JP2007178320A - Rotary encoder - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To allow eccentricity to the rotating shaft of a scale board to some extent by correcting angle errors due to the eccentricity to the rotating shaft of the scale board, in a rotary encoder. <P>SOLUTION: The rotary encoder comprises the scale board giving an angle scale, and a pair of CCD linear sensors obtaining a pair of reading angles by reading the angle scale of a symmetric position of the scale board. In each proper rotation angle of the scale board of the rotary encoder, the average Ψ and the difference Φ of the pair of the reading angles are calculated, and the average Ψ and the difference Φ satisfy the relational: Φ=Asin(Ψ+B)+C, and eccentric factors A, B, C are calculated and are stored to a storing means, in advance. In measuring the actual angles, the measured angle values are corrected by using eccentric factors including the eccentric factors A, B, C. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a rotary encoder used in a surveying instrument, and more particularly, to a rotary encoder that can correct an angular error when a scale plate is eccentric with respect to a rotating shaft.

  The surveying instrument uses a rotary encoder to measure a horizontal angle and a vertical angle. The rotary encoder is a rotating disk that rotates together with the telescope, and measures a horizontal angle and a vertical angle by reading a scale attached along the circumferential direction of the rotating disk.

  In a rotary encoder used in a conventional surveying instrument, opposing detection is performed by a pair of detectors in order to eliminate the eccentric error of the dial. However, if the position where the opposite is detected is not equidistant and accurately positioned at 180 ° with respect to the rotation center of the scale plate, an angular error occurs, so the eccentricity with respect to the rotary shaft of the scale plate is less than a predetermined value. It adjusted and assembled so that it might become (refer the following patent document 1).

JP 2001-289672 A

  However, detecting and adjusting the eccentricity of the rotary encoder scale plate is burdensome for the operator, takes time, and hinders cost reduction of the surveying instrument.

  The present invention has been made in view of the above problems, and even if the rotary encoder scale plate is mounted eccentrically with respect to the rotary shaft, the rotary encoder scale plate is corrected by correcting the angular error due to the eccentricity. It is an object to allow a certain degree of eccentricity with respect to the rotating shaft.

  In order to solve the above-mentioned problem, in the invention according to claim 1, a rotary equipped with a scale plate with an angle scale and a detector that reads the angle scale at the symmetrical position of the scale board and obtains a pair of read angles. In the encoder, each time the scale plate is rotated by an appropriate angle, the pair of reading angles is obtained, the average Ψ and the difference Φ of the pair of reading angles are calculated and stored, and the plurality of sets of stored averages An eccentric factor calculating means for calculating an eccentric factor representing the eccentricity with respect to the rotation axis of the dial plate using Ψ and the difference Φ, and an eccentric factor storing means for storing an eccentric factor including the eccentric factor calculated by the eccentric factor calculating means And angle measurement value correcting means for correcting the angle measurement value using the eccentric factor stored in the eccentric factor storage means.

  In the invention according to claim 2, in the invention according to claim 1, the eccentricity factor calculation means determines that the stored plurality of sets of average Ψ and difference Φ satisfy the relationship of Φ = Asin (Ψ + B) + C. The values of A, B, and C, which are the eccentric factors, are obtained using a square method.

  According to the first aspect of the invention, before shipping the surveying instrument, every time the rotary encoder dial is appropriately rotated, the average Ψ and the difference Φ of the pair of reading angles are calculated, and these average Ψ and An eccentricity factor representing the eccentricity with respect to the rotation axis of the dial is calculated and stored from the difference Φ, and when measuring with a surveying instrument, the angle measurement value is corrected using the eccentricity factor, so the dial rotates. Even if it is slightly decentered with respect to the axis, no angular error will occur. The calculation of the eccentric factor at this time is extremely easy because there is no troublesome measurement work. This eliminates the need to adjust the eccentricity of the rotary encoder scale plate with high accuracy, reduces the burden on the operator, shortens the work time, and reduces the cost of the surveying instrument.

