JPS6161043B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a method for temperature compensating the output of a displacement meter when the displacement of a displaced object is measured by a displacement meter. There are displacement meters based on various principles, but in all of them, when the ambient temperature or the temperature of a displaced object changes, the displacement-output characteristics thereof change. For example, in an eddy current type displacement meter, a displaced object and a coil are placed opposite each other, and the coil emits lines of magnetic force from the coil to the displaced object. At the same time, changes in impedance of the coil due to the eddy current generated in the displaced object are converted into voltage output. The impedance changes depending on the gap between the displacement object and the coil, which corresponds to the displacement of the displacement object, and the electric conductivity of the displacement object and the air in the gap, which changes with temperature.
It will also be affected by factors such as magnetic permeability. Therefore, it is necessary to temperature-compensate the output of the displacement meter, and conventionally, as a compensation method, a method has been attempted in which the temperature near the displaced object is detected and the displacement meter output is corrected accordingly. However, this correction method has the disadvantages that the correction calculation is complicated, and that accurate correction cannot be performed because the detected temperature is not the temperature of the displaced object or the displacement meter itself.
In particular, when the temperature change is rapid, the correction itself is difficult because the thermal transient responses of the displaced object, the displacement meter, and the thermometer are not the same. In addition, as shown in Japanese Patent Application Laid-open No. 50-130471, for example, it is possible to detect changes in the output of a displacement meter due to temperature,
There is also a method of adjusting the operating section of the displacement meter itself, but in this case, an adjustment section must be added to the displacement meter, and there is a problem that temperature compensation cannot be performed as is for an existing displacement meter. Therefore, when considering temperature compensation for the output taken out from a displacement meter, if temperature compensation is difficult with one displacement meter, two displacement meters with the same characteristics should be placed in the same temperature environment. The first thing to consider is whether the influence of temperature can be canceled out using the two displacement meter outputs. However, in this case, if the outputs of the two displacement meters become the same displacement output for the displacement of the displaced object, correction is impossible, so the outputs of both displacement meters will be different for the same displacement. There must be. In order to meet these conditions, it is sufficient to sandwich a displaced object and place displacement meters with the same characteristics on both sides. In this case, if the output of one displacement meter increases in response to the displacement of the displaced object, , the other will decrease, and both displacement gauges will be placed under the same temperature environment. Hereinafter, the output of each displacement meter will be first explained with reference to the drawings. In FIG. 1, which shows the arrangement of displacement meters, displacement meters 1 and 2 are arranged to face each other on the left and right with a displacement object 3 in between. As a result, when the displacement object 3 is displaced to the left or right, the respective gaps G ₁ and G ₂ between the displacement meters 1 and 2 and the displacement object 3 will increase or decrease, respectively. Now, based on the state where the displaced object 3 is located at the midpoint of the placement positions of both displacement meters 1 and 2, the displacement of the displaced object is x (positive displacement in the right direction), and the sum of the gaps G ₁ and G ₂ If we set C (constant value), the gaps G ₁ , G ₂
The relationship between and displacement X is as follows. In other words, when the displacement x changes between -C/2 and +C/2, the gaps G ₁ and G ₂ change from 0 to C and from C to
It will change between 0 and 0. Figure 2 shows the output voltages of displacement meters 1 and 2 when the displacement x changes within the range of ±C/2 as described above.
