γb: Downstream side Fixed amplification side phase delay amount βa: Upstream side Phase delay amount of LPF and buffer βb: Phase delay amount of downstream LPF and buffer

Γa and γb represent phase delays due to the AGC circuit, and if these are fixed (that is, only βa and βb), they can be treated as simple offset amounts. However, since such a phase delay changes depending on the amplification factor, the ambient temperature, the frequency, etc. as shown in FIGS. 2 and 3, it cannot be simply compensated. In addition, in FIG. 2, the measurement frequencies are 560 Hz, 760 Hz, and 960, respectively.
Fig. 3 shows that the temperature of the constant temperature bath is 20 degrees and 40 degrees.
This is a plot of the amount of phase delay when degrees and 60 degrees are plotted with different amplification factors (“□”, “Δ”, “x” marks, etc.). Although FIGS. 2 and 3 are not necessarily clear, it is sufficient to understand that the amount of phase delay by the AGC circuit is a function of the amplification factor, the ambient temperature, the frequency, and the like.

Therefore, variable amplification factor amplifiers 52a and 52b are provided.
The ROM table 64 for storing the propagation delay amount using the amplification factor setting voltage, ambient temperature, frequency, etc. as parameters is prepared in advance, and the amplification factor setting voltages of the amplification factor variable amplifiers 52a and 52b are set at the time of operation of the flowmeter. A / D converter 62
A / D conversion is performed by a and 62b and the signals 73a and 73b are taken into the arithmetic processing unit 61, and the ambient temperature of the amplification factor variable amplifiers 52a and 52b is used as the signal 74 by the temperature sensor 63, and the amplification factor variable amplifier 52a. The output frequency is captured as the signal 75. As a result, the arithmetic processing unit 61 causes the acquired signals 73a, 73
The ROM table 64 is referred to in accordance with b, 74 and 75 (interpolation calculation is also performed if necessary), and the propagation delay amounts γa and γb of the variable amplification factor amplifiers 52a and 52b are obtained.

On the other hand, the phase difference between the signals 71a and 71b obtained by AGC amplification to match the signal amplitudes on the upstream side and the downstream side is
Since it is obtained by the phase difference measuring unit 59 and given to the arithmetic processing unit 61 as a signal 72 shown by the following equation, (72) D * C * A {2α + (βb−βa) + (γb−γa)} Γb− obtained as described above from 72.
By subtracting γa, the phase delay can be corrected. The measurement of the phase difference by the phase difference measuring unit 59 is performed in the same manner as in the conventional example shown in FIG. Further, it goes without saying that the phase difference can be measured as a time difference.

From the above, the corrected phase difference becomes D * C * A {2α + (βb-βa)}. In this equation, βb−βa is a substantially constant value and can be corrected as an offset amount. Therefore, the final phase difference is given by the following equation. D * C * A * 2α

In the above description, the variable amplification factor amplifier is provided on both the upstream side and the downstream side, but it goes without saying that it may be provided on only one of them. Further, in the ROM table 64, the phase delay amount is a function of the amplification factor control voltage, the ambient temperature and the frequency, but it goes without saying that it can be a function of at least one of the amplification factor, the ambient temperature and the frequency.

[0031]

According to the present invention, when measuring the phase difference between the pickup signals on the upstream side and the downstream side,
A for matching the amplitude of the pickup signal on the downstream side
Since the GC amplifier and the correction circuit for correcting the phase delay due to the amplification factor of the amplifier are provided, the error due to the amplitude difference between the pickup signals on the upstream side and the downstream side is reduced, and the phase difference between the two signals is accurately measured. The advantage is that better measurements can be made. Further, by considering the installation mode of the AGC amplifier or the configuration method of the memory provided in the phase delay correction circuit, the configuration can be simplified, the cost can be reduced, and the flexibility can be provided according to the usage mode. .

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a characteristic diagram showing a phase delay characteristic of a variable gain amplifier with temperature.

FIG. 3 is a characteristic diagram showing a phase delay characteristic according to a frequency of a variable amplification factor amplifier.

