JP2639759B2 - Multi-sensor weigh scale - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic digital display weighing scale, and in particular to a weighing scale having a low height in relation to the size of a mounting table of the weighing scale (flat weighing scale). And a weighing scale suitable for a large weighing scale that weighs as a total value of outputs of a plurality of sensors. 2. Description of the Related Art In general, a method of concentrating a load applied to a mounting table at one point and measuring the load with one sensor is used. The so-called flat-type weighing scale has a problem in the measure of the measurement error due to the eccentric load, the method of concentrating the load at one point, the mechanical strength of the load transmission mechanism, etc. In the scale, a method called a multi-point weighing scale, in which the load is distributed and weighted to a plurality of sensors, and the outputs of the respective sensors are summed, is adopted. In the above-mentioned multi-point weighing scale, the ratio of the load applied to each sensor varies depending on the position of the load mounting table. If the output characteristics of each sensor are not the same, correct weighing cannot be performed. That is, there is a problem that a so-called eccentric load error occurs. [0004] The countermeasure against this eccentric load error is to individually remove each strain sensor, such as by removing the strain-generating portion of the sensor so that the output characteristics of each sensor used for one weighing scale under the reference load are all the same. Was adjusted mechanically. However, making the outputs of the individual sensors identical mechanically with high precision requires skill and requires a large number of adjustment steps. Therefore, an amplifier is connected to each of the individual sensors.
Mechanically, a sensor output that differs from the reference load is individually input to the amplifier, the amplification factor is adjusted, electronically corrected so that the output value from the amplifier becomes the same, and the output is added. Has been adopted. In any case, since adjusting the analog output value for each sensor requires a lot of man-hours, an A / D converter is connected for each sensor output, and when adding a digital amount after digital conversion, A method of performing correction according to individual sensor characteristics has been proposed, and this method has become widespread. However, this method requires as many A / D converters as the number of sensors, which directly affects the cost. Therefore, all the sensors are switched to one A / D converter and connected alternately, and one A / D converter is connected. A / D converter for each sensor with / D converter
A method of making the connection equivalent to connecting a converter (hereinafter referred to as a scanning method) is conceivable (FIG. 2). In this method, no problem occurs when the load is stable, but the load fluctuates. In such a case, the load ratio of each sensor changes within one measurement period, so that the total output of each sensor at different times fluctuates for each measurement period, and accurate measurement cannot be performed. [0008] The time-consuming and difficult adjustment of individual sensors as described above is eliminated, and an accurate weighing scale can be obtained even when measuring a swinging load without increasing the cost. Is to do. Means for Solving the Problems A plurality of sensors are appropriately arranged, the outputs of all the sensors are simultaneously input to one integrating A / D converter, and the input integration time of the output of each sensor is determined. The control is performed in proportion to the reciprocal of the output of each sensor at the reference load. The standard output V of the sensor at the time of the reference load W (g) is
Assuming that the number of clocks in the reference integration period obtained when inputting to the integration type A / D converter having the number of clocks in the input integration period: N, the clock cycle: T, and the reference integration voltage: VR is n, n = V / VR * N. expressed. When the output Vi when an arbitrary sensor is loaded with the reference load W (g) is input to the A / D converter, and the number of clocks in the reference integration period is ni, it is represented by ni = Vi / VR * N. By multiplying both sides by n / ni, the above equation becomes n = Vi / VR * N * n / ni. Above, when converting the output of an arbitrary sensor into a digital value by an integrating A / D converter, the number of clocks of the input integration time is multiplied by n / ni which is a number proportional to the reciprocal of the output of the sensor. Is obtained, indicating that the correction can be made by adjusting the input integration time. That is, a plurality of sensors are connected to the same integral type A / D
The output voltage at the end of the input integration period when the input integration time of each sensor is disconnected at a time (NT * n / ni) proportional to the reciprocal of the output at the reference load of each sensor when connected to the converter and Is the sum of the values corrected to The details will be described below based on embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Four sensors (load cells) as shown in FIG.
