JPS63111408A - Digital position detector - Google Patents

Abstract

PURPOSE:To realize high-accuracy position detection by generating a position signal by adding a correcting value in a correction table to the output of an original position detector. CONSTITUTION:A resolver 2 is coupled directly with the rotating shaft of a body 1 of rotation to be measured and a digital position signal is obtained by an R/D converter 4 which performs A/D conversion as to input and output signals of the object body of rotation and the resolver. Addresses of a memory 6 are accessed according to the output of the R/D converter 4. An adder 7 adds output data theta' of the R/D converter and a correcting value DELTAtheta of a position stored in the memory 6 corresponding to the data to obtain the high-accuracy position signal theta. A data processor 8 is connected to the R/D converter 4, counters 51 and 52, the memory 6, the adder 7, and a common bus 9 through a buffer to input and output data and performs various kinds of arithmetic operations, etc. Thus, the theta is made high in accuracy for the output theta' of the R/D converter 4.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides highly accurate detection by electrically correcting the detection error of a position detector that generates a digital signal in accordance with the rotation angle of a rotating body. The present invention relates to a digital position detection device capable of position detection.

[Conventional technology]

Conventionally, as a method for detecting the rotation angle of a rotating body, the eighth
As shown in the example in the figure, a synchro electric machine or a resolver 2 is directly connected to the rotating shaft of the rotating body 1 to be measured, and a digital position signal is generated by, for example, an R/D converter 4 that performs analog/digital conversion from these input/output signals. There are known ways to obtain it.

[Problems to be solved by the invention] However, due to problems such as non-linearity of the resolver and eccentricity of the resolver mounting, the detected position value usually differs from the true position of the rotating body to be measured. FIG. 9 shows an example of the relationship between the true position and the detection value of a conventional position detection device.

In other words, a detection error occurs at one or multiple cycles for one rotation of the rotating body to be measured. This error cannot be removed even if the number of pits in the R/D converter is increased, and even if the number of pits is increased, the detected value shown in FIG. 9 will only be output with a smaller resolution. The rotational speed of the rotating body is determined by detecting the positional deviation per unit time. The speed detection accuracy in this case depends on the position detection accuracy, and highly accurate position detection is required for high-g speed detection. In recent years, the demand for position and rotational speed Qjj7 control precision in general industries, machine tools, etc. has been decreasing more and more, and therefore there is a high need for higher precision in these detection devices. However, increasing the precision of electric machines such as resolvers has limitations in terms of machining, and even if higher precision than conventional machines is achieved, they are generally very expensive.

Therefore, an object of the present invention is to provide an inexpensive and highly accurate position detection device by electrically compensating for resolver nonlinearity, mounting eccentricity, etc., while using conventional electrical components.

[Means for solving problems]

A position detector that detects the rotation angle of a rotating body as a digital box, and a memory that has a table length corresponding to the data length of the position detector and stores position correction boxes corresponding to each of its output data values. A position signal is obtained by adding the output data of the position detector and the correction data of the memory corresponding to this data.

[Effect]

We focused on the fact that once the resolver, R/D transformer, etc. to be used has been determined, and the resolver is attached to the rotating object to be measured, its position detection characteristics have characteristics unique to the device. This is an attempt to accurately detect the position by adding corrections based on unique characteristics. FIG. 7 shows the basic concept of the present invention. FIG. 9A shows the relationship between the detected position value and the true position of the rotating body to be measured, and has the same characteristics as FIG. 9. A position detection device has error characteristics unique to the device at the time it is assembled. Figure (b) shows the relationship between the deviation between the true position and the detected position value with respect to the detected position value in (a). For any detected position value, a correction that is the deviation of that position detected value and the same figure (b) is added.
! ! The relationship between the detected position value in step 2 and the true position is shown in the same figure (c).
It becomes slimy.

The second position detection value maintains a proportional relationship with the true position as shown in FIG.

[Embodiments of the invention]

FIG. 1 is an explanatory diagram for explaining the principle of a position detection device according to the present invention. In addition, each direction shows an example. Here, θ is the output of the R/D converter, and is a digital value of the rotational position that changes every moment. Δθ is a table whose table length is, for example, equal to the data length of θ, and the address θ of the Δθ table is selected for any θ. This θ address contains the position correction accuracy Δ for the position signal θ.
θ is stored. As shown in the figure, θ and Δθ are added such that several pits below θ and the same number of pits above Δθ have the same weight. That is, the pits superimposed on θ of Δθ are used to correct the detection error of θ, and the other lower pits of Δθ are used to improve the resolution of position detection. The position signal θ obtained in this manner has a higher resolution than the original position signal θ, and has higher accuracy.

Note that the data of Δθ is signed data.

One of the challenges is how to create a table of Δθ. One method would be to measure the characteristics of a single unit such as a resolver and create a table with a function to correct it.

