JPH0228092B2 - DENJIRYURYOKEI - Google Patents

Description

[Detailed description of the invention]

TECHNICAL FIELD OF THE INVENTION The present invention relates to an electromagnetic flowmeter, and particularly to an electromagnetic flowmeter for ultra-low conductivity fluids. [Prior art] Conventionally, an alternating current magnetic field is generated perpendicular to the flow whose flow rate is to be measured, and the electromotive force generated in the fluid is extracted by a pair of electrodes installed on the inner wall of a pipe through which the fluid flows, in direct contact with the fluid. In a normal electromagnetic flowmeter, the measurable fluid conductivity is 2 to 5×
The lower limit is 10 ^-6 ≠/cm. On the other hand, the electrical conductivity of what is generally called "oil" is 10 ^-11 ~
10 ^-17 ≠/cm, and it is considered impossible to measure the flow rate of this "oil" as well as intermediate substances such as acetone and ammonia using the method described above. However, on the other hand, because electromagnetic flowmeters are easier to maintain than mechanical ones, there is a strong desire to realize an electromagnetic flowmeter that can measure even the above-mentioned ultra-low conductivity fluid. [Summary of the Invention] The present invention was made in view of the above circumstances, and its purpose is to make it possible to measure ultra-low conductivity fluids of, for example, 10 ^-7 ≠/cm or less, which was previously considered impossible. Our goal is to provide electromagnetic flowmeters that meet the needs of our customers. In order to achieve such an objective, the present invention
A so-called capacitively coupled electromagnetic flowmeter was used, in which a dielectric material is interposed between the fluid and the detection electrode, and the electromotive force generated in the fluid is extracted by capacitive coupling while the electrode surface is not in direct contact with the fluid. Compensation electrodes extending along the tube axis are provided on the outer periphery of the tube, and an operational amplifier is installed at the front stage of the differential amplifier to which the detection potential obtained from each detection electrode is used as a non-inverting input. By providing an output voltage approximately equal to the detected potential obtained from this voltage divider as the inverting input of the operational amplifier, attenuation due to stray capacitance between the inverting and non-inverting input terminals is compensated for. By applying an output voltage equal to a'/acosθ (a is the inner radius of the tube, a' is the outer radius) obtained by further dividing the output voltage to the above compensation voltage, the tube thickness is finite and the outer circumference is The need to ground the shield compensates for the disturbance and attenuation of the internal electric field distribution. Furthermore, by positively feeding back the output voltage amplified by the operational amplifier at a constant gain through the capacitor, the input capacitance of the operational amplifier is equal to or greater than the minute coupling capacitance between the detection electrode and the fluid. This compensated for the drop in output voltage caused by this. The present invention will be explained in detail below using examples. First, the validity of capacitive coupling will be explained, and the above-mentioned problems that must be solved in that case will be analyzed. Now, we will move on to an explanation of an embodiment as a means for solving the problem. [Example] Regarding the flow rate detection potential difference due to the Faraday effect in ultra-low conductivity fluids, there is an analysis by V. Cushing (“Induction Flowmeter”, Rev. Scl.
Instrum., 29, pp692−697, 1958). According to this, when a tube 1 made of a dielectric material has a finite thickness and its outer periphery is shielded and grounded by a metal shield tube 2, as shown in FIG. The potential difference ΔV obtained between the electrodes 3 is expressed by the following equations (1) and (2). Note that 4 indicates a fluid. ΔV=K・B・D・U…(1) K=σ ₁ +jε ₀ (ε〓 ₁ −1)ω/(σ ₁ +jε ₀ ε
〓 ₁ ω) + (σ ₂ +jε ₀ ε〓 ₂ ω)1+(a/b) ² /1−
