JPH0949749A - Electromagnetic flowmeter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the influence of flow noise even when a signal is small like a battery-driven electromagnetic flowmeter having small power consumption by responding to a fast flowing change, conducting power conservation, increasing periods of sampling the signal and making the noise uniform. SOLUTION: After the small period of exciting adjacent to positive and negative exciting periods is repeated three times, a non-exciting period T7 is provided as the former half period Ta of the large period of the exciting. Then, after the small period adjacent to negative and positive exciting periods is repeated three times, a non-exciting period T14 is provided as the latter half period Tb of the large period of the exciting. The sum of the differentiated noise E1n and the flow proportional signal E1s is output at each small period. Differentiated noises Vb', Vb are output during the non-exciting period, subtracted from the sum, and the differentiated noise is removed to calculate a flow proportional signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an improvement of a rectangular wave excitation type electromagnetic flowmeter.

[0002]

2. Description of the Related Art It is known that in an electromagnetic flow meter using a rectangular wave excitation method, a zero point variation occurs due to differential noise generated when switching excitation.

For this reason, like the electromagnetic flowmeters described in JP-A-53-75966 and JP-A-57-149919, there are four periods of positive, no, negative and no excitation as one cycle. A value excitation method is known in which differential noise generated during a non-excitation period cancels differential noise during a positive or negative excitation period to reduce and improve a zero point variation.

[0004]

Although the zero-point fluctuation is improved by the above method, the flow rate signal is measured only in the two periods of positive and negative excitation out of four periods of one cycle of excitation, and in the case of no excitation. Two
The period is used only to detect differential noise, not to detect flow rate.

The problems for this are: The response of the flow rate measurement to the rapid change of the flow rate becomes slow. . The period for sampling the signal is 1/2 of that of the normal binary excitation method without the non-excitation period, and the averaging of the flow noise included in the flow rate signal becomes small and the variation becomes large.

[0006] When the three-valued excitation method as in the prior art is used, the excitation frequency is substantially 1/2 of that of the two-valued excitation method.
And becomes weaker with respect to flow noise having a 1 / f characteristic, which becomes larger as the frequency becomes lower. In particular, a low power consumption type electromagnetic flowmeter such as a battery drive inevitably has a low flow rate signal level, so it is easily affected by the above flow noise, and a three-valued excitation method allows the excitation frequency to be a two-valued excitation method. It becomes a serious defect because the merit that the zero fluctuation is improved disappears.

Therefore, an object of the present invention is to provide an electromagnetic flow meter which can solve these problems.

[0008]

In order to achieve the above object, the invention of claim 1 provides a positive excitation period (T1, T3, T5) and a negative excitation in a rectangular wave type electromagnetic flow meter. Short period of positive / negative binary excitation (T1 + T2, T3 + T4, T5 + T6) for adjacent periods (T1 + T2, T3 + T4, T5 + T6) adjacent to each one of the periods (T2, T4, T6)
As described above, after repeating this small cycle a plurality of times, a non-excitation period (T7) corresponding to a half cycle of the small cycle is provided, and a plurality of the small cycles (T1 + T2, T3 + T4, T5 + T6) and the non-excitation period ( The total of T7) is the first half period (T
a) and the negative excitation period (T8, T10, T
12) and one of the positive excitation periods (T9, T11, T13) adjacent to each other (T8 + T9, T10 + T11, T).
12 + T13) is a short cycle of negative / positive binary excitation (T8 + T)
9, T10 + T11, T12 + T13), after repeating this small cycle a plurality of times, a non-excitation period (T14) corresponding to a half cycle of the small cycle is provided, and the small cycle (T8
+ T9, T10 + T11, T12 + T13) a plurality of times and the non-excitation period (T14), the total of the latter half period (T).
As b), the first half cycle (Ta) and the term half cycle (Tb)
Of the total period (Ta + Tb) of the exciting period (Ta + T
b), and the flow rate signals (Va1, Va2, Va3, Va4, Va) including differential noise for each small cycle of binary excitation in the first half cycle (Ta) and the second half cycle (Tb) of the large cycle.
5, Va6), and the non-excitation period (T
7) Every non-excitation period (T14) or before the large cycle (T14)
14 ') signal (Vb) (V) corresponding to the differential noise
b ′) is taken out and the flow rate signal (Va1, Va2, Va3, Va4, Va5, Va6) including the differential noise is taken out.
And an electromagnetic flowmeter which extracts a flow rate signal from which differential noise is removed based on the signals (Vb) and (Vb ') corresponding to the differential noise for each non-excitation period (T7) (T14) (T14'). Is.