Are arranged at positions where the vertices of a rectangle are formed, and as shown in the block diagram of FIG. 1, the output of each sensor is connected to a double integral type A / D converter via a switch controlled by a microcomputer. Are separated in ascending order of input integration time, and when input integration of all sensors is completed, reference voltage integration is performed to determine the number of clocks in the reference voltage integration period. The input integration time for each sensor is determined by the procedure described below. The output (V0i) of each sensor under no load is stored. Next, a load 4 that is four times the reference load
Apply W (g) to the center of the mounting table (W (g)
Then, the output (V0i) at the time of no load is subtracted from the output (Vfi) of each sensor to obtain the output: Vi = Vfi-V0i for the reference load W (g) of each sensor. Sum of outputs of each sensor with respect to the reference load: A
= ΣVi, and A / 4 (average output of four sensors with respect to the reference load) is divided by each output to obtain a correction coefficient A / 4Vi for the integration time of each sensor. The sum A of the outputs of the four sensors with respect to the reference load is a total load of 4 W
In the case of (g), a number equal to the standard output 4V, and a reference input integration time clock number N where 4V = A * N are obtained. The above is a description of a method for determining the correction coefficient of the input integration time of each sensor in units of weighing scale. A reference load is applied to each sensor, an output: Vi is obtained, and a standard voltage at the reference load: V Is multiplied by the reciprocal of Vi to obtain a correction coefficient: V / Vi for each sensor. Hereinafter, the operation of the present invention will be described in detail. For convenience of description, S1, S2, S3, and S4 are set in order from the one with the largest output to the reference load regardless of the mounting position of the sensor, and the outputs at the reference load W (g) of each sensor are V1, V2, V3, and V3, respectively. It is assumed that V4 (V1≥V2≥V3≥V4). ΣVi = A = V1 + V2 + V3 + V4, where S1 is the input integration time correction coefficient of each sensor: A / 4V1 is α S2 input integration time correction coefficient: A / 4V2 is β S3 input integration time correction coefficient Coefficient: A / 4V3 is γ S4 Input integration time correction coefficient: A / 4V4 is δ. (It has a relationship of α ≦ β ≦ γ ≦ δ) Further, the resistance value of the resistor of the integration circuit is: R, the capacitance of the capacitor is: C, and the cycle of the clock is: T. When a load of 4 W (g) is placed on the center of the mounting table (a reference load W (g) is equally applied to each sensor), the output voltage V at the end of the input integration period is It is represented by Expanding and substituting the original form for α, β, γ and δ, V = − (V1α + V2β + V3γ + V4δ) / CR * NT = − (V1A / 4V1 + V2A / 4V2 + V3A / 4V3 + V4A / 4V4) / CR * NT = −A / CR * NT. When a load of 4 W (g) is placed at an arbitrary position on the mounting table (load load ratio of each sensor S1 load load ratio = s S2 load load ratio = t S3 load load ratio = u S4 The output voltage Vh at the end of the input integration period of the load sharing ratio = v (s + t + u + v = 1) is Vh = − (4V1s + 4V2t + 4V3u + 4V4v) / CR * αNT− (4V2t + 4V3u + 4V4v) / CR * (β−α) NT− ( 4V3u + 4V4v) / CR * ([gamma]-[beta]) NT-4V4v / CR * ([delta]-[gamma]) NT, and if the same expansion as in the previous equation is performed, Vh =-(4V1s [alpha] + 4V2t [beta] + 4V3u [gamma] + 4V4v [delta]) / CR * NT =-(4V1sA / 4V1 + 4V2tA / 4V2 + 4V3uA / 4V3 + 4V4 vA / 4V4) / CR * NT =-(sA + tA + uA + vA) / CR * NT =-(s + t + u + v) A / CR * NT where s + t + u + v = 1 Therefore, Vh = −A / CR * N
T and V = Vh. In the above, the outputs of all the sensors are simultaneously input to one double-integration type A / D converter, and the input integration time of the output of each sensor is controlled in proportion to the reciprocal of the output at the reference load of each sensor. (Separating in order from the one with the shortest input integration time) indicates that the measurement is always performed accurately without being affected by the load placement position. On the other hand, consider the case where the load fluctuates and the load share of each sensor changes within the measurement period. When the load is oscillating, the load share ratio of each sensor in one measurement cycle: s, t, u, v is s + t + u
While satisfying + v = 1, each burden ratio can take a value from 0 to 1 (0 ≦ s, t, u, v ≦ 1). In the scanning method, the output at the time of moving by 1/4 of the measurement period is added, and it is not possible to cope with the change of the load burden ratio of each sensor within the measurement period. It is an unstable value that cannot be handled by processing such as truncation. In the method of the present invention, most of the measurement time (variation of the outputs V1 to V4 with respect to the reference load of each sensor is several percent) does not affect the measured value even if the load burden ratio of each sensor changes. The outputs of all the sensors, which are not provided, are measured and added, and the correction period for measuring and adding only some of the sensor outputs is 10% or less of the measurement cycle. On the other hand, the fact that the correction period for measuring and adding only some of the sensor outputs is short (several tenths of the scanning method) means that the change in the load burden ratio of each sensor during this period is small. The effect of the change in the load sharing ratio on the measured value is minute, and it is possible to obtain a much more stable and accurate measured value than the conventional scanning method, and the measurement value averaging process was fully satisfactory. Measurements can be obtained. As described above, according to the present invention, a high-precision flat weighing scale sufficiently coping with a oscillating load requires special skills without increasing production costs. You can create without.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of an embodiment of the present invention. FIG. 2 is a block diagram of a conventional example (scanning method). FIG. 3 is a block of another conventional example (individual A / D conversion method). 4A is a plan view of a main part of the weighing scale according to the embodiment of the present invention. FIG. 4B is a cross-sectional view taken along line AA of FIG. 4A. FIG. 4C is a side view of FIG. FIG. 6 is a diagram illustrating a time-voltage relationship during an input integration period of the double integration type A / D converter in the embodiment. [Description of References] 1 Base 2 Mounting table 3 Sensor 4 Focused point 5 Strain gauge 6 Leg 7 Set screw

Claims (1)

(57) [Claims] The weight of an object to be weighed on a mounting table is measured by a plurality of sensors, and the total output of each sensor is converted to a digital value by an integrating A / D converter, and the weight of the object to be weighed is measured. In a weighing scale that digitally displays the weight of a sensor, the outputs of all the sensors are simultaneously input to one integrating A / D converter, and the input integration time for each sensor is proportional to the reciprocal of the output of each sensor at the reference load. A multi-sensor weighing scale characterized in that the weighing scale is controlled.