(a/b) ² ...(2) However, B: Magnetic flux density D: Tube diameter (inner diameter) (=2a) U: Average flow velocity σ ₁ : Fluid conductivity σ ₂ : Tube conductivity ε ₀ : Vacuum Dielectric constant ε〓 ₁ : Relative dielectric constant ε〓 ₂ : Relative permittivity of the tube ω: Angular frequency of the magnetic field a: Inner radius of the tube b: Outer radius of the tube This generally gives the output signal of an electromagnetic flowmeter. This is the basic equation, and in a normal electromagnetic flowmeter that does not target ultra-low conductivity fluids, σ ₁ is sufficiently large, so it can be safely assumed that K=1. Now, since σ ₂ can be made sufficiently small in practice (for example, 10 ^-18 ≠/cm for Teflon), if we ignore it, equation (2) can be expressed as equation (3) below. K=σ ₁ +jε ₀ (ε〓 ₁ −1) ω/σ ₁ +jε ₀ (ε〓 ₁ +1
+(a/b) ² /1-(a/b) ² ε〓 ₂ ) ω) …(3) When examining the practical limit from equation (3), first of all, the actual thickness of the pipe The value cannot be made zero or infinite, and the conversion of a/b is 0<a/b
<1, but a/b _nio = 0.5 a/b _nax = 0.8 is considered to be the limit. Here, to summarize the numbers, a/b = 0.65. Next, ε〓 ₂ is determined by the lining material, that is, the inner wall of the tube, the dielectric between the detection electrode and the fluid, which requires particularly high insulation, so for example Teflon is used, and in that case ε〓 ₂ = 2 Therefore, equation (3) becomes the following equation (4). K=σ ₁ +jε ₀ (ε〓 ₁ −1) ω/σ ₁ +jε ₀ (ε〓 ₁
+5) ω…(4) ε〓 Regarding ₁ , we will consider two types of fluids. That is, one is petroleum-based, ε〓 ₁ = 2, and the other is pure water, ε〓 ₁ = 80. Alcohols are located in between these. Next, the frequency characteristics of the absolute value of K obtained from equation (4) using these values are shown in FIGS. 2 and 3.
Figure 2 shows the petroleum system, and A, B, C, and D in the diagram show the cases where σ ₁ is 10 ^-5 , 10 ^-18 , 10 ^-11 , and 10 ^-9 ≠/cm, respectively, and Figure 3 shows the pure water system. In the figure, σ ₁ is
The cases of 10 ^-7 and 10 ^-5 ≠/cm are shown. In addition, C in FIG. 3 shows theoretical pure water having the minimum σ ₁ of 5.55×10 ⁻⁸ . In each curve,
The flat part in the low range of ω corresponds to the case where the detection electrode is in direct contact with the fluid, but in process control systems, which are the main fields in which electromagnetic flowmeters are used, a response of at least 1 to 2 seconds is required. Because it will be done,
The lower limit of ω must be considered to be around 20 (3Hz in frequency). Then, as is clear from the figure, in pure water systems, it is possible to measure everything up to theoretically pure water, but in petroleum systems, the lower limit of conductivity that can be measured is 10 ^-10 ≠/cm. Furthermore, since this is a theoretical limit of a completely ideal measurement circuit, it is actually impossible to measure "oil" using the method described above, and capacitive coupling must be used. Therefore, in equation (3), σ ₁ /ω
→ Setting it to 0 corresponds to capacitive coupling. limK limK σ ₁ /ω→0=ε〓 ₁ −1/ε〓 ₁ +1+(a/b) ² /1
-(a/b) ² ...(5) Here, if a/b=1, 1imK=0, so in the tube 1 made of a dielectric material whose outer periphery is shielded and grounded as shown in Fig. 1, If the thickness is reduced, the resulting output signal will be reduced.
Therefore, it is necessary to use a tube as thick as possible, but there are restrictions due to design and cost considerations (for example, Teflon is expensive). Therefore, in the present invention, a/b=0, that is, infinite thickness, was achieved equivalently by the following method. Now, as b → ∞, a point p (θ ) is V(θ), and the potential at point q(θ) with radius a′ opposite to point p is V′(θ). Since there is no electric field in the axial direction of the tube, V′( θ) = a/a′V(θ) This means that if we place an electrode at point q and
This shows that if a potential of V'(θ) is applied, b=∞ is equivalent even if nothing exists outside the radius a'. Since a and a′, that is, the outer radius of the actual pipe, are known, if V(θ) is known, V′(θ)
is decided. V(θ) at θ=0 is the potential V of the detection electrode, and since it can be set that V(θ)=Vcosθ, V'(θ)=a/a'Vcosθ. In other words, the inner radius a and the outer radius are made of dielectric material.
If a compensation electrode extending in the tube axis direction is provided on the outer periphery of the tube a′ and a potential distribution given by the above equation is given to it, then