  According to the second aspect of the invention, the stored average Ψ and the difference Φ satisfy the relationship of Φ = Asin (Ψ + B) + C, and the eccentric factors A, B, and C are calculated using the least square method. Therefore, the eccentricity factor can be obtained easily and accurately, and an accurate angle measurement value can be obtained more easily.

  The present invention calculates the average Ψ and difference Φ of a pair of reading angles each time the rotary encoder dial is rotated by an appropriate angle before shipping the surveying instrument, and calculates the scale disk from these average Ψ and difference Φ. An eccentric factor representing the eccentricity with respect to the rotation axis is calculated and stored, and when the user measures with the surveying instrument, the angle measurement value is corrected using the eccentric factor including the eccentric factor.

  FIG. 1 is a block diagram of an embodiment according to the rotary encoder of the present invention. FIG. 2 is a diagram for explaining the generation of an angle error when the dial is eccentric in the rotary encoder. FIG. 3 is a diagram showing a change in angular error due to the eccentricity of the dial. FIG. 4 is a diagram for explaining the generation of an angle error when the CCD linear sensor is not parallel in addition to the eccentricity of the scale plate in the rotary encoder. FIG. 5 is a flowchart showing a procedure for storing the eccentricity factor necessary for correcting the angle error.

  As shown in FIG. 1, the rotary encoder includes a scale plate 10, a code pattern 12 provided near the periphery of the scale plate 10, and a pair of CCD linear sensors 14 and 14 ′ that read the code pattern 12 at a symmetrical position. (Detector), amplifiers 15 and 15 'for amplifying output signals from the CCD linear sensors 14 and 14', A / D converters 16 and 16 'for converting the amplified signals into digital signals, and A CPU18 which calculates | requires a measured angle value from the output signal of / D converter, the memory | storage means 20 which memorize | stores data and a program, and the display means 22 which displays the calculated | required measured angle value are provided. The code pattern 12 is an angle scale by a code combining a thin slit and a thick slit. The code pattern 12 is read by the pair of CCD linear sensors 14 and 14 ′, and the average of the obtained reading angles is taken to reduce errors due to the eccentricity of the dial 10. It is also possible to process the output signals of the pair of amplifiers 15 and 15 'by inputting them to one A / D converter using a changeover switch.

  In the present embodiment, an eccentric factor (to be described later) relating to the eccentricity of the scale plate 10 is stored in advance in the storage means 20, and when the user measures the angle, the measured angle value is further corrected with the eccentric factor, and the scale plate 10 The angle error due to the eccentricity with respect to the rotating shaft 11 is removed.

First, how to determine the eccentric factor will be described. Based on FIG. 2, angle errors δ1 and δ2 that occur when the dial 10 is fixed eccentrically with respect to the rotating shaft 11 will be described. The CCD linear sensors 14 and 14 'for reading the scale plate 10 are completely parallel. The arrows on the CCD linear sensors 14, 14 'indicate the direction in which the pixel number increases. The distance from the center Ca of the rotary shaft 11 of the graduation plate 10 to the CCD linear sensor 14, 14 ', respectively, and _{R 1} and _{R 2.} When the rotary shaft 11 is rotated, the locus of the center Cd of the dial plate 10 is set as T. An arrow S along the locus T indicates a direction in which the angle increases from the direction O of the scale 0 of the scale board 10. The distance between the center Ca of the rotary shaft 11 and the center Cd of the scale plate 10 is set to e, and it is rotated around the rotary shaft 11 by the rotation angle θ, and the reading points P ₁ and P _{2 of the} scale plate 10 are the rotary shafts. It is assumed that it is on a straight line passing through the 11 center Ca. Then, the reading points P ₁ , P ₂ on the CCD linear sensors 14, 14 ′ are at an angle δ ₁ , with respect to the reading points Y ₁ , Y ₂ when the scale plate 10 is not eccentric with respect to the rotating shaft 11. Since they are shifted by δ ₂ , the angles δ ₁ and δ ₂ are angle errors.

