JPS6147383B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is applied to a sensor such as an eddy current displacement transducer, and detects a change in impedance of a detection coil corresponding to the magnitude of measured displacement using a resonant circuit. The present invention relates to a method for setting circuit constants of an impedance change detection circuit.

FIG. 1 is a block diagram showing an example of a commonly used impedance change detection circuit. In the figure, R ₁ and R ₂ are resistors, C is a capacitor, and L is a coil whose inductance change is detected. Here, the resistance R ₂ represents the resistance of the coil L. The resistor R ₂ , the capacitor C, and the coil L constitute a parallel resonant circuit as shown, and this parallel resonant circuit is connected in series with the resistor R ₁ . Hereinafter, this connection point A will be referred to as the midpoint. OSC is a power supply, and SR is a synchronous rectifier circuit that synchronously rectifies the output voltage e _put detected from the midpoint A using the power supply voltage e _i . In addition, in general, in such a detection circuit, the Q value (ωL/R _{2 )} of the parallel resonant circuit is selected to be a value of 10 or more.Furthermore, the size of the resistor R ₁ is determined to ensure the optimal output voltage e _put In order to obtain approximately QωL/ compared to the impedance ωL of the coil L,
It is selected so that R ₁ ≈1, and normally the value of QωL/R ₁ should be within the range of 0.1 to 10. In addition, if we show an actual numerical example, R ₁ = 850
Ω, R ₂ =1.3Ω, L=12μH, C=1500pF, and the resonant frequency of the parallel resonant circuit is about 1.2MHz.

In an impedance change detection circuit configured in this way, when trying to detect a change in the inductance of the conduit L, conventionally the circuit constants were set so that the parallel resonant circuit was in a resonant state in order to obtain the best characteristics. . However, using a parallel resonant circuit in a resonant state is optimal when the output voltage e _put at midpoint A is directly indicated by an AC voltmeter, etc.; When the output voltage e _put is synchronously rectified using the power supply voltage e _i , the operating state is not necessarily optimal.
Therefore, the circuit constants of the detection circuit must be changed in some way, but it is particularly difficult to set the circuit constants of the resonant circuit to a point other than the resonance state unless there is a clear setting goal.

In the present invention, as described above, the output voltage e _pu at the midpoint A
It is possible to easily set the circuit constants of an impedance change detection circuit in which _t is extracted by synchronous rectification using the power supply voltage e _i , and a circuit with high sensitivity and a good S/N ratio can be realized. The purpose of this invention is to provide a method for setting circuit constants of an impedance change detection circuit.

The method for setting the circuit constants of the impedance change detection circuit of the present invention focuses on the phase of the voltage appearing at the connection point between the resistor and the parallel resonant circuit (hereinafter simply referred to as the phase of the midpoint) as the goal of setting the circuit constants. The circuit constants are set so that the voltage has a specific relationship with the power supply voltage.

Hereinafter, a circuit constant method for an impedance change detection circuit according to the present invention will be explained with reference to the drawings. In the impedance change detection circuit of FIG. 1 described above,
Expressing the relationship between the power supply voltage e _i and the output voltage e _put as e _put /e _i =1/a+jb=X+jY, X=a/a ² +b ² Y=-b/a ² +b ² a=1+R ₁ R ₂ /R〓+ω ² L ² b=R ₁ (ωC−ωL/R〓+ω ² L ² ). Here, it is assumed that the resistor R ₁ is a fixed resistor and does not change, and from the relationship between the Q value of the parallel resonant circuit and the value of the resistor R ₁ , ωL/R ₁ <1, and R ₁ R ₂ / If ω ² L ² <1, then a is almost constant for small changes in ω, L, and C, and although the change in a is small, b
The value of will change significantly. Therefore, if b is deleted from the above formula for X and Y and converted to coordinates where the real part is X and the imaginary part is Y, we get {X-1/2(1-R ₁ R ₂ /R〓+R ₁ R ₂ +ω ² L ²
)} ² +Y ² = 1/4 (1-R ₁ R ₂ /R〓+R ₁ R ₂ +ω ² L
² ) ^The following equation is obtained.