An example of V ₁ V ₂ is shown, where the horizontal axis is the displacement x and the vertical axis is the output voltage V. Also, each output voltage
At V ₁ V ₂ , the curves shown by the solid line, broken line, and dashed-dotted line are the output voltages at temperatures t ₀ , t ₁ , and t _{2 ,} respectively. In this way, the output voltages V ₁ and V ₂ of displacement meters 1 and 2
are the same at the point where the displacement x is 0, and are symmetrical around that point, and the zero point voltage and the shape of the output curve change depending on the temperature. Now, if the output voltages of displacement meters 1 and 2 at temperature t ₀ are expressed by a general formula, it will be as follows. V=K ₀ G ^l +V ₀ (2) However, V: Output voltage at temperature t ₀ K ₀ : Proportionality coefficient at temperature t ₀ V ₀ : Constant voltage l: Displacement characteristic index of displacement meter This K ₀ , V ₀ will change depending on the temperature. Now, if the temperature t, that is, the temperature difference T (=t- _{t 0 )} , the proportional coefficient is K _T , and the constant voltage is V _T , they are as follows. It can be expressed as follows. Here, α: Temperature coefficient of proportionality coefficient β: Constant voltage 〃 m: Index showing temperature characteristics of proportionality coefficient n: Constant voltage 〃 However, when m and n are even numbers, α and β when T<0 Reverse the polarity of As a result of the above, the output voltage of the displacement meters 1 and 2 at the temperature t is expressed as follows from equations (1) to (3). K ₀ , V ₀ , α, β, n, and m in the above equation are constants determined depending on the type of displacement meter. Next, we will use these two output voltages V ₁ and V ₂ to compensate for the temperature, but it is impossible to remove the temperature term by simply calculating the ratio of V ₁ and V ₂ . . Next, considering the difference between the two output voltages V ₁ and V ₂ , we get the following. V ₁ −V ₂ =K ₀ (1−αT ^m ) {(C/2+x) ^l −(C/2x) ^l } (5) This difference is due to the fact that there is only one temperature term, and the difference between (In the case of l = 1, there is originally a linear relationship, but even in the case of l≠1, the linearity of V ₁ - V ₂ is improved compared to V ₁ or V ₂ ), but The effect of temperature cannot be excluded. Then, a possible method is to convert V ₁ −V ₂ to V ₁ or
Although dividing by V ₂ does not eliminate the temperature term, the linearity deteriorates. By the way, when considering the sum of both output voltages V ₁ +V ₂ , since V ₁ and V ₂ have mutually opposite characteristics, it can be regarded as an approximately constant value with respect to the displacement x. Therefore, by calculating the ratio of V ₁ −V ₂ and V ₁ +V ₂ mentioned above, there is less risk of degrading the linearity, and since the temperature term is given in the form of a ratio as shown in the following equation, the calculation The degree of influence of temperature on the overall value is also reduced. That is, V ₁ −V ₂ /V ₁ +V ₂ =K ₀ /2V ₀ {(C/2+x) ^l −(C/2−x) ^l }/K ₀ /2V ₀ {(C/2+x)
^l + (C/2+x) ^l }+ (1+βT ⁿ )/(1+αT ^m )…(6) However, even if this is done, the temperature term (1+βT
ⁿ )/(1+αT ^m ) will remain. Therefore, in order to remove the temperature term, the term βT ⁿ or αT ^m is multiplied by a predetermined proportionality coefficient a, and aβ
It is sufficient that T ⁿ is approximated to αT ^m . In other words, 1+aβT ⁿ /1+αT ^m ≈1 (7). Therefore, in equation (7) above, if the exponents n and m of both temperature terms are the same, equation (7) becomes 1 by setting the proportionality coefficient a to α/β, and the influence of temperature is completely eliminated. removed. However, if n and m do not match, instead of the proportionality coefficient a, which is a predetermined constant,
Equation (7) cannot be set to 1 unless a is a function term of temperature. However, even if such complete temperature compensation cannot be performed, aβT ⁿ can be approximated to αT ^m by selecting an appropriate value for the proportionality coefficient a, and the temperature term approximated in this way can be When the ratio is calculated using , it approaches 1 as shown in equation (7). For example, for the temperature term αT ^m where T = 0 to 60, α = β = 0.0001, and m = 2 to simplify the explanation, an index n = different from m
When the temperature term βT ⁿ having 1.5, 1, and 1/2 is approximated by proportionality coefficient a = 7.2, 48, and 300, respectively, the ratio becomes 1 compared to the case where approximation by proportionality coefficient a is not performed as shown below. The value is close to .