When the center Cd of the graduation plate 10 is on the line connecting the centers Ca of read point P ₁ and the rotary shaft 11 of the dial 10, the scale angle of the graduation plate 10 and theta _0. Now, when the scale plate 10 is rotated about the rotation axis 11 by the rotation angle θ, there are angular errors δ ₁ and δ _{2 due} to the eccentricity of the scale plate 10 as described above, so at the reading point P ₁ of the scale plate 10. The reading angle φ ₁ of the scale plate 10 read by the CCD linear sensor 14 is an angle smaller than θ ₀ −θ by an angle error δ ₁ and is as follows.
φ ₁ = θ ₀ −θ−δ ₁
= Θ ₀ −θ−tan ⁻¹ {(e * sin θ) / (R ₁ −e * cos θ)}
≈θ ₀ −θ− (e * sin θ) / R ₁
(1)
The reading angle φ ₂ of the scale plate 10 read by the CCD linear sensor 14 ′ at the reading point P ₂ of the scale plate 10 is larger than θ ₀ -θ by π (180 °) and an angle error δ _2. It becomes as follows.
φ ₂ = θ ₀ −θ + π + δ ₂
= Θ ₀ −θ + π + tan ⁻¹ {(e * sin θ) / (R ₂ + e * cos θ)}
≈θ ₀ −θ + π + (e * sin θ) / R ₂
(2)
As can be seen from equations (1) and (2),
The angle error δ ₁ = − (e * sin θ) / R ₁ and δ ₂ = (e * sin θ) / R ₂ are sine with respect to the rotation angle θ as shown in FIGS. 3A and 3B. Become a wave.

Next, as shown in FIG. 4, CCD linear sensor 14, 14 'is not perfectly parallel, when intersect at an angle η, the read points _{P 1} and _{P 2} are CCD linear sensor 14, 14' are mutually since rotates by each angle eta / 2 from the reading point _{P 1} and _{P 2} when parallel, rotation angle with respect to the reading point _{P 1} and _{P 2} of the graduation plate 10 are respectively θ-η / 2, θ + η / 2 , and the The equations (1) and (2) are corrected as follows.
φ ₁ = θ ₀ − (θ−η / 2) − {e * sin (θ−η / 2)} / R ₁
= Θ ₀ −θ + η / 2− {e * sin (θ−η / 2)} / R ₁
(3)
φ ₂ = θ ₀ − (θ + η / 2) + π + {e * sin (θ + η / 2)} / R ₂
= Θ ₀ −θ−η / 2 + π + {e * sin (θ + η / 2)} / R ₂
(4)

Here, the difference Φ between the reading angles φ ₁ and φ ₂ of the scale plate 10 read at the reading points P ₁ and P ₂ of the scale plate 10 is as follows. However, φ ₂ is calculated from the reduced by π.
Φ = φ ₁ − (φ ₂ −π)
= Η- {e / (R ₁ R ₂ )} √ (R ₁ ² + R ₂ ² + 2R ₁ R ₂ cos η)
* Sin [θ + tan ⁻¹ {(R ₁ −R ₂ ) / (R ₁ + R ₂ ) * tan (η / 2)}]
(5)
Here, when an angle θr indicating a correct rotation angle not including an error of the rotary encoder is set to θr = θ ₀ −θ, the expression (5) is finally rewritten as follows.
Φ = η + {e / (R ₁ * R ₂ )} √ (R ₁ ² + R ₂ ² + 2R ₁ R ₂ cos η)
* Sin [θr−θ ₀ + tan ⁻¹ {(R ₂ −R ₁ ) / (R ₁ + R ₂ )
* Tan (η / 2)}] (6)