FIG. 4 is a principle configuration diagram of a Coriolis mass flowmeter.

FIG. 5 is an explanatory diagram for explaining a vibration state of a U-shaped pipe.

FIG. 6 is an explanatory diagram of a vibration mode generated by the torsion torque of the Coriolis force of the U-shaped pipe.

FIG. 7 is a waveform diagram showing an example of a pickup signal when Coriolis force is generated in the U-shaped pipe.

FIG. 8 is a configuration diagram showing a phase difference detection circuit using a counter method.

FIG. 9 is a structural diagram showing an example of a straight tube type Coriolis mass flowmeter.

FIG. 10 is a configuration diagram showing a conventional example of a differential phase difference detection circuit.

FIG. 11 is a waveform diagram for explaining operation waveforms of respective parts of FIG.

FIG. 12 is an explanatory diagram for explaining a measurement error in the case of FIG.

13 is a configuration diagram showing a modified example of FIG.

14 is a graph showing a delay of the AGC amplifier shown in FIG.

[Explanation of symbols]

1 ... U-shaped pipe, 2 ... Magnet, 3 ... Base, 4 ... Electromagnetic drive coil, 5 ... Support frame, 11, 12 ... Electromagnetic pickup, 1
5 ... Straight pipe, 22, 29 ... Amplifier, 23, 34 ... Comparator, 24 ... Exclusive OR circuit, 26 ... Counter, 27 ...
Reference clock, 28 ... Differentiator, 30 ... Sign controller, 31
... Frequency multiplier, 32, 55 ... Integrator circuit, 33 ... Constant current circuit, 35, 52a, 52b ... AGC amplifier, 50a,
50b ... Low-pass filter (LPF), 51a, 51b
... buffer, 53a, 53b ... peak detector, 54 ...
Subtraction circuit, 56a, 56b ... Amplification factor setting circuit, 57 ... Addition circuit, 58 ... Upstream / downstream vibration amplitude difference detection unit, 59 ... Amplitude control circuit, 60 ... Phase difference measurement unit, 61 ... Arithmetic processing device,
62a, 62b ... A / D converter, 63 ... Temperature sensor, 6
4 ... ROM table.

Claims (3)

[Claims] 1. A Coriolis mass flowmeter for causing a fluid to flow in an oscillating pipe, twisting and vibrating the pipe by a Coriolis force generated by the angular vibration of the flow and the pipe, and obtaining a mass flow rate from a phase difference or a time difference of the vibration. , A pair of detectors for detecting the asymmetric flexural vibration of the pipe, a variable amplification factor amplifier connected to at least one output side of the pair of detectors to amplify its output, and an amplitude of the amplifier output. Amplitude difference detection means for detecting a difference and amplitude control means for matching the output amplitudes of the amplifiers by controlling the amplification factor of the amplifier according to the output from the amplitude difference detection means, and the same amplitude control means. Phase / time difference measuring means for obtaining the phase difference or time difference between the amplifier output signals, and the output from the phase / time difference measuring means is compensated according to the transmission delay amount of the amplifier. Coriolis mass flowmeter, characterized in that a correcting means for. 2. When the amplifier is provided for both the pair of detectors, the amplification factor of one amplifier is made variable according to the output from the amplitude difference detection means, and the amplification factor of the other amplifier is set. Is fixed, the Coriolis mass flowmeter according to claim 1. 3. The correcting means includes voltage detecting means for detecting an amplification factor control voltage of the amplifier, frequency detecting means for detecting an operating frequency of the amplifier, and temperature detecting means for detecting an ambient temperature of the amplifier. Storing means for storing in advance the relationship between at least one of the amplification factor control voltage of the amplifier, the frequency and the ambient temperature and the amount of transmission delay of the amplifier, and the stored contents according to the output from each of the detecting means. 2. The Coriolis mass flowmeter according to claim 1, wherein the Coriolis mass flowmeter comprises a read-out unit and an arithmetic processing unit for correcting the output value of the phase / time difference measuring unit using the read value.