JPH0577252B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention comprises a strain gauge affixed to a predetermined position of a strain-generating body, and the strain gauge is assembled into a bridge circuit.
The present invention relates to a method of compensating for changes in zero point temperature of a load cell that measures load from the output of the bridge circuit.

[Prior art]

Conventionally, strain gauges Z ₁ to _{Z 4} are attached to predetermined thin-walled portions 10a to 10d of a strain generating body 10 having a structure as shown in FIG.
There is a load cell having a structure in which the strain gauges Z ₁ to _{Z 4} are assembled into a bridge circuit as shown in FIG. In this load cell, when a load in the direction of arrow A is applied to one end of the strain-generating body 10, the strain gauges _Z1 and
_Z4 receives a compressive force, and strain gauges _Z2 and _Z3 receive a tensile force, resulting in an output _V0 corresponding to the load from the bridge circuit.

In the load cell configured as described above, the output V ₀ changes in response to temperature changes, so temperature compensation is required. This temperature compensation includes so-called span compensation that compensates for changes in the output V ₀ caused by the Young's modulus of the thin parts 10a to 10d of the strain body 10 due to temperature changes, and the resistance values of the strain gauges Z ₁ to _{Z 4} that change due to temperature changes. There is a so-called zero point compensation that compensates for the change in the zero point due to the change in .

For the above span compensation, insert low antibodies for span compensation at points a and b in Figure 3, and for zero point compensation, insert zero point compensation at points c, d, e, f, g, h, i, and j in the same figure. Each compensation is performed by inserting a resistor.

[Problem that the invention seeks to solve]

In the load cell with the above configuration, the change in zero point output V ₀ with respect to temperature T is as shown in Fig. 4.
Contains non-linear components.

The conventional method of compensating for temperature changes at the zero point simply compensates for the linear change (gradient) at the zero point due to temperature changes, as shown in Figure 5. If there is a non-linear component in the output change, that nonlinear component is compensated. It was not possible to compensate for the linear component, which caused measurement errors due to temperature changes in the load cell.

The present invention has been made in view of the above-mentioned points, and provides a method for compensating for changes in zero point temperature of a load cell, which compensates for both changes when there are linear changes and non-linear changes in changes with respect to temperature at the zero point of a load cell. It's about doing.

[Means for solving problems]

In order to solve the above-mentioned problems, the present invention includes a metal strain gauge that has a first-order temperature coefficient and a second-order temperature coefficient and whose resistance value changes depending on temperature, and also incorporates the metal strain gauge into a bridge circuit. In the load cell that measures the load from the output of the bridge circuit,
The non-linear change and linear change in the zero point temperature change of the load cell are measured by using a resistor of a different type than the metal strain gauge, which has a first-order temperature coefficient and a second-order temperature coefficient for which the following equation holds on the side of the bridge circuit. By inserting it in series as a compensation resistor, zero point temperature compensation is established for the square term of temperature change, thereby compensating for zero point temperature change of the load cell.

Z・Bz=−R・Br Here, Z is the nominal resistance value of the strain gauge, B is the square term of the temperature coefficient of the load cell, R is the temperature compensation resistance value, and Br is the square term of the temperature coefficient of the temperature compensation resistance. It is.

[Effect]

In the bridge circuit shown in FIG. 3, the zero point output V ₀ (output when no variable load is applied) of the bridge circuit is expressed by equation (1).