On the other hand, when the average Ψ of the reading angles φ ₁ and φ ₂ of the scale plate 10 that has been read is obtained, the following is finally obtained.
Ψ = {φ ₁ + (φ ₂ −π)} / 2
= Θr + (1/2) {− (e / R ₁ ) * sin (θ−η / 2)
+ (E / R ₂ ) * sin (θ + η / 2)}
(7)
Here, since the second term of the equation (7) can be ignored as compared with θr, Ψ≈θr can be established. Then, the equation (6) can also be expressed as follows.
Φ = Asin (Ψ + B) + C (8)
However, A, B, and C are eccentric factors used for correcting the angle measurement value, and have the following values. Here, R ₁ and R ₂ are also eccentric factors used for correcting the angle measurement value.
A = (e / R ₁ / R ₂ ) √ (R ₁ ² + R ₂ ² + 2R ₁ R ₂ cos η) (9)
B = −θ ₀
+ Tan ⁻¹ {(R ₂ −R ₁ ) / (R ₁ + R ₂ ) * tan (η / 2)} (10)
C = η (11)

Therefore, the scale plate 10 is actually rotated by an appropriate angle, and the scale plate 10 is read each time. A difference between a pair of reading angles Φ = φ ₁ − (φ ₂ −π) and an average of a pair of reading angles Ψ = By calculating {φ ₁ + (φ ₂ −π)} / 2 and applying the least square method to the form of equation (8), the values of A, B, and C in equation (8) can be obtained. it can. Since the least square method is well known to those skilled in the art, the description thereof is omitted. If the values of A, B, and C are obtained, the distance e between the center Ca of the rotating shaft 11 and the center Cd of the scale plate 10, the scale angle θ ₀ of the scale plate 10 described above, and the pair of CCD linear sensors 14, 14 The angle η formed by 'can be obtained. When only the result is described, it becomes as follows.
e = AR ₁ R ₂ / √ (R ₁ ² + R ₂ ² + 2R ₁ R ₂ cosC) (12)
θ ₀ = −B + tan ⁻¹ {(R ₂ −R ₁ ) / (R ₁ + R ₂ ) * tan (C / 2)}
(13)
η = C (14)

Thus, when the eccentric factors R ₁ , R ₂ , A, B and C or R ₁ , R ₂ , e, θ ₀ and η are obtained, the error of the measured angle value read by this rotary encoder can be corrected from this. it can.

  Next, a method of correcting the angle measurement value using the eccentric factor will be described.

Equations (3) and (4) can be written as follows when θr = θ ₀ −θ.
φ ₁ = θr + (η / 2) + (e / R ₁ ) sin (θr−θ ₀ + η / 2) (15)
φ ₂ = θr + π− (η / 2) − (e / R ₂ ) sin (θr−θ ₀ −η / 2) (16)
Using the equations (15) and (16), the average Ψ of the reading angles φ ₁ and φ ₂ of the scale plate 10 read is finally obtained as follows.
Ψ = {φ ₁ + (φ ₂ −π)} / 2
_{= Θr- {e / (2R 1} R 2)} √ (R 1 2 + R 2 2 -2R 1 R 2 cosη)
* Sin [θr + B
−tan ⁻¹ {(R ₂ −R ₁ ) / (R ₁ + R ₂ ) * tan (η / 2)}
+ Tan ⁻¹ {(−R ₁ −R ₂ ) / (R ₁ −R ₂ ) * tan (η / 2)}] (17)
Or Ψ = {φ ₁ + (φ ₂ −π)} / 2
= Θr- (A / 2) { √ (R 1 2 + R 2 2 -2R 1 R 2 cosC)
/ √ (R ₁ ² + R ₂ ² + 2R ₁ R ₂ cosC)}
* Sin [θr + B
−tan ⁻¹ {(R ₂ −R ₁ ) / (R ₁ + R ₂ ) * tan (C / 2)}
+ Tan ⁻¹ {(−R ₁ −R ₂ ) / (R ₁ −R ₂ )} * tan (C / 2)}] (18)