Therefore, from the above circuit conditions, 1/2 (1-R ₁ R ₂ /R + R ₁ R ₂ + ω ² L ² ) is almost constant, so the resistance R ₁ R ₂ , capacitor C, and coil L Angular frequency ω of each impedance and power supply voltage e _i
^No _{matter which parameter of} ^_
)(1). Here, this X+
As mentioned above, jY is the power supply voltage e _i and the output voltage e _p
Since it expresses the relationship (voltage transfer coefficient) with _ut , the unit voltage (e _i =1) as the power supply voltage e _i
If , the vector connecting the origin and the point X+jY directly represents the output voltage e _put . Figure 2 shows this situation.

In this way, the output voltage e
It was found that the locus of _put moves on the circumference of a radius r, but investigating the characteristics of such an L-C parallel resonant circuit is difficult to determine how the locus of output voltage e _put moves with respect to variations in each parameter. This corresponds to checking whether it moves on the circumference in such a manner.

In Figure 2, the radius r' is Δr, and the phase angle θ is Δ
Output fluctuation Δe _put when θ fluctuates is expressed as in the following equation.

This situation is shown in FIG. That is, FIG. 3 illustrates the amount of change Δe _put in the vector e _put when the radius r and phase angle θ, which are elements that determine the magnitude and direction of the vector e _put , change. The relationship among α, θ, and r in the above equation is as shown in the following equation.

α=2θ (3) θ=tan ^-1 −b/a (4) r=1/2(1−R ₁ R ₂ /R〓+R ₁ R ₂ +ω ² L ²
)(5) Here, a=1+R ₁ R ₂ /R〓+ω ² L ² (6) b=R ₁ (ωC-ωL/R〓+ω ² L ² ) (7) Transforming equation (2), we get the following become that way.

Here, x represents any one of R ₁ , R ₂ , C, L, and ω. Equation (8) shows how much the output voltage e _put changes when the parameters constituting the impedance change detection circuit change by Δx, that is, the output sensitivity, and dθ/dx,
If dr/dx is a calculable function, Δe _put can be analytically determined for each parameter.

Next, output sensitivity de _put /dx (hereinafter, Δe _put (x)
). From equations (4) and (5) above, d
You can calculate θ/dx and dr/dx as they are, but
Since the calculation formula is complicated, abbreviations will be made to the extent that it is practically impossible to calculate. As mentioned above,
In a commonly used impedance change detection circuit, the ratio between the impedance ωL of the coil L and the resistance R ₂ is ωL/R ₂ ≫ 1, and it is safe to assume that R ₁ ≫ R ₂ , so as mentioned above (4 ) and (5) become the following equations (9) and (10).

θ=tan ^-1 {-Q ₁ Q ₂ /1+Q ₁ Q ₂ R ₁ (ωC-1
/ωL)}(9) r=1/2 (1-1/1+Q ₁ Q ₂ ) (10) However, Q ₁ and Q ₂ are expressed as in the following equation.

Q ₁ = ωL/R ₁ (11) Q ₂ = ωL/R ₂ (12) Here, assuming that Q ₁ and Q ₂ do not change much near the resonance point, dθ/dx and dr/dx are ( 13) to (16).

dθ/dω=-Q ₁ Q ₂ /1+Q ₁ Q ₂ R ₁ (C+1/
ω ² L) cos ² θ(13) dθ/dL=-Q ₁ Q ₂ /1+Q ₁ Q ₂・R ₁ /ωL
² cos ² θ(14) dθ/dC=-Q ₁ Q ₂ /1+Q ₁ Q ₂ R ₁ ωcos ² θ(1
5) dr/dω=dr/dL=dr/dC≈0(16) Note that it is assumed that R ₁ R ₂ does not change, and the formula for it is omitted.

The output sensitivity Δe _put (x) is calculated by substituting the relationships obtained from equations (13) to (16) into equation (8).

Δe _put (ω)=-r・2Q ₁ Q ₂ /1+Q ₁ Q ₂ R ₁ (C+1/ω ² L) cos ² θΔω (17) Δe _put (L)=-r.2Q ₁ Q ₂ /1+Q ₁ Q ₂・R ₁ /ωL ² cos ² θΔL (18) Δe _put (C)=−r.2Q ₁ Q ₂ /1+Q ₁ Q ₂ R ₁ cos ² θΔC (19) (17) to (19) are expressed as (9) ) and rearranging, we get equations (20) to (22).