According to a second aspect of the present invention, in the electromagnetic flowmeter according to the first aspect, the moving average value of the differential noise extracted for each of a plurality of non-excitation periods is a flow rate signal (Va1, for each small period).
Va2, Va3, Va4, Va5, Va6) and subtracts the differential noise to extract the flow rate signal.

The invention of claim 3 relates to claim 1 or 2.
The residual magnetic excitation method is used in the above electromagnetic flowmeter.

[0011]

[Function] Each small cycle (T1 + T2, T3 + T4, T5 + T
6, T8 + T9, T10 + T11, T12 + T13) from the flow rate signal output (Va1, Va2, ..., Va6)
The differential noise (Vb ') obtained during the non-excitation period in the immediately preceding large cycle is subtracted to obtain the flow rate signal output (Va1, Va
2, ..., Va6) to remove the differential noise, only the flow rate proportional signal is taken out.

Since the electrochemical direct current noise does not change in a short period of time, it is canceled and canceled when the signal (Va ₀ ) is taken out from the difference of each half cycle between each small cycle.

Similarly, in the differential noise (Vb ₀ ), the electrochemical direct current noise is removed and not included. In the invention of claim 2, since the moving average value of the differential noise is further used,
Therefore, the accuracy of the flow rate proportional signal is improved.

Further, in the invention of claim 3, the exciting power is further reduced, which is useful for power saving of the electromagnetic flowmeter.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION

[First Embodiment] FIG. 1 is a block diagram showing a first embodiment of the present invention, and FIG. 2 is a diagram showing the timing of sono operation and the waveform of each part.

When an exciting current having the waveform of FIG. 2A is passed through the coil 3 to generate a magnetic field in the flow tube 1, two electrodes 2a are formed.
A flow rate proportional signal having a waveform similar to the excitation waveform of FIG. 2 (a) is generated between −2b and amplified by the differential amplifier 4 to become E1s. An excitation circuit 7 is controlled by the timing circuit 8.

Differential noise is generated between the electrodes 2a and 2b between the electrodes 2a and 2b irrespective of the flow rate due to switching of excitation, and the differential amplifier 4
The waveform E1m is amplified by the waveform E1m shown in FIG. In reality, the output of the differential amplifier 4 has a waveform E ₁ that is a combination (addition) of E1s and E1m, but for the sake of easy understanding, it is independently shown in FIG. 2 (a) and FIG. 2 (b).
s) and E1m.

The output obtained by sampling the combined signal, E1 = E1s + E1n, at the timing of the switches S1 and S2 by the sample and hold circuit 5 is shown in FIG. 2C, the solid line is the output Va of the integrator 52, and the shaded portion is the switch. The value Va _{0 held} by the capacitor C2 and the amplifier 53 having the amplification factor of 1 at the timing of S4 is shown by a dotted line.

Further, 51 is an inverting amplifier having an amplification degree of -1,
The amplification degree of the integrator 52, which is determined by the circuit constants of the resistors R1 and R2 and the capacitor C1 and the integration time when the switches S1 and S2 are turned on, will be described as 1.

The integrator 52 integrates the respective values E and N of the latter half of the periods T1, T2, ..., T14 of the waveforms of FIGS. ). In the first small half cycle T1, the preceding small half cycle T14 '
Is a non-excitation period, and the differential noise is N / 2, which is half that in the period following the normal excitation period, as shown in FIG.
Therefore, the integrator 52 in the first short cycle T1 + T2
Output Va1 of the above becomes Va1 = 2E + 3 / 2N.

This differential value Va1 is held in the capacitor C2 at the timing of the switch S4, and then the switch S4.
It is reset at the timing of 3. Next small cycle T3 + T
4 and T5 + T6 are similarly integrated and sample-held, but since there is no non-excitation period immediately before these small cycles, the sample value Va2 of each small cycle T3 + T4 and the sample value Va3 of each small cycle T5 + T6 are Va2 = Va3 = 2E + 2N.

In the last non-excitation period T7 of the period T1 to T7, that is, the first half period Ta of the large cycle, Ta = T1 + T2 + ... + T7, the same configuration as the sample hold circuit 5 is set at the timing of the switch S6. The output E1 of the differential amplifier 4 is sampled by another sample hold circuit 6.