V ₀ = V _io {Z ₂ / (Z ₄ + Z ₂ ) −Z ₁ / (Z ₃ + Z ₁ )} (1) Therefore, V _io = {Z / (R _Z + Z)}・E (2) R _Z = R _a + R _b (3) (However, R _a and R _b are the resistance values of the resistors inserted in a and b in Figure 3, and Z is the input resistance value of the bridge circuit) Z = (Z ₄ + Z ₂ ) (Z ₃ +Z ₁ ) / (Z ₁ +Z ₂ +Z ₃ +Z ₄ ) (4) In general, factors that cause the zero point of a load cell to change are related to the amount of water in the strain gauge in addition to temperature. Now, the functions of zero point change factors that each strain gauge Z ₁ to _{Z 4} have are aggregated into one strain gauge Z ₃ , and the functions of the other strain gauges Z ₁ , Z ₂ , and Z ₄ are unrelated. Then, Z ₃ =Z ₃₀ +f(T)+g(H). Here, Z ₃₀ is the resistance value of the strain gauge Z ₃ that is not affected by temperature and humidity, g (H) is the resistance value that changes depending on the moisture content of the load cell strain gauge, and f (T) is the resistance value that changes depending on the temperature of the strain gauge. This is the resistance value that changes accordingly. Further, the above g(H) is expressed as g(H)=G(h, T, t). Here, h is relative humidity, T is temperature,
t indicates the time spent under a certain environmental condition.

In the above equation, when h is small, g(H) becomes negligibly small, and the temperature change at the zero point of the load cell is approximately related only to the temperature. According to experiments, g(H)
Under the condition of ≈0, the change in zero point output due to temperature is approximated by a quadratic function. Therefore, as mentioned above, the change resistance value f due to temperature concentrated in the strain gauge _Z3
(T) is also approximated by the following quadratic function.

f(T)=a ₀ +a ₁ T+a ₂ T ² , that is, under the condition of g(H)≈0, Z ₃ =Z ₃₀ +a ₀ +a ₁ T+a ₂ T ² . Therefore, if the load cell is used in a low-temperature environment or if it is completely coated with a moisture-proof coating, the zero point can be reduced by inserting a resistor having a temperature coefficient of the quadratic term in the bridge to cancel out the quadratic term. Non-linearity of temperature change can be corrected. In addition, the linear temperature change at the zero point, including the temperature coefficient of the linear term of the resistor inserted to compensate for nonlinearity (which occurs when the originally linear temperature change is added or subtracted), is less than or equal to the quadratic term. (approximately 1/100 or less of the quadratic term of the temperature coefficient of the resistance value for nonlinearity compensation).
Now, if we express the resistance value R of the temperature compensation resistor using a quadratic equation, we get R=R ₀ (C+AT+BT ² ), and if we express the above Z ₃ = Z ₃₀ + a ₀ + a ₁ T+a ₂ T ² according to the above equation, we get , Z ₃ = Z ₃₀ ′(a ₀ ′+a ₁ ′T+a ₂ ′T ² ), and here, using a resistor with a first-order term or less,
The resistance change of only the next term is Z ₃ = Z ₃₀ ′a ₂ ′T ² and R =
R ₀ BT ² , and in order to cancel the quadratic term of Z ₃ with the compensation resistor, it is sufficient to make the resistance value R equal to the resistance value Z ₃ of the strain gauge Z ₃₀ , and Z ₃₀ ′a _{2 ′} T ² = − R ₀ BT ² , and Z ₃₀ ′a ₂ ′=−R ₀ B. Here, Z ₃₀ ′ is the nominal resistance value Z of the strain gauge,
Since a ₂ ' can be expressed as the square term Bz of the temperature coefficient of the load cell, R ₀ can be expressed as the resistance value R of the temperature compensation resistor, and B can be expressed as the square term Br of the temperature coefficient of the temperature compensation resistor, Z・
z=-R·Br, and if the temperature compensation resistor satisfies this condition, it is possible to compensate for the zero point temperature change of the load cell.

〔Example〕

Hereinafter, one embodiment of the present invention will be described based on the drawings.

The strain body of the load cell is the same as the strain body 10 having the shape shown in FIG.
Attach strain gauges Z ₁ to Z ₄ to 0a to 10d. The nominal resistance value of the strain gauges Z ₁ to _{Z 4} is 350Ω.
When strain gauges Z ₁ to _{Z 4} are connected to the bridge circuit shown in FIG. 1 and the power supply voltage E is 12V, the zero point output of the bridge circuit at each temperature listed below was as shown below.

