JPH10145231A - A / D converter and data correction method in D / A converter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a more accurate A / D converter and D / A converter by minimizing a gain / offset error and a non-linearity error which cannot be completely corrected by a conventional data correction method. Get. In an A / D converter, a corrected value is obtained from a difference between a digital signal obtained by using a least-squares method and a predetermined ideal digital signal in a converted digital signal. The gain / offset error and the non-linearity error are corrected and calculated based on the correction value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data correction method in an A / D converter used for a plant controller or the like and a data correction method in a D / A converter.

[0002]

2. Description of the Related Art FIG. 11 is a block diagram showing a circuit configuration including a conventional A / D converter, and FIG. 12 is a diagram showing its conversion characteristics. In FIG. 11, reference numeral 100 denotes a plant controller including an A / D converter 101 and a CPU 102 of the plant controller. 1 is a plant 30
This is an analog signal input to the plant controller 100 from, for example, a plurality of signals. A / D converter 10
In 1, reference numeral 2 denotes an analog input circuit for receiving an analog signal 1, and reference numeral 3 denotes one of analog signals having a plurality of channels.
A multiplexer circuit for selecting one of them, 4 a channel select signal to the multiplexer circuit 3, 5 an analog signal selected by the multiplexer circuit 3, 6 an A / D conversion circuit for converting the analog signal into a digital signal, Numeral 8 processes the digital signal 7 digitized by the A / D conversion circuit 6, and the CPU 1 of the plant controller
The CPU 9 transmits the data 9 to the data 02.

FIG. 12 is a graph showing characteristics when a correction operation is performed on a digital signal digitized in the CPU 8 shown in FIG. 11, in which the horizontal axis represents an analog value and the vertical axis represents a digital value. Expressed as a value. In FIG. 12, reference numeral 26 denotes an A / D conversion characteristic straight line before the correction operation is performed, and reference numeral 18 denotes a characteristic after the A / D conversion characteristic line 26 is subjected to the correction operation. 18 is A / D
The graph shows an ideal A / D conversion characteristic straight line in the conversion device 101.

Next, the operation will be described. In FIG. 11, a plurality of analog signals 1
Is a multiplexer circuit 3 via an analog input circuit 2.
Is input to The multiplexer circuit 3 selects one of a plurality of analog signals based on the channel select signal 4 input from the CPU 8 and inputs the selected analog signal to the A / D conversion circuit 6. The A / D conversion circuit 6 converts an analog signal into an arbitrary digital signal corresponding to the analog signal and inputs the digital signal to the CPU 8.
However, the digital signal input to the CPU 8 often does not correspond to the analog signal input first while passing through these circuits. The error is corrected by the CPU 8 calculating the error. This correction operation requires a correction value for each circuit, and in order to derive the correction value, it is necessary to know the degree of error of the A / D converter 101 in advance. Therefore, in FIG.
Digital values at two points on the A / D conversion characteristic line 26 of the / D conversion result and at two points on the ideal A / D conversion characteristic line 18 are obtained. Here, as an example, description will be made assuming that two points, ie, points a1 and a2 and points b1 and b2 are respectively taken.

From the digital values at two points on each straight line obtained above, the following conversion formulas (1), (2),
A correction value is obtained according to (3). A / A
If the correction operation is performed at an arbitrary point on the A / D conversion characteristic line 26 of the D conversion result, the A / D conversion characteristic line 26 can be made closer to the ideal A / D conversion characteristic line 18. At this time, the A / D conversion characteristic straight line 26 is represented by points a1 and a2.
Then, the vertical axis direction and the inclination are corrected so that the points b1 and b2 on the A / D conversion characteristic line 18 overlap.

[0006]

(Equation 1)

[0007]

(Equation 2)

[0008]

(Equation 3)

Note that Y is the digital value after the correction operation, X is the digital value before the correction operation, and x1, x2, y1, and y2 are the digital values shown in FIG. A and B indicate correction values.