Since θr in [] of sin in equation (18) can be set to ψ, equation (18) can also be written as follows.
Ψ = θr- (A / 2) {√ (R 1 2 + R 2 2 -2R 1 R 2 cosC)
/ √ (R ₁ ² + R ₂ ² + 2R ₁ R ₂ cosC)}
* Sin [Ψ + B
−tan ⁻¹ {(R ₂ −R ₁ ) / (R ₁ + R ₂ ) * tan (C / 2)}
+ Tan ⁻¹ {(−R ₁ −R ₂ ) / (R ₁ −R ₂ ) * tan (C / 2)}] (19)

If the values of R ₁ , R ₂ , A, B and C are obtained in advance, the reading angles φ ₁ and φ _{2 of} the scale plate 10 read from the pair of CCD linear sensors 14 and 14 ′ from this equation (19). It can be seen that the angle error due to the eccentricity of the dial 10 with respect to the rotating shaft 11 can be reduced by adding the following error correction amount Ccomp to the average Ψ of
Ccomp = (A / 2) { √ (R 1 2 + R 2 2 -2R 1 R 2 cosC)
/ √ (R ₁ ² + R ₂ ² + 2R ₁ R ₂ cosC)}
* Sin [Ψ + B
−tan ⁻¹ {(R ₂ −R ₁ ) / (R ₁ + R ₂ ) * tan (C / 2)}
+ Tan ⁻¹ {(−R ₁ −R ₂ ) / (R ₁ −R ₂ ) * tan (C / 2)}] (20)

  The CPU 18 corrects the angle measurement value using the equations (19) and (20), thereby calculating an accurate angle measurement value. The CPU 18 performs an operation for correcting the angle measurement value using the equations (19) and (20).

Now, when the surveying instrument is completed at the factory, the scale plate 10 is rotated by rotating the telescope at an appropriate angle. The CCD linear sensor 14 reads the scale plate 10 each time, and the reading angles φ ₁ and φ ₂ of the scale plate 10 are read. Get. At this time, the values of R ₁ and R _{2 in} the equation (20) are also measured from the image of the scale plate 10 projected onto the CCD linear sensors 14, 14 ′ and stored in the storage means 20. The CPU 18 obtains the values of A, B, and C in the equation (8) from the average Ψ and the difference Φ of the multiple sets thus obtained by using the least square method, and the values of A, B, and C are also stored in the storage means. 20 stored. The storage means 20 may store the amplitude and the initial phase representing the error correction amount Ccomp in the equation (20). A procedure for obtaining the values of A, B, and C will be described with reference to FIG.

First, after the surveying instrument is completed, the eccentricity factor calculation program is started. Then, it progresses to step S1, and the scale board 10 is read to CPU18 with a pair of CCD linear sensors 14 and 14 ', and reading angle (phi) ₁ and (phi) ₂ are obtained. Then, the process proceeds to step S2, CPU 18 stores seeking read angle phi ₁ and phi ₂ of the average Ψ and difference Φ read by the pair of CCD linear sensor 14, 14 '. Next, proceeding to step S3, the scale plate 10 is rotated at an appropriate angle by rotating the telescope of the surveying instrument at an appropriate angle (for example, about 30 °).

Next, it progresses to step S4 and CPU18 judges whether this program may be complete | finished. This is determined by whether the reading angles φ ₁ and φ ₂ have been acquired a predetermined number of times or more. If steps S1-S3 are completed at different angles at least three times or more, the process can be completed. Normally, however, steps S1-S4 are repeated, and it is determined that the process is completed when the telescope is rotated 360 degrees or more. If it is not determined to end, the process returns to step S1.

  If it is determined in step S4 that the process is terminated, the process proceeds to step S5, and the CPU 18 obtains the values of A, B, and C in the equation (8) by using the least square method. Steps S <b> 1 to S <b> 5 correspond to the eccentricity factor calculating means described in claim 1. In step S6, the values of A, B, and C are stored in the storage unit 20. This step S6 corresponds to the eccentric factor storage means described in claim 1. This ends the program.