25℃ -4.8991mV 50℃ -4.9041mV -10℃ -4.9241mV Change in zero point output due to temperature of this load cell.
Summarizing Z ₃ , the resistance value f(T) is f(T)=−0.70246+a ₁ T+a ₂ T ² a ₁ =350×(−9.2820×10 ⁻⁷ ) a ₂ =350×(−6.6051×10 ^{− 9} ).

As mentioned above, when a ₂ < 0, a low antibody (for example, Nickel) with a positive quadratic term temperature coefficient is used as a strain gauge Z ₃ as a resistor to compensate for the nonlinearity of zero point temperature change.
Insert the resistor _X3 or _X2 on the same side as the strain gauge _Z2 or on the same side as the strain gauge Z2. Furthermore, if the resistor has a negative quadratic term temperature coefficient (for example, platinum), it is inserted as the resistor X ₁ or X ₄ on the same side as the strain gauge Z ₁ or on the same side as the strain gauge Z ₄ in the opposite direction. Also, a ₂ >
In the case of 0, if the resistor has a positive quadratic term temperature coefficient, place the resistor on the same side as strain gauge Z ₁ or Z ₄ .
Insert as X ₁ or X ₄ . For example ^, as a resistor that compensates for the non-linearity of the zero ^point temperature change of the load cell, a resistor _R1 of approximately 3.27Ω (0°C ) strain gauge
Insert resistor X ₃ or X ₂ on the same side as Z ₃ or Z ₂ . Also, as a resistor that compensates for linear zero point change, the temperature coefficient of resistor _R2 is approximately 3.8Ω (0℃), which has a temperature coefficient of 1 + 0.004368 × 10T.
Place resistor X ₁ or resistor on the same side as strain gauge Z ₁ or Z ₄ .
Insert as _X4 . In this case, the zero balance is 4mm
V collapse from -4.9mV to -8.9mV. but,
At the same time, if a non-temperature sensitive resistor (approximately 50 ppm decrease) having the same resistance value at 0° C. as the resistors R ₁ and R ₂ is inserted into the corresponding bridge, the zero balance will not be lost.

Here, in order to create an easy-to-understand model that incorporates all cases, we have explained using a resistor with a fictitious temperature coefficient, but we have used a nickel resistor to correct non-linear changes in the zero point, and If a copper resistor is used to correct this, the zero balance will not be disturbed much, and there will be no need to insert a non-temperature-sensitive resistor. In FIG. 1, SR ₁ and SR ₂ are span compensation resistors.

Next, a specific experimental example of the compensation method for the zero point temperature change of the load cell will be explained.

In Figure 1, span compensation resistor SR ₁ and
The following is used as SR ₂ .

SR ₁ = 55.5 × (C ₁ + A ₁ × T + B ₁ × T ² ) Ω, (T
is temperature) (C ₁ = 1.00002, A ₁ = 3.56112×10 ^-3 , B ₁ = -
6.11861×10 ^-7 ) SR ₂ = 26× (C ₂ + A ₂ × T + B ₂ × T ² ) Ω (C ₂ = 0.916777, A ₂ = 3.26167×10 ^-3 , B ₂ = −
4.31664×10 ^-6 ) Also, as strain gauges Z ₁ to _{Z 2} , strain gauges with a nominal value of 350Ω are pasted to the thin parts 10a to 10d of the flexure element 10, and the The zero point output values were measured at three points and were as follows. Note that the input voltage E is constant at 12V.

Low temperature -8.4℃ 5.2783mV (100.672%) Room temperature 24.8℃ 5.1977mV (100%) High temperature 51.2℃ 5.1191mV (99.345) Resistance values of strain gauges Z ₁ to Z ₄ Z ₁ = Z ₂ = Z ₄ = 350Ω (5 ) and Z ₃ as Z ₃ = 350 + 350 × (A × T + B × T ² + C) (6) and calculate the coefficients A, B, and C of the load cell by applying them to the formulas (1) to (4) above. , A = 4.8017×10 ^{-7 B} = -4.17685 −0.068%
It was hot. This can be regarded as the maximum error in the nonlinearity of the zero point output with respect to temperature.

