JP2006266802A - Measurement apparatus with linearizer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce an error value in a span without increasing the number of calibration points. <P>SOLUTION: The measurement apparatus with a linearizer for an object to be measured having nonlinear characteristics performs broken line approximation on a measurement range with a broken line having a plurality of folding points, performs calibration for adjusting a measured value to a true value on each folding point, sets a predetermined conversion coefficient within a span between folding points, and determines a display value by multiplying the measured value by the conversion coefficient. At part of the folding points, a calibration point and a switching point of the conversion coefficient are set to be in a dissimilar relationship. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a measuring device called, for example, a digital voltmeter, and more particularly to a measuring device including a linearizer for matching a deviation between a voltage value to be measured and a display value.

  FIG. 5 shows a schematic configuration of the digital voltmeter. In the figure, 1A indicates a voltage measurement terminal, and 1B indicates a current measurement terminal. When measuring voltage, the switch 2 selects the voltage measuring terminal 1A, and when measuring current, the switch 2 selects the current measuring terminal 1B. At the time of current measurement, the current to be measured is input to the current-voltage converter 3, the current to be measured is converted into a voltage value, and the voltage value is input to the AD converter 5 through the range switching circuit 4. The measured voltage AD-converted by the AD converter 5 matches the measured value and the displayed value by the linearizer 6, and the matched value is input to the digital display 7 and displayed as a measured value.

Here, an example of a linearizer calibration method will be briefly described. For example, the effective value V _RMS of AC voltage is

により求められる。
（１）式より、実効値変換後の値は図６図線Ａに示すように入力電圧V(t)に対して直線でなく、二次曲線Ａで表わされる。ここで仮に曲線Ａの各点を微細（ほぼ連続的）に分割し、各分割点毎に校正を施すことにより変換精度の高いリニアライザを得ることができる。

Is required.
From the equation (1), the value after the effective value conversion is represented not by a straight line but by a quadratic curve A with respect to the input voltage V (t) as shown in FIG. Here, it is possible to obtain a linearizer with high conversion accuracy by dividing each point of the curve A minutely (substantially continuously) and calibrating each divided point.

  However, when a large number of calibration points are taken, a lot of labor is required for calibration, and an increase in cost is inevitable. That is, a standard voltage source capable of generating a voltage having a known value is prepared for calibration, a voltage corresponding to each point on the curve A is generated from the reference voltage, and the displayed value is changed according to the applied state of the voltage. This refers to the work of adjusting the conversion coefficient of the linearizer so that it becomes the voltage value during input. Adjustment of the conversion coefficient of the linearizer is performed, for example, by an operation such as cutting a resistor on a semiconductor substrate with a laser beam or the like. For this reason, if a large number of calibration points are taken, there arises a disadvantage that the manufacturing cost increases. Furthermore, even if such a linearizer can be realized, when it is incorporated in a measuring device, a conversion coefficient is determined between each calibration point (span), and if a measured voltage within the span is input, it is input. The displayed value must be calculated by multiplying the measured voltage by the conversion factor. In this case, since it is necessary to determine which conversion coefficient is used for calculation, there is a disadvantage that the more calibration points, the longer the processing for the determination.

Therefore, in general, a calibration method is adopted in which a representative voltage position on the curve A is defined as a break point, and a conversion result having a value close to the curve A is obtained by folding line approximation. A point FS shown in the curve A shown in FIG. 6 is a full scale point, 1/10 FS is a fold line indicating 1/10 of the full scale FS, 1/100 FS is a fold point indicating 1/100 of the full scale FS, and zero Represents the zero point. Calibration is performed at these four break points FS, 1/10 FS, 1/100 FS, and zero, and curve A is approximated by a broken line. Incidentally, the zero point is stored obtains a measurement value output from the AD converter 3 while the 0 input voltage as an offset voltage e _n (noise component), subtracts the offset voltage e _n for each breakpoint , Remove the effect of offset voltage. By performing calibration at each break point 1/100 FS, 1/10 FS, and FS on the curve A, conversion coefficients used in the respective spans M1, M2, and M3 divided by the break points are determined.