[0010]

In the conventional data correction method described above, it can be said that the gain / offset error included in the A / D conversion result can be sufficiently corrected if the characteristic is linear. . However, in practice, this also includes a non-linearity error, and in many cases, its characteristics are curved as shown by a line 27 in FIG. As shown in the line 28, the points other than the two points a1 and a2 for which the correction values were obtained are values far from the ideal straight line 18, and the correction effect by the calculation does not increase so much, and the operation status of the plant is accurately grasped. It is difficult to output accurate control signals to a plant or a plant.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and has an A / D converter and a D / A converter.
The conversion device absorbs the gain / offset error and reduces the non-linearity error, so that a more accurate A / D
It is an object of the present invention to provide a data correction method capable of performing conversion and D / A conversion.

[0012]

According to a first aspect of the present invention, there is provided a data correction method, wherein an A / D converter converts a digital signal obtained by converting an analog signal into a digital signal using a least squares method. And a correction value obtained from a difference between the predetermined value and a given ideal digital signal, and a gain / offset error and a non-linearity error are corrected and calculated based on the correction value.

According to a second aspect of the present invention, in the A / D converter, three or more digital signal values obtained by converting three or more different input analog signal values,
A correction value is obtained from a difference between the values of the three or more digital signals obtained by converting the values of the accurate three or more input analog signals respectively corresponding to the three or more different input analog signals. , A gain / offset error and a non-linear error are corrected and calculated.

In a data correction method according to a third aspect of the present invention, in a D / A converter, a digital value obtained by adjusting an analog signal converted from a predetermined ideal digital signal to an ideal analog signal output is obtained. A correction value is obtained from a difference between a digital signal obtained by using the least squares method and an ideal digital signal given in advance, and a gain / offset error and a non-linear error are corrected based on the correction value. It is characterized by calculating.

[0015] The data correction method of the fourth invention is different from the third method.
The values of the corresponding digital signals when the analog signals of three or more points obtained by converting the values of the predetermined ideal digital signals of three or more points are converted into ideal analog signal outputs, respectively, It is characterized in that a correction value is obtained from a difference from the value of the ideal digital signal given in advance, and a gain / offset error and a non-linear error are corrected based on the correction value.

[0016]

BEST MODE FOR CARRYING OUT THE INVENTION

Embodiment 1 FIG. Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a circuit configuration including an A / D converter according to the first embodiment and a second embodiment to be described later. FIG. 2 shows an A / D according to the first embodiment.
FIG. 4 is a diagram illustrating conversion characteristics. FIG. 3 is a flowchart showing processing of the data correction method according to the first embodiment.

In FIG. 1, reference numeral 1 denotes an analog signal input from a plant 30 or the like to a plant controller 100 having an A / D converter 103 and a CPU 102, and a plurality of analog signals are input. In the A / D converter 103,
2 is an analog input circuit for receiving the analog signal 1, 3 is a multiplexer circuit for selecting one of the analog signals having a plurality of channels, and 4 is a multiplexer circuit 3.
5 is an analog signal selected by the multiplexer circuit 3, 6 is an A / D conversion circuit for converting an analog signal into a digital signal, and 8 is a digital signal digitized by the A / D conversion circuit 6. Processes the signal 7 and sends the data 9 to the CPU 102 of the plant controller.
Is the CPU that transmits the data.

The analog signal 1 from the plant 30 is input to the multiplexer circuit 3 via the analog input circuit 2. Next, the analog signal 5 selected by the multiplexer circuit 3 is converted into an m-bit digital signal 7 by an A / D conversion circuit 6 in the next stage, and input to the CPU 8.
The CPU 8 is a CPU 102 of the plant controller.
The data 9 is transmitted to. FIG. 2 shows the conversion characteristics at this time and the conversion characteristics when the data correction method of the first embodiment is used. In FIG. 2, reference numeral 17 denotes a characteristic curve of an A / D conversion result including a gain / offset error and a nonlinearity error, and shows a relationship between the analog signal 1 and the m-bit digital signal 7 in FIG. . Reference numeral 16 in FIG. 2 is a characteristic straight line derived from the characteristic curve 17 using the least squares method.
In FIG. 2, as an example, a characteristic line 16 is derived at five points a1 to a5. Reference numeral 18 denotes a characteristic line obtained by performing a correction operation on the characteristic line 16, which is an ideal characteristic line as a result of the A / D conversion. Reference numeral 19 denotes a characteristic curve indicating a final correction result obtained by the data correction method.