Signal attenuation due to the finite thickness of the tube can be compensated for. Approximately, this can be achieved by providing a band-shaped compensation electrode 11 divided into a plurality of parts in the axial direction, as shown in FIG. can. FIG. 5 is a block diagram showing an embodiment (part) of the present invention, in which a 90° angle is divided into four parts, and eight compensation electrodes are provided in each of the upper and lower parts. On the other hand, in this embodiment, an operational amplifier 12 as a preamplifier which uses the detection potential obtained at the detection electrode 3 as a non-inverting input is provided before the differential amplifier (not shown).
It is equipped with a voltage divider 13 consisting of resistors 14 and 15 that divides the output of the operational amplifier 12 into 1/A (A is the gain of the operational amplifier 12) (in other words, the gain A is
is determined by the ratio of the resistance values R _A and R _B of resistors 14 and 15). The resistor 15 further serves as a voltage divider that divides the input of the operational amplifier 12 obtained by the above voltage division, that is, the output voltage corresponding to the detected potential, by a/a' cos θ (θ is the angle at which each compensation electrode is arranged). The voltage obtained at each output point is applied to the compensation electrode 11 at the corresponding angle. Note that by feeding back the output equivalent to the detection potential obtained from the connection point between the resistors 14 and 15 to the non-inverting input terminal of the operational amplifier 12, the differential mode stray capacitance between the non-inverting and inverting input terminals is reduced. This compensates for the signal attenuation caused by (C _D ). Although omitted in the figure, the lower detection electrode also has a circuit with a similar configuration. In the illustrated example, the compensation electrode 11 is provided offset from the detection electrode 3 (θ≠0°), but this is for convenience of drawing out, and it may be arranged so that θ=0°. Needless to say. The outside of the compensation electrode 11 is covered with a normal insulating material,
Cover the outside with a metal tube for shielding and ground it. For example, this can be easily realized by using a flexible printed circuit board as the compensation electrode 11, pasting it on the outside of the tube 1 made of Teflon, for example, and grounding the outside as a shield. For the tube 1, for example, Teflon is used, and a structure is used in which a detection electrode is embedded near the inner wall, but for the part that comes into direct contact with the fluid, ceramic, which has higher wear resistance, is used mainly from the viewpoint of wear resistance. Good too. In that case, for example, a detection electrode may be attached to the outer periphery of the ceramic tube, and the outside thereof may be molded with Teflon. With the above configuration, while using a tube with a finite thickness whose outer periphery is shielded and grounded, it is possible to obtain the same signal amount as using a dielectric tube with an infinite thickness. It is also possible to compensate for the attenuation due to Here, the input capacitance of the operational amplifier 12, which is the first-stage amplifier that inputs the signal from the detection electrode, becomes a problem. That is, although the signal extracted by the detection electrode through capacitive coupling is originally very small, the input capacitance becomes equal to or greater than the coupling capacitance, and the signal is significantly attenuated. That is, in FIG. 6, 21 is the coupling capacitance,
The capacitance value C of the coupling capacitor 21 is determined by the design of the flowmeter main body, with 22 being a bias resistor. That is, the area of the detection electrode 3 is S, the gap between this detection electrode and the fluid is t, and the relative permittivity of the dielectric between them is ε〓 ₂
C is expressed as C=S/tε ₀ 〓 ₂ , but in actual design, for example, t=0.15 (cm) S=2.5×1.5 (cm ² ) ε 〓 ₂ = 2 (in case of Teflon) C=4.43×10 ^-12 (F) is considered to be the structural limit. On the other hand, the input capacitance 23 (C _IN ) is 3~
It is about 4 pF, which is almost the same as the coupling capacitance mentioned above, that is, the signal source series capacitance. Therefore, by using an operational amplifier with a FET input with a small bias current and connecting a high bias resistor of 3000 MΩ, the voltage division ratio of the CR coupling part can be adjusted.