As described above, if a large number of calibration points are used, the conversion accuracy of the linearizer is improved, but a lot of labor is required for calibration, resulting in a problem that the manufacturing cost increases. Therefore, it is convenient if the number of folding points (number of calibration points) that approximates the broken line as much as possible can be reduced. However, if a wide span is taken between the calibration points, the rate at which an error value generated within the span range approaches or exceeds the allowable value increases.
FIG. 7 shows the state of errors that occur between calibration points. In this type of measuring apparatus, it is considered “good” if a predetermined measurement accuracy is satisfied between full scale FS and about 1/20 FS. Even in the measurement region between Zero and 1 / 20FS, a large error value is allowed. For this reason, how the error occurs from the full scale FS to 1/100 FS in the curve A shown in FIG. 6 will be described. Since the occurrence of an error between the full scale FS and the break point 1/10 FS in the curve A is slight, here, an error occurring in a span between the break points 1/10 FS to 1/100 FS will be described. FIG. 7 shows an error occurring in the span M2 between the break points 1/10 FS and 1/100 FS. At the break points 1/10 FS and 1/100 FS, the error is fixed to almost zero by calibration, but an error (deviation) occurs as shown by a curve B in FIG. Since this error swells in the vicinity of 1/20 FS, it may exceed the allowable value. Even if the error exceeds the allowable value between Zero and 1/20 FS, this is allowed. However, it is not permissible to exceed the allowable value in a measurement region with a value greater than 1 / 20FS. As shown by the curve B, if it is in a state close to the allowable value near 1/20 FS, the rate of occurrence of a product that exceeds the allowable value among the mass-produced products is increased, resulting in a disadvantage that the yield is deteriorated.

For this reason, in order to reduce the error in the vicinity of 1/20 FS, for example, it is conceivable to set the break point 1/20 FS as a calibration point. When the break point 1 / 20FS is selected as the calibration point, the error at the break point 1 / 20FS becomes zero, but this time, the error between the 1 / 20FS point and zero becomes large.
This is shown in FIG. Curve B shown in FIG. 8 is an error curve when zero and break points 1/100 FS and 1/10 FS are calibration points, and curve C is zero, when break points 1/20 FS and 1/10 FS are calibration points. The error curve, curve D, shows the error curves when zero, the break point 1/40 FS, and 1/10 FS are used as calibration points. As apparent from the curves C and D, when the calibration point is changed from the break point 1/100 FS to the break point 1/40 FS or the break point 1/20 FS, the error value between the break points 1/10 FS to 1/20 FS can be reduced. However, the error value in the observation region smaller than the break point 1 / 20FS becomes abnormally large.

Although the error value between Zero and 1 / 20FS is “good” even if it is large, the curve C is zero, the error curve when the break points 1 / 20FS and 1 / 10FS are the calibration points, and the curve D is zero. And error curves when the break point 1/40 FS and 1/10 FS are set as calibration points are shown. As apparent from the curves C and D, when the calibration point is changed from the break point 1/100 FS to the break point 1/40 FS or the break point 1/20 FS, the error value between the break points 1/10 FS to 1/20 FS can be reduced. However, the error value of the measurement region smaller than the break point 1 / 20FS becomes abnormally large.
Although the error value between Zero and 1 / 20FS is “good” even if it is large, the error values shown by the curves C and D are too large. Therefore, as a target, it is desirable that the error value between 1/10 FS and 1/20 FS is sufficiently smaller than the allowable value, and the error value between Zero and 1/20 FS also exhibits an error characteristic indicating a reasonably small value.

  An object of the present invention is to provide a measuring apparatus including a linearizer that can be suppressed to an error value sufficiently smaller than an allowable error value in a measurement region where accuracy is required, and can be suppressed to a small error value even in a measurement region where accuracy is not required. It is what we are going to propose.

A measuring apparatus having a linearizer according to the present invention performs a calibration for approximating a measuring object having a nonlinear characteristic by a nonlinear curve having a plurality of bending points and matching a measured value to a true value at each folding. In addition, a linearizer that determines the specified conversion coefficient within the span between the breakpoints and multiplies the measurement value by the conversion coefficient to obtain the displayed value is provided. It is characterized by being set to a mismatch function.
The measuring apparatus having the linearizer according to the present invention further determines, at the conversion coefficient switching point, a span length that generates an error value close to or exceeding an allowable error value in each span approximated by a broken line. The calibration is characterized in that the calibration point is displaced in a direction shorter than the span length.

  According to the measuring apparatus including the linearizer according to the present invention, it is possible to suppress an error in a span in which a relatively large error is generated to a small value without increasing the number of calibration points. As a result, there is an advantage that improvement in measurement accuracy can be realized while suppressing an increase in cost.

  The measuring device provided with the linearizer according to the present invention is best applied to a digital voltmeter.