Although this method has the same type, when correcting resolver errors that occur during manufacturing, it is very expensive to measure individual error characteristics, which is impractical. Therefore, the present invention provides the following method. That is, during a free run in which the rotational speed of the rotating body is gradually decreasing due to mechanical loss only, the rate of decrease in speed per unit time can be considered to be constant for a short period of time during which the rotating body rotates at most several times.

Under these conditions, by examining the transition of the position signal θ during the period in which the rotating body rotates twice, the true value θ for θ is determined.
It turns out that it can be found by calculation.

In the present invention, the true value θ for θ is determined based on the amount of information while the rotating body is free running, and a correction table Δθ is created from this. This makes it possible to correct not only resolver errors but also position detection errors due to nonlinearity of the R/D converter, resolver mounting eccentricity, and the like.

Next, the position signal θ obtained by the conventional position detection device
In contrast, the principle of obtaining a position signal θ closer to the true value will be explained in detail.

FIG. 2 shows an example of speed changes during free running of the rotating body, where the horizontal axis is time t, the vertical axis is rotational speed ω, and it is assumed that dω/dt is constant during this period. Therefore, ω is given by the following equation.

(1) With the position at 1-0 as the origin, the rotational position θ is obtained by integrating the above equation as shown in the following equation.

Now, the time t=τ1 is the time when the rotating body has made one revolution and returned to the origin, and the time -τ2 is the time when the rotating body has made one more revolution (two revolutions) and returned to the origin. Since the position detected at this time is the origin defined at the beginning, it can be considered to be straight. Therefore, in equation (2), when t-τ1, θ-2π
When , is rearranged by substituting the relationship θ-4π when -τ2, the following equation is obtained.

Here, since time t is proportional to the reference clock count [K from 1-0 (θ-0), the following equation is obtained from equation (6).

・・・・・・(4) Here, K+ * K2 are respectively Okuθ≦2π, 2π
The position aθ is determined by counting the reference clocks from t-0 (θ-0) so that the position aθ is reduced. A characteristic feature of equation (4) is that θ is given as a ratio of the clock counting frequency. In other words, the calculation 1i of θ is based on the clock frequency if the clock frequency is constant during the clock counting period. Absolute accuracy has a sharp (point) (relationship. The counting period of the clock is, for example, about 100 m5ec,
Position θ can be calculated with high accuracy without being affected by temperature drift of clock frequency.

Next, an example will be described in which the rotational position θ is determined and a table of Δθ is created from this. For example, let's assume that the controlled object is an electric motor and that it is subject to variable speed control. Since the control device requires complex calculations, it is convenient to use a digital device using a microcomputer. at first,
All Δθ tables are reset to 10" and the rotational speed is increased to a predetermined speed. After that, the motor current is reduced to zero to create a freeze state. At this time, the control device is released from normal control and performs the following: The test mode operates as described in .The test mode is roughly divided into three modes.The first mode will be explained first.

In @1 mode, (4
) This is a mode in which data for finding 1 in K and 1 required in the calculation of the equation is stored. FIG. 3 is an explanatory diagram for explaining this.

In this N, two counters A and B are used to calculate the number of pulses of the reference clock (how long it takes for one bit to change) during a period in which the contents of the position signal θ change by one bit. Counter A performs counting operation when the lowest pit (LSB) of θ is “0”.
Figure (C) Reference input counter B performs a counting operation when the LSB of θ is 1'' (see Figure 3 (D)). The table in Figure (E) is used to store this count value. , this can be shared with the Δθ table mentioned earlier.

The microcomputer displays the count value of counter A when the LSB of 7 is 1s, and the count value of counter B when the LSB of 7 is '0''.
Input the count value of , and store each data in the address indicated by θ of the Δθ double. Further, after inputting the count value of the counter, each counter is reset. Δ
This mode ends when all data is input to the θ table and the detected position value returns to the origin.

The second mode follows the first mode and is a mode in which 2 is required to be calculated by the calculation of equation (4) during the period in which the rotating body makes one more rotation. The microcomputer calculates 2 by adding the counts of counter A and counter B one after another until θ starts from ul'' and returns to the origin ``0'' after one rotation.

The third mode is a mode for creating the final target Δθ table. By calculating equation (4), K1 can be found by adding all the contents of the Δθ table.

K2 is determined in the second mode. K is found by adding the contents from address 1 of the Δθ table to the address pointed to by θ. From the above, θ closer to the true value can be found for all ′θ゛. Once θ is determined, Δθ can be determined using the following equation.

Δθ=θ−θ 4 (5) A flowchart for creating a Δθ table in this mode is shown in FIG. The third mode starts when the content of θ becomes "1#" (see step ■), and all data is input into the Δθ table, and when the content of θ becomes "1" again, the third mode starts. (Refer to step ①), all modes of the test mode are ended.

From the above information when the rotating body to be measured is free running,
Although the method of creating a correction table for the position Δθ has been described, it is also possible to rotate the rotating body to be measured at a constant rotational speed using a drive device, and then calculate θ using the following equation (6).