Δe _put (ω)=r{sinα −2Q ₂ /1+Q ₁ Q ₂ (1+cosα)}Δω/
ω(20) _{Δe put} (L)=- _Q2r /1+ _Q1Q2 (1+cosα)
ΔL/L(21) Δe _put (C)=r{sinα −Q ₂ /1+Q ₁ Q ₂ (1+cosα)}ΔC/C(
22) In equations (20) and (22), this circuit is assumed to be used near the resonance point as described above, and is used at a point far away from the resonance point (where 1 + cos α is almost 0). Q ₁ <1, Q ₂ ≫1, and Q ₂ /1+Q ₁ Q ₂ >1.
Considering that, the sin α term can be almost omitted, the output sensitivity to inductance and capacitance fluctuations is almost equal, and the output sensitivity to frequency fluctuation is about twice that.

Here, if r and Q ₂ /1 + Q ₁ Q ₂ are considered constant, the output sensitivity given by equations (20) to (22) is a function only of α, that is, the phase angle θ of the midpoint A, and considering the output sensitivity At this time, it is no longer necessary to consider all the parameters ω, L, C, R ₁ and R ₂ .

In equations (20) to (22), output sensitivity Δe _put
(x) is shown as a function only of the phase angle θ, so the phase angle θ that yields maximum sensitivity is dΔe _put (x)/d
All that is required is to find α such that α=0.

dΔe _put (ω)/dα=rsin(α+β ₁ )Δω/
ω(23) dΔe _put (L)/dα=rQ ₂ /1+Q ₁ Q ₂ sinα
ΔL/L(24) dΔe _put (C)/dα=rsin(α+β ₂ )ΔC/C(
25) However, β ₁ and β ₂ are expressed by the following formula.

β ₁ = tan ^-1 1+Q ₁ Q ₂ /2Q ₂ (26) β ₂ = tan ^-1 1+Q ₁ Q ₂ /Q ₂ (27) From equations (23) to (25), α at which the sensitivity is maximum is αΔe _put (ω)=−β ₁ (23) αΔe _put (L)=0 (29) αΔe _put (C)=−β ₂ (30). αΔe _put (x) in the above equation is the value of α that maximizes the output sensitivity when ω, L, and C vary slightly. In equations (28) to (30), 1+Q ₁ Q ₂ /Q ₂ <1, so ω, L, C
It can be seen that the output sensitivity to fluctuations in is maximum when α=0, that is, near the resonance point.

The above analysis was a relational expression that holds true when the output voltage e _put appearing at midpoint A is directly indicated by an AC voltmeter or the like as described above. Therefore, it is difficult to apply the above results to an impedance change detection circuit including a synchronous rectification circuit as shown in FIG. Therefore, next midpoint A
Let us find the optimal value α when the output voltage e _put appearing in is synchronously rectified with the power supply voltage e _i .

As shown in Figure 4, the phase direction of the synchronous rectification output is a, and the direction of the output change Δe _put at the midpoint A is b.
Then, since Δe _put is only due to the change in α, the phase angle between the output change Δe _put at the phase angle θ at the midpoint A and the power supply voltage e _i is (π/2−α). In this state, when the inductance L changes by ΔL, the synchronous rectification output _Vput can be obtained from the following equation.

V _put =-re _i Q ₂ /1+Q ₁ Q ₂ (1+cosα) sin(π/2-α)ΔL/L=-re _i Q ₂ /1
+Q ₁ Q ₂ (sinα+1/2sin2α)ΔL/L (31) In formula (31), the maximum sensitivity α can be obtained by finding the point where dV _put /dα=0 in the same way as formula (24) above. Differentiating equation (31) with respect to α, dV _put /dα=-re _i Q ₂ /1+Q ₁ Q ₂ (cosα+cos2α)ΔL/L (32) There are two points for the desired α: ±π/3 and π. I will do it. Here, since the point α=π is a point where the sensitivity is zero, the optimum value that can actually be used is α=±π/3, and the phase angle θ at this time is θ=±π/6. As mentioned above, this point is the point at which the sensitivity is maximum, but if we also consider that dV _put /dα = 0, even if parameters such as L and C change slightly due to disturbance, the sensitivity will not change, for example The point is that it is not affected by disturbances such as temperature changes.