Next, the operation will be described. In FIG. 1, a plurality of analog signals 1
Is a multiplexer circuit 3 via an analog input circuit 2.
Is input to The multiplexer circuit 3 selects one of a plurality of analog signals based on the channel select signal 4 input from the CPU 8 and inputs the selected analog signal to the A / D conversion circuit 6. The A / D conversion circuit 6 converts the analog signal 5 into a corresponding digital signal 7 and inputs the digital signal 7 to the CPU 8.
However, the digital signal 7 input to the CPU 8 often does not correspond to the analog signal input first while passing through these circuits. When this is represented in a graph, it becomes like a characteristic curve 17 in FIG. In the first embodiment, the characteristic curve 17 is changed to an ideal characteristic straight line 18.
At the point where the influence of the error is minimized, and the least squares method is used for the correction calculation.

Hereinafter, the description will be continued with reference to FIG. Characteristic curve 1
7, a characteristic straight line 16 is derived in advance by the least square method. At this time, the least-squares method draws a perpendicular at each of the points a1 to a5 and draws a distance L1 to the intersection with the characteristic line 16.
A characteristic line 16 that minimizes the sum of L5 is obtained, whereby the characteristic line 16 can be a line that minimizes the effect of the nonlinearity error. Thereafter, the points a1 'and a5' on the characteristic line 16 and the ideal line 18
The correction value is obtained from the two points b1 and b5. And
By correcting and calculating an arbitrary value of the characteristic curve 17 using the correction value, the gain / offset error is corrected, and the characteristic curve 19 in which the influence of the nonlinearity error is minimized can be derived. Note that the correction operation performed after the least squares method uses the correction operation formula of the related art.

Next, this series of operations will be described with reference to the flowchart of FIG. As described above, in this data correction method, since a correction value must be calculated in advance,
The operation modes of the A / D converter 103 include:
It is assumed that two modes of “correction value calculation mode” and “A / D conversion mode” are prepared (step 301). First, since it is necessary to obtain a correction value for each channel due to a circuit error in the A / D converter 103, the correction value calculation mode is set. In the correction value calculation mode, first, the CPU 8 outputs the channel select signal 4 in order to determine which of the plurality of channels a correction value is to be calculated (step 302). At this time, since it is necessary to input an accurate value to the analog signal 1, a voltage generator (not shown) or the like is connected to the analog input circuit 2. An accurate analog value corresponding to the point b1 in FIG. 2 is input from the voltage generator (step 303), and a digital value after A / D conversion is read (step 304). This digital value is point a1 in FIG. The same method is applied to other points to determine the values of points a2 to a5 in FIG. Point a1
The characteristic curve 17 shown in FIG. Next, the straight line 16 in FIG. 2 is obtained from these values by using the least squares method (step 305), and the point a of the minimum value is obtained.
1 ′ and the maximum point a5 ′ are obtained. These two points a1 ', a
If the minimum value point b1 and the maximum value point b5 on 5 ′ and the ideal characteristic line 18 are known, correction values A and B can be obtained (step 306), and the obtained correction values are stored in a memory IC or the like. (Step 307). Thus, the correction value of the corresponding channel has been calculated. If it is necessary to calculate the correction value of another channel, the channel select signal 4 from the CPU 8 is set to an arbitrary value, and the same method as described above is used. Calculate the correction value.

When the correction value of each channel can be calculated, the operation mode of the A / D converter 103 is set to "A / D conversion mode". In this A / D conversion mode, the actual plant 3
Since 0 is a mode for normal use, a signal connection from the actual plant 30 is connected to the analog signal 1. Then, even if an arbitrary analog signal is input from the plant 30, a correction operation is performed using the above-described correction value (step 30).
8, 309), the gain / offset error is completely corrected, and the non-linearity error is corrected to a value that minimizes the effect. Therefore, the characteristic shown in curve 19 in FIG. it can.

Since the analog signal from the plant can be received with higher accuracy by such a data correction method, the operating condition of the plant can be accurately grasped, and the controllability of the controller is improved.