FIG. 1 shows an example of a linearizer calibration method used in a measuring apparatus equipped with a linearizer according to the present invention. In the figure, FS is the full scale value of the measurement range, 1/10 FS is the break point set to 1/10 of the full scale FS, 1/40 FS is the break point set to 1/40 of the full scale FS, 1 / 100FS represents a conversion coefficient switching point set to 1/100 of the full scale FS, and Zero represents a zero point.
In other words, in the present invention, in order to suppress the occurrence of an error particularly between 1/100 FS and 1/10 FS in each of the spans M1, M2, and M3 divided at each break point 1/10 FS to 1/100 FS, In order to reduce the length of the span where the breakpoint 1 / 100FS was set as the calibration point, the calibration point should be moved from the breakpoint 1 / 100FS to the breakpoint 1 / 40FS in this case. It is a feature. However, calibration is not performed at the break point 1/100 FS, and is used as a conversion coefficient switching point. That is, in the span M1 between Zero and 1/100 FS, the conversion coefficient of the linearizer is A1,
In the span M2 between 1 / 100FS and 1 / 10FS, the conversion coefficient is A2,
In the span M3 between 1/10 FS and FS, the conversion coefficient is A3,
The conversion coefficient for each span is set so that

  When the calibration point and the conversion coefficient switching point are set in this way, an error between 1/100 FS and 1/10 FS is generated between 1/40 FS and 1/10 FS as shown by curve E in FIG. The error value can be kept small compared to the allowable value. That is, 1/40 FS is used as the calibration point, and the error generation is suppressed to almost zero at 1/40 FS, and the 1/100 FS to 1/10 FS span is shortened to the 1/40 FS to 1/10 FS span. The error between / 40FS and 1 / 10FS can be reduced. At the same time, since 1/100 FS is not a calibration point, the error value moves to the positive side of the opposite polarity as shown in FIG. At the same time, the error between Zero and 1 / 100FS becomes larger than that in the case of the conventional curve B shown in FIG. 7, but can be made smaller than those in the curves C and D shown in FIG. Such an error is allowed in the region.

FIG. 3 is a table showing the relationship between the conventional calibration method and the calibration points and break points (coefficient switching points) used in the present invention. FIG. 4 shows the measured values of the input value and the display value in comparison with the prior art and the present invention.
In the above description, the linearizer for measuring the effective value has been described as an example. However, the present invention is not limited to this, and can be applied to other linearizers for measurement. In short, when the error generation is close to the allowable value within the specified span, the coefficient switching point is left as it is, and only the calibration point is moved in the direction that shortens the span, and the error value within the span is suppressed. It is characterized by this.

  The linearizer calibration method according to the present invention and the measuring apparatus equipped with the linearizer constructed by this calibration method are utilized in the field of digital voltmeters.

The graph for demonstrating the calibration method of the linearizer used for the measuring apparatus provided with the linearizer by this invention. The graph for demonstrating the error generation characteristic of the linearizer comprised with the linearizer calibration method shown in FIG. The figure for demonstrating the difference between the calibration point used by this invention and a prior art, and a coefficient switching point. The figure which shows the example of an actual measurement of the display value obtained by this invention and a prior art. The block diagram for demonstrating the outline | summary of a digital voltmeter. The graph for demonstrating a prior art. The graph for demonstrating the characteristic of the error which generate | occur | produces by a prior art. The graph for demonstrating the inconvenience at the time of shortening both a calibration point and a coefficient switching point in the span where a big error generate | occur | produces.

Explanation of symbols

1/10 FS to 1/100 FS Breakpoint 4 Range switching circuit M1, M2, M3 Span 5 AD converter
1A Voltage measurement terminal 6 Linearizer
1B Current measurement terminal 7 Digital display
2 changeover switch
3 Current-voltage converter

Claims (2)

For a measurement object with nonlinear characteristics, the measurement range is approximated by a broken line with multiple break points, calibration is performed to match the measured value to the true value at each break point, and within the span between the break points. Then, in a measuring apparatus including a linearizer that determines a predetermined conversion coefficient and obtains a display value by multiplying the measurement value by the conversion coefficient,
A measuring apparatus provided with a linearizer, wherein a calibration point and a conversion coefficient switching point are set to be inconsistent in a part of the break point. 2. A measuring apparatus comprising the linearizer according to claim 1, wherein a span length that generates an error value close to or exceeding an allowable error value within each span approximated by a broken line is determined at a conversion coefficient switching point. A measuring apparatus provided with a linearizer, characterized in that the calibration point is displaced in a direction shorter than a predetermined span length.

Images