Since the period T7 is a non-excitation period, there is no flow rate proportional signal E1s, and only the differential noise E1n is sampled, and the value becomes N / 2, which is Ta in the first half period Ta of the large cycle.
Ends. As described above, T1 + T2 and the like are called small cycles, and half the period T1 and the like are called small and half cycles. The period represented by T1 + T2 ... + T14 = Ta + Tb is called a large cycle and is called a large cycle. However, it is not a technical term that is commonly used. However, expressions such as half cycle, small half cycle, and large cycle are used as they are considered appropriate for explaining the present invention.

The second half period Tb of the large cycle, Tb = T8 + T
9 + ... + T14, the same operation as above is performed, and the period T
The sample value Va ₄ Va4 = 2E + (3/2) N of the small cycle of 8 + T9 and the sample values Va ₅ and Va ₆ of the small cycles of the periods T11 + T12 and T13 + T14 are sample-held, respectively, Va5 = Va6 = 2E + 2N. However, since the excitation timing is inverted as shown in FIG. 2A as compared with the period Ta in the first half of the large cycle, the sampling timing by the switches S5 and S6 is the switch S in the period Ta.
The relationship is opposite to the sampling timing of 1 and S2.

The last period T1 of the latter half period Tb of the large cycle
Reference numeral 4 denotes a non-excitation period, and the switch S5 is set to the value Vb = N / 2 sampled at the timing of the switch S6 in the period T7.
At this timing, differential noise having a polarity opposite to that of the period T7 is differentially sampled and Vb =
N. This value Vb is held in the capacitor C4 at the timing of the switch S8 and becomes the output Vb ₀ of the amplifier 63 having the amplification factor of 1.

From the above, one time T of the large sampling cycle
End 1 + T2 + ... + T14. In the first half of the above period T
The period a and the latter half period Tb constitute a so-called first half period Ta and second half period Tb of the large period Ta + Tb.

The output Va0 of the sample and hold circuit 5 shown by the dotted line in FIG. 2C is a value obtained by sampling the sum E1 of the flow rate proportional signal E1s and the differential noise E1m. The value of the output Vb0 of the sample hold circuit 6 is subtracted.

However, since Va1 = Va4 = 2E + (3/2) N Va2 = Va3 = Va5 = Va6 = 2E + 2N, the output Vb0 of the sample hold circuit 6 is passed through the amplifier 11 to the negative input of the differential amplifier 10. There is.

The amplification degree of the variable amplifier 11 is 3/2 times when the signal S9 of the timing circuit 8 is at the H level and is doubled when the signal S9 of the timing circuit 8 is at the L level.
The differential noise is offset and removed from the output V0 of the output V0 of the output V0 of V0 = 2E.

This value V0 corresponds to each small cycle T1 + T2, T
3 + T4, T5 + T6, T8 + T9, T10 + T11,
It is output for every T12 + T13, ...
It has the same response speed as an electromagnetic flowmeter with value excitation.

Although not specifically shown in the timing chart of FIG. 2, the electrochemical direct current noise generated between the electrodes 2a and 2b is the sample value Va1 of the sample hold circuit 5.
Va2, ..., Va6 ... by a difference for each half cycle, for example, a period T1
And T2 are obtained from the difference between the outputs E1 of the differential amplifier 4 and the like, so that they are canceled and eliminated at every small cycle. Similarly, when the sample value Vb ₀ of the sample hold circuit 6 is obtained, the electrochemical direct current noise is removed.

[Second Embodiment] FIG. 3 is a block diagram of a second embodiment of the present invention. A sampling circuit 5A comprises the sample-hold circuit 5 in FIG. 1 including a switch S4, a capacitor C2 and an amplifier 53 having an amplification factor of 1. The hold portion is removed, and the output V01 of this is removed from the A / D converter 20.
After being converted into a digital amount by, the differential noise is arithmetically processed by the MPU (microprocessor) 21.

The MPU 20 outputs operation signals to the excitation circuit 7 and the switches S1, S2 and S3, respectively, and outputs the analog output V01 of the sampling circuit 5A to A /
The D converter 20 outputs a control signal P1 which is converted into a digital amount and is taken into the MPU 21. In addition, MPU is A / D
Of the digital amount taken in from the converter 20, the differential noise is canceled and removed, and only the flow rate proportional signal is calculated.