Embodiment 2 FIG. In the first embodiment,
The characteristic straight line 1 is obtained from the characteristic curve 17 of FIG.
6, which is an effective method when the value of the nonlinearity error is relatively small. However, when a very large non-linearity error occurs, that is, when the characteristic curve is steep, etc., this method can find the minimum value of the linearity error, but even if it can be found, No correction can be made. Therefore, in the second embodiment, the correction value is calculated for each of the points a1 to a5 on the characteristic curve 17 without using the least squares method, and the correction calculation is performed. Even in such a case, they can be corrected to approach the ideal value.

This will be described with reference to FIG. FIG.
FIG. 4 shows the correction values obtained for the points a1 ′ and a5 ′ of FIG.
It is determined for each of the points a1 and a2, between the points a2 and a3, between the points a3 and a4, and between the points a4 and a5. If the correction operation is performed within the range of each point, the nonlinearity as shown in FIG. Since the error can be reduced and the gain / offset error can be corrected at the same time, it can be made closer to the ideal characteristic line 18.

This series of operations will be described with reference to the flowchart of FIG. In the data correction method according to the second embodiment, as in the first embodiment, since the correction value must be calculated in advance, the mode of the A / D converter 103 (see FIG. 1) is “ Correction value calculation mode "and" A / D
It is assumed that two modes of "conversion mode" are prepared (step 501). First, since it is necessary to obtain a correction value for each channel due to each circuit error in the A / D device, a correction value calculation mode is set. In the correction value calculation mode, first, the CPU 8 outputs the channel select signal 4 to determine which of the plurality of channels a correction value is to be calculated (step 502). Since it is necessary to input an accurate value, a voltage generator (not shown) is connected to the analog input circuit 2 (see FIG. 1).
An accurate analog value corresponding to the point b1 is input (step 503), and a digital value after A / D conversion is read (step 504). This digital value is point a1 in FIG. Next, the point b in FIG.
Input an accurate analog value corresponding to 2 and read the digital value after A / D conversion. This digital value is point a2 in FIG. The same method is applied to points b3, b4, and b5 in FIG. 4 to determine the values of points a3, a4, and a5. Points a1
After reading the value of a5, between points a1 and a2,
Correction values are calculated for three points, between points a3 and a4, and between points a4 and a5 (steps 505 to 509). Also, when calculating the correction value for other channels,
After setting the channel select signal 4 from the CPU 8 to an arbitrary value, a correction value is calculated in the same manner as described above.

When the correction value of each channel can be calculated, the operation mode of the A / D converter 103 is set to "A / D conversion mode". Since this mode is a mode in a case where the actual use is performed in the actual plant 30, the analog signal 1 is connected to the signal from the actual plant 30. And
Even if an arbitrary analog signal is input from the plant 30,
By performing the correction operation (steps 510 and 511) using the above-described correction values, the gain / offset error is completely corrected, and the non-linearity error is also corrected so as to approach the ideal straight line 18. Therefore, the curve 20 in FIG. Can be obtained. In the above description, the correction value is calculated at five points a1 to a5. However, if the number of points is increased, the correction value can be made closer to the ideal straight line.

Since the analog signal from the plant can be received with higher accuracy by such a data correction method, the operation state of the plant can be accurately grasped, and the controllability of the controller is improved.

Embodiment 3 In the first and second embodiments, the A / D conversion has been described. However, the same data correction method can be applied to the D / A conversion. Embodiment 3 will be described with reference to FIGS. 6, 7, and 8. FIG. FIG. 6 is a block diagram including the circuit configuration of the D / A converter 201 according to the third embodiment. The control signal output from the CPU 202 of the plant controller to the plant 30 is performed via the D / A converter 201. . From the plant controller 200 to the D / A converter 2
The control data 15 sent to the CPU 8 is digital data, and the CPU 8 transmits the control data 15 to the D / A conversion circuit 1 in the next stage.
0 is output as a digital signal (digital data) 7. Analog signal 11 converted by D / A conversion circuit 10
Is input to the multiplexer circuit 12, and is input to the analog output circuit 13 of an arbitrary channel by the channel select signal 4 from the CPU 8. Then, the analog output signal 14 is output to each plant 30. At this time, while the analog output signal 14 passes through these circuits, a gain / offset error and a non-linear error occur as in the case of the A / D converter. FIG. 7 shows the conversion characteristics at this time and the conversion characteristics when the data correction method of the third embodiment is used. The correction calculation is performed by the CPU 8, and outputs an m-bit digital signal 7 in which a circuit error is expected in advance.