θ − 2 Needless to say, it is also possible to find

FIG. 5 is an example of a hardware configuration showing an embodiment of the present invention. 1
-4 are similar to the conventional position detection device shown in FIG. 8, and counters 51 and 52 correspond to counters A and B in FIG. 3, respectively. 6 is a Δθ table, which is stored in a write-free non-volatile memory, such as EEPRO.
M or Batch IJ - consists of backed up RAM. The address of memory 6 is accessed by the output of R/D converter 4. An adder 7 adds the output data θ of the R/D converter and the position correction value Δθ corresponding to the data and stored in the memory B to obtain a highly accurate position signal θ. Reference numeral 8 is a data processing device such as a microphone stomach computer, and the data processing device 8 is a data processing device such as a microphone computer, and the data is processed through a buffer (not shown) to 4, 51, 52, 6.
．． 7 and a common bus 9, and performs data input/output and various calculations.

High precision is achieved for the output θ of the R/D converter 4, and θ is used as a position signal for normal control purposes, such as speed detection, localization stop, rotor position detection in vector control, etc. . Further, the memory 6, which is a Δθ table, may store position correction values only for representative data selected from the output data of the R/D converter 4; It becomes possible to reduce the storage capacity of 6. For example, the memory 6 is connected to the R/D converter 4
It has an address corresponding to a plurality of upper pits of the output data of the R/D converter 4, and one correction data is commonly stored in the address for the plurality of output data of the R/D converter 4.

FIG. 3 is an example of a hardware configuration showing another embodiment of the present invention. In the figure, addresses in memory 6 are accessed by a processing device (microcomputer) 8. The memory 6 stores position correction values for only representative data selected from the output data of the R/D converter 4. For the output data θ of the R/D converter 4, the processing device 8
A plurality of representative data closest to the data are selected, and a position correction value Δθ for the representative data is input from the memory 6. The processing device further calculates a position correction value for θ using a known interpolation method from θ, a plurality of representative data for θ, and a position correction value Δθ for the representative data, and obtains data obtained by adding θ to this correction 1. Let be the position signal.

Although the original position detection device has been described above as a combination of an R/D converter and a resolver, the present invention is not limited to the above position detection device. The present invention can be similarly applied to machines, encoders, etc.

〔Effect of the invention〕

According to the present invention, the correction table for correcting the position corresponding to the output of the conventional position detection device is rJGml, and the position signal is a value obtained by adding the correction value of the correction table to the output of the conventional original position detection device. Therefore, highly accurate position detection can be achieved with an inexpensive method. Especially in those that use data processing devices such as microcomputers,
By having the processing device itself create the correction table in its initial state, it is possible to obtain data to be input into the correction table easily and at low cost.

[Brief explanation of the drawing]

Fig. 1 is a principle explanatory diagram for explaining the present invention in detail, Fig. 2 is a characteristic diagram showing an example of speed change characteristics, and Fig. 3 is an explanatory diagram for explaining the first step of the test mode according to the present invention. Similarly, FIG. 4 is a flowchart for explaining the third step of the test mode, FIG. 5 is a block diagram showing 11 examples of the present invention, and FIG. 3 is a diagram showing another embodiment of the present invention. A configuration diagram, FIG. 7 is an explanatory diagram for explaining the basic concept of the present invention, and FIG. 8 is a configuration diagram showing a conventional example of a position detection device.
FIG. 9 is a characteristic diagram showing conventional position detection characteristics. Code explanation 1...Rotating body, 2...Resolver, 6.
・1 oscillator, 4...R/D converter, 51, 5
2...Counter, 6...Memory, 7.
・Time adder, 8... Data processing device (microcomputer), 9... Song common bus. Agent Patent Attorney Akio Namiki Agent Patent Attorney Kiyoshi Matsuzaki @IQ
Eco W 2 Figure 0 te l:2 -χ Yuki 3 ■ Figure 4 lf'r6! f4 −

Claims (1)

[Claims] 1) A position detector that detects the rotation angle of a rotating body as a digital value, and a position correction device having a table length corresponding to the output data length of the position detector and corresponding to each of the output data values. What is claimed is: 1. A digital position detection device comprising: a memory storing a value; 2) In the digital position detecting device according to claim 1, the data length of the memory is made shorter than the output data length of the position detector, and the memory stores representative information from the output data of the position detector. A digital position detection device characterized in that a position correction value corresponding to only selected representative data is stored. 3) In the digital position detecting device according to claim 1, the data length of the memory is shorter than the output data length of the position detector, and the memory contains representative data from the output data of the position detector. A digital position system characterized in that a position correction value corresponding to only selected representative data is stored, and data other than the representative data is determined by interpolation from the position correction values stored in the memory. Detection device. 4) In the digital position detection device according to any one of claims 1 to 3, when the output data of the position detector is obtained when the object to be measured is rotated at a predetermined rotational speed, Also, if the arbitrary pit of the data is “
The digital device is characterized in that the position of the rotating body is calculated by counting the number of reference clocks in a period of “0” or “1”, and the deviation between the calculated value and the output of the position detector is used as data in the memory. type position detection device.