FIG. 4 is a diagram showing the timing and the waveform of each part of the operation of the second embodiment of FIG. 3, and FIGS. 4 (a) and 4 (b).
Is exactly the same as in FIGS. 2A and 2B, and FIG. 4E shows the output V01 of the sampling circuit 5A. Switch S
After the output E1 of the differential amplifier 4 is sampled at the timings of S1 and S2, the output V01 of the sampling circuit 5A is converted to an A / D converter at the timing of the control signal P1 during the hold period in which all the switches S1, S2 and S3 are turned off. 20 in A
/ D conversion.

As a result, the output of the sampling circuit 5A in the short cycle T1 + T2 of the first half cycle Ta of the large cycle is V
a1, sampling circuit 5A in short cycle T3 + T4
Output is Va2, and the output of the sampling circuit 5A in the short cycle T5 + T6 is Va3.

Similarly, in the latter half period Tb of the large period, the sampling values Va4, Va5 and Va6 are set to the small periods T8 + T9,
The output of the sampling circuit 5A at T10 + T11 and T12 + T13 is A / D converted and converted into a digital amount. These include the flow rate proportional signal and differential noise.

These respective values are the same as in the first embodiment, the sampling values Va1 and Va4 are Va1 = Va4 = 2E + (3/2) N, and the sampling values Va2, Va3, Va5, Va.
6 is Va2 = Va3 = Va5 = Va6 = 2E + 2N. Further, the A / D converted values Vb1 and Vb2 of the output of the sampling circuit 5A in the periods T7 and T14 are Vb1 =
Vb2 = N / 2. Therefore, the following calculation is performed in the MPU 21, and the outputs E01, E02, ..., E06 of only the flow rate proportional signals excluding the differential noise in the non-excitation period are extracted.

The output of the first small cycle T1 + T2 is E01 = Va1- (3/2) (Vb'1 + Vb'2) The output of the second small cycle T3 + T4 is E02 = Va2-2 (Vb'1 + Vb'2) The third small The output of the cycle T5 + T6 is E03 = Va3-2 (Vb′1 + Vb′2) The output of the fourth short cycle T8 + T9 is E04 = Va4- (3/2) · (Vb′1 + Vb′2) The output of the fifth short cycle T10 + T11 is E05 = Va5-2 (Vb'1 + Vb'2) The output of the sixth short cycle T12 + T13 is E06 = Va6-2 (Vb'1 + Vb'2).

It should be noted that Vb'1 and Vb'2 have a large cycle Ta + T.
The output value of the sampling circuit 5A during the non-excitation period of the first half cycle and the second half cycle of the large cycle immediately before b is the output Vb.
1 and Vb2, and these correspond to differential noise in the non-excitation period. Only Vb'2 is shown in FIG. 4 (e),
Vb'1 is not represented.

E01 to E06 are flow rate proportional signals from which differential noise is removed. These values are one cycle T1 of each small cycle.
+ T2, T3 + T4, T5 + T6, T8 + T9, T10
+ T11, T12 + T13, ... Is output for each D / A
It is converted into an analog amount by the converter 22 and output as a flow rate of 4 to 20 mA, or is calculated as it is by the MPU 21 and displayed as an integrated flow rate value or an instantaneous flow rate value on a liquid crystal display or the like not shown. [Third Embodiment] FIG. 5 is a third embodiment of the present invention, in which the A / D converter 20 of the second embodiment of FIG. 3 is replaced with a comparator 30 and a reference voltage V is applied to the integrator 52B of the sampling circuit 5B.
A double integration function of inversely integrating e through a resistor Re and a switch Se is provided.

This embodiment does not use the A / D converter 20 which generally consumes a large amount of power as in the second embodiment,
The active parts of all electronic circuits are composed of elements capable of reducing power consumption, such as C-MOS elements. 4 for MPU31
Bit or 8-bit low power consumption type C-MOS
IC is used.

The switches S1, S2, S3 and Se are also C-
By using a MOS IC, these power consumptions can be reduced to substantially zero. Further, the UP amplifiers constituting the amplifiers 51 and 52B and the comparator 30 can also be operated at about 2 to 10 μA.

This embodiment is intended to realize a two-wire type electromagnetic flowmeter or a battery-driven electromagnetic flowmeter which is required to save power. In the explanation of FIG. 5, a general excitation method is used. However, this is an embodiment that exhibits the most power saving effect when combined with the residual magnetic excitation method.