FIG. 7 is a graph showing D / A conversion characteristics according to the third embodiment. In FIG. 7, reference numeral 22 denotes a characteristic curve of a D / A conversion result including a gain / offset error and a nonlinearity error.
And the relationship between m and the m-bit digital signal 7 is shown.
Reference numeral 21 in FIG. 7 is a characteristic straight line derived from the characteristic curve 22 by using the least squares method. In FIG.
A characteristic straight line 21 is derived at the point. Reference numeral 23 denotes a characteristic straight line obtained by performing a correction operation on the characteristic straight line 21, which is an ideal straight line of the D / A conversion result. And 24
Is a characteristic curve showing a final correction result obtained by this correction method.

The way of thinking up to this point is almost the same as that of the A / D converter according to the first embodiment. However, since the realization method is different, a series of operations will be described with reference to the flowchart of FIG. As in the case of the first embodiment, in the data correction method of the third embodiment, since the correction value must be calculated in advance, the operation mode of the D / A converter 201 is “correction value”. Calculation mode "and" D / A
It is assumed that two modes of “conversion mode” are prepared (step 801). First, since it is necessary to obtain a correction value for each channel due to a circuit error in the D / A converter 201, the correction value calculation mode is set. A high-precision voltmeter (not shown) is connected to the analog output circuit 13. In the correction value calculation mode, first, in order to determine which of the plurality of channels a correction value is to be calculated. 7 outputs the channel select signal 4 from the CPU 8 (step 802), then outputs the value of the point b1 in FIG. 7 from the CPU 8, and checks the value of the previously connected high-precision voltmeter (step 803). CPU 8
Is adjusted (step 804). This value is point a1 in FIG. In the same way, points b2, b3, b
4 and b5, the points a2, a3 and a4 in FIG.
Find the value of a5. This is the characteristic curve 22 in FIG.

Next, the characteristic line 21 of FIG. 7 is obtained from the values of these points a1 to a5 by the least square method, and the minimum point a1 'and the maximum point a5' are obtained (step 80).
5). If these two points a1 'and a5' are known, the correction values C and D
(Step 806), and the obtained correction value is stored in a memory IC or the like (step 80).
7). Thus, the correction value of the corresponding channel has been calculated. If it is necessary to calculate the correction value of another channel, the channel select signal 4 from the CPU 8 is set to an arbitrary value, and the same method as described above is used. Calculate the correction value.

When the correction value of each channel can be calculated, the operation mode of the D / A converter 201 is set to "D / A conversion mode". In this D / A conversion mode, the actual plant 3
Since 0 is a mode for normal use, the connection of the analog output signal 14 is set to a signal connection from the actual plant 30. Then, even if an arbitrary digital signal is output from the plant controller 200, by performing a correction operation using the above-described correction value (steps 808 and 809),
Since the gain / offset error is completely corrected, and the nonlinearity error is corrected to a value that minimizes the effect, the characteristic can be obtained as shown by the characteristic curve 24 in FIG.

With such a data correction method, a control signal from the plant controller can be sent to the plant with higher accuracy, leading to an improvement in the controllability of the controller.

In the D / A conversion, the method of calculating the correction value derived from the circuit error is different from the case of the A / D conversion, and can be calculated by the following equations (4), (5), and (6).

[0036]

(Equation 4)

[0037]

(Equation 5)

[0038]

(Equation 6)

Note that Y is the digital value after the correction operation, X is the digital value before the correction operation, C and D are the correction values, x1, x
2, y1, y2 and digital values (see FIG. 12).

Embodiment 4 FIG. In the third embodiment, the characteristic straight line 21 is obtained from the characteristic curve 22 of FIG.
This is an effective method when the value of the nonlinear error is relatively small. However, if a very large non-linear error occurs, that is, if the characteristic curve is steep, this method can find the minimum value of the linearity error, but even if it can be corrected, Can not do. This is the same as in the case of the A / D converter.