Now this factor is the square term B of the temperature coefficient of the load cell.
_{Assuming that this is caused by} ^_ Considering that a nickel wire with ₃ = 1, A ₃ = 0.0057457, and B ₃ = 0.0000065 is inserted in series with the strain gauge Z ₃ , from equations (6) and (7) above, 350B = -R _NO B ₃ If the condition (8) is satisfied, the quadratic terms of the temperature coefficients in both equations can be canceled out. Therefore, R _N0 = 350 × 4.17685 × 10 ^-9 / 6.5 × 10 ^-6 = 3.5 × 4.17685 × 10 ^-1 / 6.5 = 0.2249Ω nickel wire should be inserted in series on the side of strain gauge Z ₃ . In reality, a nickel resistor with a measured value of 0.2249Ω at 0°C, including cutting errors, was inserted on the strain gauge _Z3 side. In order to correct the slope that is combined with the linear temperature change in the output caused by inserting this resistor and the linear change in the zero point of the original load cell, the side of the strain gauge Z ₁ or Z ₄ is A copper wire having a resistance value as shown below was inserted in series.

0.224×(1+0.00427T)Ω The calculated values and actual measured values after inserting these values are shown in Figures 6 and 7.

FIG. 6 shows the case where compensation is performed using a nickel wire of 0.2249Ω (0°C) and a copper wire of 0.2240Ω (0°C), where a shows the calculated value and b shows the measured value.
In the figure, linearity LI indicates the maximum difference between the calculated value and the measured value, that is, the maximum bulge, by connecting the value at 0° C. and the value at 50° C. with a straight line, and the gradient GR indicates the gradient of the straight line.

FIG. 7 shows the case where compensation is performed only with a copper wire of 0.1060Ω (0° C.), where a shows the calculated value and b shows the actually measured value. In the figure, linearity LI indicates the maximum bulge connecting the 0°C value and 50°C value with a straight line, as in the case of Fig. 6, and gradient GR indicates the slope of the above straight line.

In the above embodiment, a first resistor having a quadratic term and a second resistor having a small quadratic term and a large first-order term are used as different resistors, and the non-linear change in the temperature change of the zero point output is detected by the first resistor. The linear change and the linear term of the first resistor are canceled by the second-order term of the first resistor.
The first resistor with the first order term of the temperature coefficient is erased by the first order term of the resistor with a small first order term of 2.
By using a second resistor with a large next term, the linear change in temperature change of the zero point output is canceled by the linear term of the first resistor, and the non-linear change and the quadratic term of the first resistor are eliminated by the second resistor. It may also be erased by a quadratic term of a resistor (in this case, for example, a nickel or platinum resistor is used as the first resistor, and manganin or constantan is used as the second resistor). In addition, by using a first resistor whose temperature coefficient has a small first-order term and a large second-order term, and a second resistor whose second-order term has a small second-order term and a large first-order term, the non-linear change in temperature change of the zero point output can be suppressed by using the first resistor. The linear change is eliminated by the quadratic term of the field, and the linear change is
(In this case, for example, the first resistor has a manganin resistance, the second
(using a copper resistor as the resistor).

In the above embodiment, each side of the bridge circuit is made up of strain gauges, but it is also possible to have one or two active gauges and the others are dummy resistors. Further, in the above embodiment, the shape of the strain body of the load cell shown in FIG. 2 was used, but the shape of the load cell is not limited to this, and the shape of the load cell may be arbitrary.

In addition, even if different types of resistors have their own temperature coefficients of first-order and second-order terms, if they are a combination of items with different characteristics of the second-order terms, the proportions of each may be changed in a timely manner to improve the load cell output. It is possible to correct for linear (slope) and non-linear changes with temperature. For example, there are the following combinations of resistors.

Nickel + platinum.