The method for creating a non-excitation period by the residual magnetic excitation method is described in Japanese Patent Laid-Open No. 60-242318 and will not be described. In the third embodiment of FIG. 5, the timings of the switches S1 and S2 are exactly the same as those of the timing chart of FIG. 4, so that in the timing chart of FIG. Only showing.

The switch Se is set to MP after the sampling is completed.
It is turned on by the timing signal output from U31, and the integrator 5
2B output signal, that is, the output V02 of the sampling circuit 5B
When the value of becomes zero and the comparator 30 is inverted, the MPU 31 detects this and turns off.

By counting the clock with the counter built in the MPU 31 during the period τ during which the switch Se is on, the sample value in the sampling circuit 5B is converted into a digital value.

The counter may be externally attached to the MPU 31. The switch S3 is an integrator 52
This is a switch for discharging the electric charge charged in the B capacitor C1.

When the low power consumption type comparator 30 is used, since it takes time for the switch Se to turn off after the output of the integrator 52B becomes zero, the output of the integrator after the A / D conversion is completed. Switch S3 is useful for zeroing.

In particular, when the input becomes too large and the maximum time of inverse integration is exceeded, the switch S3 is indispensable for resetting with the switch S3 and starting the sampling operation in the next small cycle from zero.

7 (f), Se and S3 are A / D of the third embodiment of FIG. 5 in the non-excitation period of the period T7 or T14.
Only the conversion operation is shown in (f), Se, and S of FIG.
Similar to 3.

The A / D conversion operation in FIGS. 6 and 7 is converted into a digital value by counting with an internal or external counter in the MPU 31, as shown in FIG.
A general A with large power saving as in the second embodiment of FIG.
The / D converter 20 is not used.

The A / D-converted values of each small period are arithmetically processed by the MPU 31 just as in the case of the second embodiment, and only the flow rate proportional signal excluding the differential noise is extracted. Since the third embodiment of FIG. 5 is intended for battery operation, the digital amount of output is large in power saving.
No D / A conversion is performed by the D / A converter 22 and the liquid crystal display or the like displays the instantaneous flow rate value or the integrated flow rate value.

In the above embodiment, the differential noise Vb is subtracted from the integrated values Va1 to Va6 including the sampled differential noise to extract only the flow rate proportional signal. However, the value of the differential noise Vb is once in each large cycle. The measured values may be used, or the moving average values of about 10 times obtained before each large cycle may be used.

When the moving average value is used as the value of the differential noise Vb a plurality of times in this way, in the present invention in which the number of non-excitation periods is smaller than the number of excitation periods, in the non-excitation period for which the value of the differential noise Vb is obtained. Since the accuracy of the moving average value can be increased by increasing the number, there is an advantage that the measurement accuracy of the flowmeter can be improved as a result (the invention of claim 2).

Further, in each of the above embodiments, the number of small cycles in the first half and the latter half of the large cycle is three, but it is possible to set it to two or more times. Fluctuations in the magnitude of the differential noise occur due to fluctuations in the degree of deposits on the electrodes, fluctuations in the composition of the fluid, and fluctuations in the concentration due to chemical injection, but these fluctuations are gradual in time, and every small cycle of excitation. This is because it is not necessary to make a non-excitation period each time to compensate.

In particular, when the residual magnetic excitation method is combined with the present invention (the invention of claim 3), there is no sudden change in the flow rate in a normal stable state, and the conductivity of the fluid is kept substantially constant, Since there is no sudden change in conductivity, the change in differential noise occurs slowly in time, so the interval for placing the non-excitation period may be long, and only once in a small cycle of 10 to 100 times is required. In the case where occurs rapidly, it is preferable to make the interval variable so that one non-excitation period is placed every 2 to 5 small cycles.

As a method of determining the continuous length of the binary excitation period as an appropriate value, the value of the differential noise Vb for each time is stored in the memory in the microcomputer about 10 to 100 times in the past, and the latest value is stored. It can be judged from the degree of temporal change in comparison with the value of the differential noise Vb.

The judgment of the degree of this temporal change is not made based on the value of the latest differential noise Vb only once and the value before that, but the moving average value of the differential noise Vb of the latest 5 times is calculated. By comparing with the moving average value of the past several tens to several hundreds of times before that, more stable judgment can be performed with high accuracy.

Further, the number of small cycles within one large cycle may be set in advance according to the liquid and the usage conditions. For example, it can be set to be large for tap water and small for chemical liquids having large fluctuations in conductivity.