Therefore, in the fourth embodiment, the correction value is calculated for each of the points a1 to a5 on the characteristic curve 22 without using the least square method, and the correction calculation is performed. Even if they do, they can be corrected to approach the ideal value. This will be described with reference to the graph shown in FIG. In FIG. 9, the correction values obtained for the points a1 ′ and a5 ′ in FIG.
It is determined for each of a3, a3 and a4, and a4 and a5, and if a correction operation is performed within the range of each point, the nonlinearity error can be reduced as shown in FIG. Since the offset error can also be corrected, the ideal linear characteristic 23 can be made closer.

This series of operations will be described with reference to the flowchart of FIG. In the data correction method according to the fourth embodiment, the correction value must be calculated in advance, as in the third embodiment. Therefore, the mode of the D / A converter 201 (see FIG. 6) is “ Correction value calculation mode ”and“ D /
It is assumed that two modes of "A conversion mode" are prepared (step 901).
Since it is necessary to obtain a correction value for each channel due to a circuit error in the circuit, a correction value calculation mode is set, and a high-precision voltmeter (not shown) is connected to the analog output circuit 13 (see FIG. 6). deep. In the correction value calculation mode, first, the CPU 8 outputs the channel select signal 4 in order to determine which of the plurality of channels a correction value is to be calculated (step 902). Next, the value of the point b1 is output from the CPU 8 and the output data of the CPU 8 is adjusted to an arbitrary value while checking the value of the high-precision voltmeter connected in advance (step 903) (step 90).
4). This value is point a1 in FIG. Point b in the same way
2, b3, b4, b5, and the points a2,
The values of a3, a4, and a5 are obtained. This is the characteristic curve 22 in FIG. After reading the values of points a1 to a5, point a
1, a2, point a2, a3, point a3, a4, point a
The respective correction values C and D are calculated between 4 and a5 (steps 905 to 909). When obtaining correction values for other channels, the channel selection signal 4 from the CPU 8 is set to an arbitrary value, and then the correction values are calculated in the same manner as described above.

When the correction value of each channel can be calculated, the operation mode of the D / A converter 201 is set to "D / A conversion mode". In this D / A conversion mode, the actual plant 3
Since 0 is a mode for normal use, the connection of the analog output signal 14 is a signal connection to the actual plant 30. Even if an arbitrary control signal is output from the plant controller 200, the gain / offset error is completely corrected by performing the correction operation (steps 910 and 911) using the above-described correction value, and the nonlinearity error is also reduced. Since the correction is made to approach the ideal characteristic line 23, FIG.
The characteristic shown by the curve 25 of FIG. In the above description, the correction values are calculated at five points a1 to a5.
If the number of points is increased, the ideal characteristic straight line 23 can be brought closer.

According to such a data correction method, the control signal from the plant controller can be sent to the plant with higher accuracy, which leads to an improvement in the controllability of the controller.

[0045]

As described above, according to the first aspect, A
In the / D converter, a correction value is obtained from a difference between a digital signal obtained by using the least squares method of the converted digital signal and a value of a given ideal digital signal,
Since the gain / offset error and the non-linearity error are calculated based on the correction value, the gain / offset error and the non-linearity error which cannot be corrected by the conventional data correction method can be minimized. And an effect of providing a more accurate A / D converter can be obtained.

According to the second aspect of the present invention, in the A / D converter, the values of the three or more digital signals obtained by converting the values of the three or more different input analog signals are respectively different from the values of the three different digital signals.
A correction value is obtained from a difference between the value of the digital signal of three or more points obtained by converting the value of the input analog signal of three or more points corresponding to the value of the input analog signal of the point or more, and based on the correction value Since the gain / offset error and the non-linearity error are calculated, the gain / offset error and the non-linearity error that could not be corrected by the conventional data correction method can be minimized.
In addition, the effect is obtained even when the non-linearity error is extremely large, so that an effect that a more accurate A / D converter can be provided can be obtained.