【table】

[Table] Based on the above idea, the present invention, when correcting the output of displacement meters placed on both sides of a displaced object, calculates the difference between the two outputs by adding a predetermined offset value to the sum of both outputs. βT ⁿ
is multiplied by a predetermined proportionality coefficient, thereby performing temperature compensation. First, regarding the temperature term in equation (7) above, the proportionality coefficient a
A method of calculating (1+aβT _o ) multiplied by (1+aβT o ) will be explained. Of the temperature term (1+βT ⁿ ), it is impossible to linearly multiply only βT ⁿ by the proportionality coefficient a. By the way, if we add the constant X to (1+βT ⁿ ), we get
The added value is obtained by multiplying the term βT ⁿ by a proportionality coefficient determined by X, and takes the form of the sum of “1”. In ^{other words ,}
It is sufficient to set X so that it matches T ⁿ ). Therefore, a=1/x+1, x=1/a-1 (9) As shown above, if we add a constant X that satisfies equation (9) to (1+βT ⁿ ), the effect of temperature will be reduced in the temperature term. It will be removed. Next, by replacing equation (8) with (1+βT ⁿ ) in equation (6), an arithmetic expression for compensating for the temperature term can be obtained. In other words, K ₀ /2V ₀ {(C/2+x) ^l −(C/2-x) ^l }/K ₀ /2V ₀ {(C/2+x) ^l + (C/2-x) ^l }+
{X+(1+βT ⁿ )}/(1+αT ^m ) =K ₀ (1+αT ^m ) {(C/2+x) ^l −(C/2−x) ^l }/K ₀ (1+αT _n ){(C/2+x) ^l +(C
/2-x) ^l }+2V ₀ (1+βT ⁿ ₎ +2V ₀ X = V ₁ -V ₂ /V ₁ +V ₂ + _{2V 0}
Therefore _{, 2V 0} Compensation will be available. Therefore, first, aβT ⁿ and αT ^m are approximated by a
After calculating the constant X and determining the offset voltage obtained by multiplying X by the constant voltage, the correction calculation described above may be performed. Although the method for determining the offset voltage in the case where the output characteristics of the displacement meter are determined by a mathematical formula has been exemplified above, the offset voltage V _c may also be determined experimentally by the following procedure. (i) Measure the maximum displacement, for example, the output voltage of ±C/2, at several points or more within the measurement temperature range. (ii) Output voltage with a displacement of ±C/2 at one temperature t _i V+c/2, i, V-c/2, i and a similar output voltage at another temperature t _i+1 V+c/2, i+1, V -c/2, i+1, the offset voltage V
Set up an equation assuming that _c is unknown and find V _c . (iii) Repeat the operation in (ii) above at each pair of temperatures to find the offset voltage V _c for each pair. (iv) When there is some difference in such offset voltage V _c (because aβT ⁿ is an approximation of αT ^m )
Let the average value be the offset voltage. As described above, in the present invention, since the displacement gauges are placed opposite to each other with the displacement object in between, both displacement gauges are affected by the same effect even in a state of transient temperature change. Since the calculation is performed by dividing the difference between the outputs by a weighted value obtained by adding a predetermined offset value to the sum of the outputs, an accurately temperature-compensated calculation value can be obtained. In addition, in the present invention, in calculation, the difference between outputs with the same characteristics and mutually opposite characteristics with respect to displacement is divided by the sum of outputs that can be regarded as a substantially constant value, so the linearity of displacement - calculated value is greatly improved. Depending on the characteristics of the displacement meter, there may also be a side effect of obtaining a performance value proportional to displacement with sufficient accuracy without the need for a linearizer.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram showing the arrangement relationship of the displacement meter in the present invention, and FIG. 2 is an example of a characteristic diagram showing the displacement-output characteristic of the displacement meter at temperatures t ₀ , t ₁ , and t ₂ . . 1, 2: displacement meter, 3: displacement object.

Claims (1)

[Claims] 1 Displacement meters are placed facing each other with a displacement object in between, the outputs of both displacement meters are taken out, the difference between them is calculated, it is used as the dividend, and the sum of the outputs of both displacement meters is further weighted by a predetermined offset value. A temperature compensation method for a displacement meter, in which a value is determined and used as a divisor, and a ratio between the dividend and the divisor is determined to obtain a temperature-compensated displacement output.