[0060]

【The invention's effect】

(1) Electrochemical direct current noise can be removed as usual, and by subtracting the differential noise during the non-excitation period, the differential noise can be removed by positive and negative binary excitation during normal repetition of small-cycle excitation. , The stability of the zero point can be improved without sacrificing the output response unlike the conventional three-valued excitation.

(2) Since the operation is the same as that of the conventional binary excitation method while the small-cycle binary excitation is repeated, the excitation frequency is 1/2 of that of the conventional three-valued excitation method, and the 1 / f characteristic is obtained. It is not necessary to sacrifice the characteristics against flow noise having a noise as in the case of the conventional three-valued excitation method. Therefore, the flow noise characteristic does not deteriorate even in the case of an electromagnetic flowmeter with a small signal such as a battery-driven low power consumption type electromagnetic flowmeter.

(3) In the invention of claim 2, in addition to the effects of (1) and (2) above, the substantial accuracy of the value of the differential noise used for compensation can be improved, and the accuracy of flow rate measurement can be improved accordingly. Get better.

(4) Since the normal operation during the repetition of the small period excluding the non-excitation period is the binary excitation method, the invention of claim 3 particularly has the effect of the invention of claim 1 or 2 It is possible to maximize the power saving effect of magnetic excitation and realize a battery-operated electromagnetic flowmeter with minimal power consumption.

[Brief description of drawings]

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

2 is a diagram showing the timing and waveforms of the first embodiment of FIG. 1, where (a) is the flow rate proportional signal E1s of the exciting current and the output E1 of the differential amplifier 4, and (b) is the difference amplifier 4; Output E
The differential noise E1n, (c) of 1 is the output Va of the sampling circuit 52 and the output Va0 of the hold circuit, (d).
Is a diagram showing the differential noise sampling output Vb and hold outputs Vb0, S1, ..., S9 during the non-excitation period, showing the on / off timings of the switches S1 ,.

FIG. 3 is a block diagram of a second embodiment of the present invention.

FIG. 4 is a diagram showing the timing and waveforms of the second embodiment of FIG. 3, where (a) is the flow rate proportional signal E1s of the exciting current and the output E1 of the differential amplifier 4, and (b) is the differential amplifier. The differential noise E1n and (e) of the output E1 of FIG. 4 are the outputs V01 of the sampling circuit 5A, and S1, S2 and S3 are the switches S.
A diagram showing the on / off timing of 1, S2, S3, P
1 is a diagram showing the timing of the control signal P1 from the MPU 21.

FIG. 5 is a block diagram of a third embodiment of the present invention.

FIG. 6 is a diagram showing timings and waveforms in a short period of excitation of the third embodiment of FIG. 5, (f) is an output V02 of the integrator 52B, Se and S3 are on / off timings of switches Se and S3. , (G) are diagrams showing clocks counted inside the MPU 31.

7 is a diagram showing timings and waveforms in the non-excitation period of the third embodiment of FIG. 5, where (f), Se, and S3 correspond to (f), Se, and S3 of FIG. 6, respectively.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Flow tube 2a, 2b Electrode 3 Excitation coil 4 Differential amplifier 5 Sample hold circuit 5A, 5B Sampling circuit 6 Sample hold circuit 7 Excitation circuit 8 Timing circuit 10 Differential amplifier 11 Variable amplifier 20 A / D converter 21, 31 MPU 30 Comparators 52, 52A, 52B, 62 Integrators T1, T2, T3, T4, T5, T6, T8, T9, T
10, T11, T12, T13, excitation period T1 + T2, T3 + T4, T5 + T6, T8 + T9, T
10 + T11, T12 + T13 Small cycle T7, T14, T14 'Non-excitation period Ta First half cycle Tb Second half cycle Ta + Tb Large cycle Va1, Va2, Va3, Va4, Va5, Va6
Flow rate signal including differential noise Vb, Vb 'Signal corresponding to differential noise

[Procedure amendment]

[Submission date] September 8, 1995

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0008

[Correction method] Change

[Correction contents]

[0008]