According to the third aspect, in the D / A converter, the converted analog signal is calculated by a difference between a value of an analog signal obtained by using the least squares method and a value of a predetermined ideal analog signal. Since the correction value is obtained and the gain / offset error and the non-linearity error are corrected and calculated based on the correction value, the gain / offset error and the non-linearity error which cannot be corrected by the conventional data correction method are calculated. This has the effect of minimizing it and providing a more accurate D / A converter.

According to the fourth invention, in the D / A converter, the values of the three or more analog signals obtained by converting the values of the three or more different input digital signals, respectively, and the values of the three or more different input digital signals are different. Correction values are obtained from differences between the values of the three or more analog signals obtained by converting the values of the three or more accurate input digital signals corresponding to the respective signal values, and the gain / offset error and the Since the non-linearity error is calculated, the gain / offset error and the non-linearity error that could not be corrected by the conventional data correction method can be minimized. Even if it is very large, the effect is obtained, so that an effect that a more accurate D / A converter can be provided is obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram including a circuit configuration of an A / D converter according to Embodiments 1 and 2 of the present invention.

FIG. 2 is a graph showing A / D conversion characteristics according to the first embodiment.

FIG. 3 is a flowchart relating to the operation of the first embodiment.

FIG. 4 is a graph showing A / D conversion characteristics according to the second embodiment.

FIG. 5 is a flowchart relating to the operation of the second embodiment.

FIG. 6 is a block diagram including a circuit configuration of an A / D converter according to Embodiments 3 and 4 of the present invention.

FIG. 7 is a graph showing D / A conversion characteristics according to the third embodiment.

FIG. 8 is a flowchart relating to the operation of the third embodiment.

FIG. 9 is a graph showing a D / A change characteristic according to the fourth embodiment.

FIG. 10 is a flowchart relating to the operation of the fourth embodiment.

FIG. 11 is a block diagram including a circuit configuration of a conventional A / D converter.

FIG. 12 is a graph showing a conventional A / D conversion characteristic.

FIG. 13 is a graph showing a conventional A / D conversion characteristic.

[Explanation of symbols]

1 analog signal, 2 analog input circuit, 3, 12
Multiplexer circuit, 4 channel select signal, 5
Analog signal, 6 A / D conversion circuit, 7 m bit digital signal, 8 CPU, 9,15 data, 10 D
/ A conversion circuit, 11 analog signals, 13 analog output circuits, 14 analog output signals, 30 plants, 1
00, 200 Plant controller, 102, 202
CPU of plant controller, 103 A / D converter, 201 D / A converter.

Claims (4)

[Claims] 1. An A / D converter for converting an input analog signal into a digital signal and outputting the digital signal, the digital signal obtained by converting the analog signal into a digital signal using a least square method. And a correction value obtained from a difference between a predetermined value and an ideal digital signal given in advance, and a gain / offset error and a non-linearity error are corrected and calculated based on the correction value. Data correction method. 2. An A / D converter for converting an input analog signal into a digital signal and outputting the digital signal.
The values of the three or more digital signals obtained by converting the values of the input analog signals of three or more points and the values of the three or more input analog signals corresponding to the different values of the three or more input analog signals are respectively calculated. A data in an A / D converter, wherein a correction value is obtained from a difference between the converted digital signal values of three or more points and a gain / offset error and a non-linear error are corrected based on the correction value. Correction method. 3. A D / A converter for converting an input digital signal into an analog signal and outputting the analog signal, wherein an analog signal converted from a given ideal digital signal is adjusted to an ideal analog signal output. The digital value at that time is obtained, and a correction value is obtained from a difference between a digital signal obtained from the digital signal using the least squares method and a given ideal digital signal. Based on the correction value, a gain / offset error and A data correction method in a D / A converter, wherein a correction operation is performed for a nonlinearity error. 4. A D / A converter that converts an input digital signal into an analog signal and outputs the analog signal.
The values of the corresponding digital signals when the analog signals of three or more points obtained by converting the values of the predetermined ideal digital signals of three or more points are converted into ideal analog signal outputs, respectively, A D / A converter is characterized in that a correction value is obtained from a difference between the above-mentioned predetermined ideal digital signal value and a gain / offset error and a non-linearity error are corrected based on the correction value. Data correction method.