〔Effect of the invention〕

As explained above, according to the present invention, a capacitive coupling method is used, a compensating electrode is provided on the outer periphery of the dielectric tube and extends along the tube axis direction, and a constant gain is obtained using a detection potential as a non-inverting input. The pipe thickness is finite by applying to the compensating electrode a voltage obtained by further dividing an output voltage approximately equal to the detected potential obtained by dividing the output of the pre-operational amplifier into a'/acosθ. The input capacitance of the operational amplifier is equal to or greater than the minute coupling capacitance of the signal generation source by compensating for signal attenuation based on the signal generation source and positively feeding back the output of the operational amplifier via a capacitor. It is possible to compensate for signal attenuation caused by electromagnetic flowmeters, making it possible to measure the flow rate of ultra-low conductivity fluids, which was considered impossible with conventional electromagnetic flowmeters.

[Brief explanation of drawings]

Figure 1 is a diagram for explaining the potential difference obtained in an electromagnetic flowmeter, Figure 2 is a diagram showing the ω dependence of the correction term K in petroleum-based fluids using the conductivity of the fluid as a parameter, and Figure 3 is a diagram for explaining the potential difference obtained in an electromagnetic flowmeter. A similar diagram for a water system; FIG. 4 is a diagram for explaining the principle of compensating for signal attenuation based on the finite thickness of the pipe; FIG. 5 is an example of the configuration of a compensation electrode in an embodiment of the present invention. 6 is a diagram showing an example of the configuration of a circuit that compensates for signal attenuation based on the input capacitance of a pre-operational amplifier.
FIG. 7 is a diagram for explaining the compensation effect, and FIG. 8 is a diagram showing the configuration of the preamplifier portion in one embodiment of the present invention. 1... Pipe made of dielectric material, 2... Shield tube, 3...
Detection electrode, 4...Fluid, 11...Compensation electrode, 12...Pre-operational amplifier, 13...Voltage divider, 21...Coupling capacitor,
23...Input capacitance, 31...Capacitor, 41, 42
...An amplifier that constitutes the pre-operational amplifier 12.

Claims (1)

[Scope of Claims] 1. The pipes are arranged so as to sandwich the fluid flowing in the pipes, which are made of a dielectric material whose outer periphery is shielded and grounded, and to be in contact with the fluid through the dielectric material, and to reduce the electromotive force generated in the fluid by capacitive coupling. In a capacitively coupled electromagnetic flowmeter that includes a pair of detection electrodes to be taken out and a differential amplifier that amplifies the potential difference obtained between these detection electrodes, a compensation electrode is provided on the outer periphery of the tube and extends in the tube axis direction. At the same time, an operational amplifier which receives the detection potential obtained from each detection electrode as a non-inverting input is provided in front of the differential amplifier, and a voltage divider which divides the output voltage of this operational amplifier. An output voltage approximately equal to the detection voltage is set as the inverting input of the operational amplifier, and the output voltage approximately equal to the detection voltage is further divided to obtain a'/acosθ (a is the inner radius of the tube, a' is the outer radius). radius)
A double output voltage is applied to the compensation electrode in a direction forming an angle θ with the detection electrode with respect to the center point in a cross section perpendicular to the tube axis direction, and the output of the operational amplifier is connected to the input side via a capacitor. An electromagnetic flowmeter characterized by having a positive feedback circuit.