Therefore, in order to obtain the best characteristics in _the impedance _change detection circuit as shown in FIG. phase angle θ
It is sufficient to set the circuit constants so that the angle is around +30° or -30° with respect to the power supply voltage e _i .

Furthermore, the above calculation was derived assuming that the radius r does not vary in the vector diagram shown in Figure 2, but when calculating the synchronous rectification output V _put from the vector diagram, V _put = e _put · e _i = e _i ² rcosα (33) By setting α=±π/2, the term rcosα can be set to zero, and the signal obtained as α changes can be obtained as shown in the following equation.

V _put = e _i ² rcos (π/2−Δα) = e _i ² rsinΔα(34
) Therefore, by setting the phase angle θ of the midpoint A to around +45° or -45°, the resistance of the coil L can be reduced.
The circuit can be made less susceptible to the influence of _R2 .

FIG. 5 is a block diagram showing another embodiment of an impedance change detection circuit to which the circuit constant setting method of the present invention is applied. In the figure, parts similar to those in FIG. 1 are designated by the same reference numerals. R ₃ R ₄ is the resistance. The impedance change detection circuit shown in the figure utilizes a bridge having a resonant circuit.
In a bridge with a resonant circuit, using the bridge in a state that is intentionally shifted from the equilibrium state,
A bridge that is constructed so as not to be affected by interference signals is particularly called an unbalanced bridge. When setting the circuit constants for such a bridge, it is relatively easy to set the circuit constants for a normal bridge because the bridge should be set so that it is balanced, but for an unbalanced bridge, the circuit constants can be set so that the bridge is balanced. Since it is necessary to set the output signal to a state where the amplitude of the output signal does not change depending on the interference signal, rather than setting it to a state where Have difficulty.

The method for setting circuit constants of an impedance change detection circuit according to the present invention is particularly effective when applied to such an unbalanced bridge. When setting the circuit constants for a bridge as shown in Figure 5,
As in the case of the circuit shown in Fig. 1, if the circuit constants are set so that the phase of the middle point A of the bridge is around +30° or -30° with respect to the power supply voltage e _i , the unbalanced bridge can be eliminated. It is possible to easily set the circuit constants at Tsuji.

In the above description, the circuit constant setting method of the present invention is applied to a circuit for detecting changes in inductance of a coil, but it can be similarly applied to a circuit for detecting changes in capacitance of a capacitor. In addition, although Fig. 5 shows an example of a bridge that has a parallel resonant circuit including a measuring coil or capacitor on one side, the same method can be applied to a bridge that has a parallel resonant circuit for temperature compensation on the other side. Constants can be set.

As established above, in the circuit constant setting method of the impedance change detection circuit of the present invention, attention is paid to the phase of the connection point between the resistor and the parallel resonant circuit, and this phase is +30° or -30° with respect to the power supply voltage. Since the circuit constants are set so as to be close to each other, the circuit constants can be set by simple calculations to obtain a detection circuit with high sensitivity and a good S/N ratio. Furthermore, since the goal of setting is clear, it is easy to change circuit constants.

[Brief explanation of the drawing]

1 and 5 are configuration diagrams showing an example of a detection circuit to which the circuit constant setting method for an impedance change detection circuit of the present invention is applied, and FIGS.
FIG. 3 is a vector diagram showing the operation of the impedance change detection circuit shown in the figure. OSC...Power supply, SR...Synchronous rectifier circuit.

Claims (1)

[Claims] 1. A series circuit of a resistor and a parallel resonant circuit, a power supply that applies an alternating voltage to this series circuit, and synchronous rectification that synchronously rectifies the voltage appearing at the connection point between the resistor of the series circuit and the parallel resonant circuit using the power supply voltage. In the impedance change detection circuit, the circuit constants are set so that the phase of the voltage appearing at the connection point between the resistor of the series circuit and the parallel resonant circuit is around +30° or -30° with respect to the power supply voltage. A method for setting circuit constants of an impedance change detection circuit.