According to the present embodiment, since the eccentric factors R ₁ , R ₂ , A, B, and C are known, the reading angle obtained from the pair of CCD linear sensors 14 and 14 ′ when the user measures the angle. By adding the error correction amount Ccomp of the equation (20) to the average Ψ of φ ₁ and φ ₂ to correct the angular error due to the eccentricity with respect to the rotary shaft 11 of the dial plate 10, a more accurate angle measurement value is obtained and displayed. It can be displayed on the means 22. For this reason, a slight eccentricity with respect to the rotation axis of the dial plate 10 may be allowed, the mounting adjustment of the CCD linear sensors 14 and 14 'becomes easy, and the eccentric factors R ₁ , R ₂ , A, B And the labor of the operator for obtaining C and C can be reduced, so that the work for attaching the scale plate to the rotating shaft is facilitated, the time is shortened, and the cost can be reduced.

  By the way, the present invention is not limited to the above-described embodiments, and various modifications are possible. For example, as in the previous embodiment, the eccentricity factor was obtained by the least square method assuming that there is a relationship of Eq. (8) between Φ and ψ, but the following equation (21) is obtained between Φ and ψ. E0, E1, and F1 may be obtained assuming a relationship in which the second harmonic wave (n = 2) or more is ignored in the Fourier series.

Φ = E0 + E1 cos Ψ + F1 sin Ψ (21)
Since methods for calculating the Fourier coefficients E0, E1, and F1 are also well known to those skilled in the art, description of how to obtain the Fourier coefficients is omitted. Equation (21) can be easily transformed into the same form as Equation (8) to determine the values of eccentric factors A, B, and C, and thereafter the angle measurement value is corrected in the same manner as in the previous embodiment. Can do.

  In addition, with respect to the above-described embodiment, the present invention can be realized even if a design change is made to move step S3 in FIG. 5 to a position on the way from step S4 to step S1.

  The rotary encoder having the error correction function is not only used for angle measurement by a surveying instrument, but can be widely used for devices that measure angles other than surveying instruments.

The block diagram of the rotary encoder of this invention. The figure explaining the generation | occurrence | production of an angle error when the scale board is eccentric with respect to a rotating shaft in a rotary encoder. The figure which shows the change of the said angle error. In a rotary encoder, the figure which explains generation | occurrence | production of an angle error in case a scale plate is eccentric with respect to a rotating shaft, and a CCD linear sensor is not mutually parallel. The flowchart which shows the procedure which memorize | stores the eccentric factor required for correction | amendment of a measured angle value in the rotary encoder of this invention.

Explanation of symbols

10 Scale board 11 Rotating shaft 12 Scale 14 CCD linear sensor (detector)

Claims (2)

In a rotary encoder comprising a scale plate with an angle scale and a detector that reads the angle scale at a symmetrical position of the scale board and obtains a pair of reading angles,
Each time the scale plate is rotated by an appropriate angle, the pair of reading angles is obtained, the average Ψ and the difference Φ of the pair of reading angles are calculated and stored, and the stored plural sets of the average Ψ and the difference are stored. Using Φ, an eccentric factor calculating means for calculating an eccentric factor representing the eccentricity with respect to the rotation axis of the dial plate,
An eccentric factor storage means for storing an eccentric factor including the eccentric factor calculated by the eccentric factor calculation means;
A rotary encoder comprising: an angle measurement value correcting unit that corrects an angle measurement value using the eccentric factor stored in the eccentric factor storage unit.   The eccentricity factor calculating means uses the least square method and assumes that the eccentricity factors A, B, and C, assuming that the stored average Ψ and difference Φ satisfy a relationship of Φ = Asin (Ψ + B) + C. The rotary encoder according to claim 1, wherein a value of is obtained.

Images