〔Effect of the invention〕

As explained above, according to the present invention, the following excellent effects can be obtained.

ZBz=-
RBr (where Z is the nominal resistance value of the strain gauge, Bz is the square term of the temperature coefficient of the load cell, R is the temperature compensation resistance value, and Br is the square term of the temperature coefficient of the temperature compensation resistance). By inserting a different type of resistor in series with a metal strain gauge that has a coefficient and a quadratic temperature coefficient, zero point temperature compensation is established for the square term of temperature change, so the zero point temperature change of the load cell is non-linear. and linear changes can be compensated for.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing a bridge circuit of a strain gauge applied to the compensation method for zero point temperature change of a load cell according to the present invention, Fig. 2 is a diagram showing a schematic configuration of a load cell, and Fig. 3 is a diagram showing a zero point temperature change of a load cell according to the present invention. A diagram showing a bridge circuit of a strain gauge to explain the principle of the compensation method, FIG. 4 is a diagram showing changes in load cell output with respect to temperature, and FIG. 5 is a diagram showing a conventional example of correcting the linear change. Figure 6 shows 0.2249Ω (0
This figure shows the case of compensation with nickel wire of 0.1060Ω (0℃) and copper wire of 0.1060Ω (0℃). Figure a shows the calculated value, figure b shows the measured value, and Figure 7 shows the 0.2240Ω.
This is a diagram showing a case where compensation is performed only with a copper wire at (0° C.), where a shows a calculated value and a diagram b shows an actual measured value. In the figure, Z ₁ to _{Z 4} ... strain gauges, and X ₁ to _{X 4} ... compensation resistors.

Claims (1)

[Scope of Claims] 1. A metal strain gauge having a first-order temperature coefficient and a second-order temperature coefficient and whose resistance value changes depending on temperature, and the metal strain gauge is assembled into a bridge circuit, and the output of the bridge circuit is provided. In a load cell that measures a load from a load cell, the metal strain gauge has a first-order temperature coefficient and a second-order temperature coefficient on the side of the bridge circuit that satisfy the following equation to express non-linear changes and linear changes in zero-point temperature change of the load cell. By inserting a different type of resistor in series as a temperature compensating resistor, the temperature change can be reduced by
A method for compensating for changes in zero point temperature of a load cell, characterized by establishing zero point temperature compensation for a multiplicative term. Z・Bz=−R・Br Here, Z is the nominal resistance value of the strain gauge, Bz is the square term of the temperature coefficient of the load cell, R is the temperature compensation resistance value, and Br is the square term of the temperature coefficient of the temperature compensation resistance. It is. 2 first and second 2 as the temperature compensation resistor;
The non-linear change in the zero point temperature change is canceled by the quadratic term of the temperature coefficient of the first resistor having a quadratic term, and the linear change and the linear term of the first resistor are eliminated. 2. A method of compensating for a zero point temperature change in a load cell according to claim 1, wherein the temperature coefficient of the second resistor has a small quadratic term and a large linear term. 3. The first and second two as the temperature compensating resistor.
The linear change in the zero point temperature change is canceled by the linear term of the temperature coefficient of the first resistor, and the non-linear change and the
2. A method for compensating for a zero point temperature change in a load cell according to claim 1, wherein the next term is canceled by a second-order term of a temperature coefficient of a second resistor having a small first-order term and a large second-order term. 4 the first and second two as the temperature compensation resistor;
A first resistor whose temperature coefficient has a small first-order term and a large second-order term eliminates the non-linear change in the zero-point temperature change, and a linear change is canceled by a first resistor whose temperature coefficient has a small first-order term and a large first-order term. 2. A method for compensating for a zero point temperature change of a load cell according to claim 1, wherein the compensation method is performed using a resistor of 2. 5. Any one of claims 1 to 4, characterized in that when the zero balance deviates from a predetermined value, the zero balance is further corrected by a zero balance correction resistor having a temperature coefficient of approximately zero. Compensation method for load cell zero point temperature change described in .