In order to achieve the above object, the invention of claim 1 provides a positive excitation period (T1, T3, T5) and a negative excitation in a rectangular wave type electromagnetic flow meter. Short period of positive / negative binary excitation (T1 + T2, T3 + T4, T5 + T6) for adjacent periods (T1 + T2, T3 + T4, T5 + T6) adjacent to each one of the periods (T2, T4, T6)
As described above, after repeating this small cycle a plurality of times, a non-excitation period (T7) corresponding to a half cycle of the small cycle is provided, and a plurality of the small cycles (T1 + T2, T3 + T4, T5 + T6) and the non-excitation period ( The total of T7) is the first half period (T
a) and the negative excitation period (T8, T10, T
12) and one of the positive excitation periods (T9, T11, T13) adjacent to each other (T8 + T9, T10 + T11, T).
12 + T13) is a short cycle of negative / positive binary excitation (T8 + T)
9, T10 + T11, T12 + T13), after repeating this small cycle a plurality of times, a non-excitation period (T14) corresponding to a half cycle of the small cycle is provided, and the small cycle (T8
+ T9, T10 + T11, T12 + T13) a plurality of times and the non-excitation period (T14), the total of the latter half period (T).
As b), a half cycle after a first half period (Ta) (Tb)
Of the total period (Ta + Tb) of the exciting period (Ta + T
b), and the flow rate signals (Va1, Va2, Va3, Va4, Va) including differential noise for each small cycle of binary excitation in the first half cycle (Ta) and the second half cycle (Tb) of the large cycle.
5, Va6), and the non-excitation period (T
7) Every non-excitation period (T14) or before the large cycle (T14)
14 ') signal (Vb) (V) corresponding to the differential noise
b ′) is taken out and the flow rate signal (Va1, Va2, Va3, Va4, Va5, Va6) including the differential noise is taken out.
And an electromagnetic flowmeter which extracts a flow rate signal from which differential noise is removed based on the signals (Vb) and (Vb ') corresponding to the differential noise for each non-excitation period (T7) (T14) (T14'). Is.

[Procedure 3]

[Document name to be amended] Statement

[Correction target item name] 0015

[Correction method] Change

[Correction contents]

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [First Embodiment] FIG. 1 is a block diagram showing a first embodiment of the present invention, and FIG. 2 is a diagram showing the timing of its operation and the waveform of each part.

Claims (3)

[Claims] 1. A rectangular wave excitation type electromagnetic flow meter, wherein a positive excitation period (T1, T3, T5) and a negative excitation period (T2, T4, T6) are adjacent to each other (T1).
+ T2, T3 + T4, T5 + T6) is defined as a small cycle (T1 + T2, T3 + T4, T5 + T6) of positive / negative binary excitation, and after this small cycle is repeated a plurality of times, a non-excitation period (T7) corresponding to a half cycle of the small cycle. By providing a plurality of small cycles (T1 + T2, T3 + T4, T5 + T6) and the non-excitation period (T7), the first half cycle (Ta)
In addition, the negative excitation period (T8, T10, T12) and the positive excitation period (T9, T11, T13) are adjacent to each other (T8 + T9, T10 + T11, T12 + T13) for negative / positive binary excitation. Small cycle (T8 + T9, T10 + T11,
T12 + T13), after repeating this small cycle a plurality of times, the non-excitation period (T1) corresponding to a half cycle of the small cycle.
4) is provided, and the small cycles (T8 + T9, T10 + T1) are provided.
1, T12 + T13) and the non-excitation period (T1
4) The total of the latter half period (Tb) is defined as the latter half period (Tb), and the total period (Ta + Tb) of the first half period (Ta) and the term half period (Tb) is defined as a large period of excitation (Ta + Tb). Ta) and flow rate signals (Va1, Va2, Va3, Va4, Va5, Va6) including differential noise for each small period of binary excitation in the second half period (Tb).
And the signals (Vb) and (Vb ′) corresponding to the differential noise are extracted for each non-excitation period (T7) (T14) or for each non-excitation period (T14 ′) before the large cycle, and the differential noise Flow rate signal (Va1, Va2, Va
3, Va4, Va5, Va6) and the non-excitation period (T
7) An electromagnetic flow meter, which extracts a flow rate signal from which differential noise is removed based on signals (Vb) and (Vb ') corresponding to differential noise for each of (T14) and (T14'). 2. A moving average value of the differential noise extracted for each of a plurality of non-excitation periods is used as a flow rate signal (Va) for each small period.
1, Va2, Va3, Va4, Va4, Va5, Va6) to extract a flow rate signal from which differential noise has been removed to obtain an electromagnetic flowmeter. 3. The electromagnetic flowmeter according to claim 1 or 2, wherein a residual magnetic excitation method is used.