JPH0692890B2 - Absolute position detector - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an absolute position detection device that detects an absolute position of a current position of a linear table, a rotary table, or the like by applying an absolute position detector, and more particularly to an absolute position detection device. , An apparatus that prepares another detector on a table or the like and calculates the absolute amount of the detector with high accuracy by a hybrid method.

[Background technology and its problems]

The previously proposed absolute position detection method is
For example, considering a machine tool, a rotation amount is extracted from a drive system in the X-axis direction with respect to a certain coordinate axis direction (X-axis direction), and the rotation amount is calculated in two or three stages. The absolute value is obtained by combining the values of the rotation detectors attached to the respective deceleration rotation shafts within one revolution.

However, in this conventional example, when an attempt is made to expand the effective measurement range, the deceleration mechanism becomes large and the moment of inertia thereof also increases. The values on the rotary shafts have different weights, that is, the rotary shafts are different from each other. Since the amount of movement of the member to be moved (eg, table) per rotation is different, there is a problem that an error is likely to occur. For this reason, the mechanical accuracy of each element of the reduction mechanism, mainly gears, bearing members, etc., must always be maintained at high accuracy despite vibrations and wear associated with the operation of the machine tool.

Therefore, the present applicant has previously proposed Japanese Patent Application No. 57-199882 as a solution to such a problem. This is a method of detecting the amount of mechanical change from a predetermined reference state position with respect to the mechanical movement of the member to be measured as an absolute value. A detection means including a plurality of detectors that emit signals is prepared, an electric signal corresponding to each cycle is taken out from the detection means, stored and held, and then one of the detection means in accordance with the mechanical change amount. The relative change amount between the first (first detector) and the member to be measured,
In obtaining the integer value from the integer multiple of the period corresponding to the first detector and the value less than one period of the same period, the integer value corresponds to the other detectors in the detecting means and the others. The absolute position is detected by using the procedure of making a decision using the value obtained from the detector and stored and held.

As a result, this method is a method to calculate the absolute position by combining the data from the detectors having multiple cycles, so there is no weight between the measurement data from each detector as in the conventional method, and there is no measurement error. In addition to the fact that there is almost no restriction, it is possible to reduce the moment of inertia even in the gear train, and it is almost unnecessary to take measures against the occurrence of an error due to friction of the gear.

However, in the case of this method, since it is necessary to prepare a plurality of detectors, at least three or more, which emit electric signals having different periods corresponding to different predetermined mechanical changes to the member to be measured, it becomes expensive. There are drawbacks. Further, since these detectors are connected to the drive shaft and data is taken out from the drive shaft, there is a drawback that errors due to twisting of the drive shaft and play between the drive shaft and the member to be measured are included.

[Object of the Invention]

Here, an object of the present invention is to apply the above-mentioned absolute position detection method to relatively measure the current position of a member to be measured such as a linear table or a rotary table even if the drive shaft is twisted or played. An object of the present invention is to provide an absolute position detection device that has a simple configuration and that detects an absolute amount with high accuracy.

[Means and Actions for Solving Problems]

Therefore, the absolute position detection device of the present invention is
Regarding a mechanical movement of a member to be measured which is moved by rotation of a drive shaft, a device for detecting an amount of movement from a predetermined reference state position as an absolute value, the device being provided along a moving direction of the member to be measured. A first detection means for generating a periodic electric signal corresponding to the amount of movement of the member to be measured; and an absolute value which is connected to the drive shaft and matches the amount of movement of the member to be measured can be detected, and Second detection means in which an error based on twist or play of the drive shaft is within a range of ± λ / 4 of the period λ of the first detection means;
Arithmetic means for determining the absolute position of the member to be measured based on output signals from the first and second detecting means, the arithmetic means being included in the output signal of the second detecting means. A first means for obtaining the absolute value of the member to be measured by adding the amount of movement based on the output signal of the first detection means to the product of the integer value of the cycle of the first detection means and the cycle of the first detection means; When the difference obtained by subtracting the product from the output signal of the second detection means is a value of λ / 4 or less and the output signal of the first detection means is a value of 3λ / 4 or more, and the difference is 3λ / 4. When the above value and the output signal of the first detection means are a value of λ / 4 or less, the first value with respect to the absolute value
And a second means for correcting the detection means for one period λ.

〔Example〕

 FIG. 1 shows an embodiment of the present invention.

In the figure, a ball screw 63 as a drive shaft is connected to one end side of an output shaft 62 of the motor 50, and a table 51 is screwed onto the ball screw 63. This allows the table
The 51 is moved back and forth in the X-axis direction by the rotation of the motor 50. On the other end of the output shaft 62 of the motor 50, a coupling 64
An absolute position detector 53 as a second detection means capable of detecting an absolute value that matches the amount of movement of the table 51 is connected via. Also, the table
A linear inductosyn 71 is provided along the moving direction of the table 51 as a first detecting means for generating a periodic electric signal corresponding to the moving amount of the table 51. The straight-type inductosyn 71 includes a slider 58 and a stator 59, and has a structure in which the slider 58 slides with respect to the stator 59 as the table 51 advances and retracts.

Here, in the primary side excitation winding of the absolute position detector 53 and the stator 59 of the linear induct thin 71,
Electric wires 54 and 60 for supplying an exciting current are connected from the driver / converter circuit 52. Also, each (53,71)
The secondary side output is fed back to the driver / converter circuit 52 by the electric wires 59 and 61. The signals PA and Pi fed back to the driver / converter circuit 52 are converted into position data and then sent to the CPU 57 as a calculation means via the line 56. The CPU 57 highly accurately calculates the absolute position by the hybrid method according to the procedure described later based on the obtained position data.

Calculation of Absolute Position (1) In FIG. 1, it is assumed that the absolute position detector 53 and the linear inductosyn 71 have been adjusted in advance so that the zero points coincide with each other.

From this state, when the motor 50 is rotated in order to move the table 51 in the X-axis direction, the output signal PA from the absolute position detector 53 and the output signal Pi from the linear induct sync 71 are changed to the driver / converter. The circuit 52 converts the data into position data PA (A) and Pi (i), respectively (see FIG. 2). FIG. 2 shows position data A ₀ , i ₀ obtained when the motor 50 rotates arbitrarily.

At this time, if the ball screw 63 has a twist or backlash, the obtained position data A ₀ itself cannot represent an accurate absolute position. Therefore, the absolute position X ₀ ^* is accurately calculated with the accuracy of λ / 2000 [μm / pulse] using the value of the position data i ₀ according to the process of the flowchart shown in FIG. Note that λ is the amount of movement per cycle (1 wavelength) obtained from the linear inductosyn 71. Further, it is assumed that the error in the absolute position detector 53 is within the range of ± λ / 4.

In FIG. 3, in STPl, the absolute position is calculated using the position data A ₀ , i ₀ , and this is set as X ₀ . That is, X ₀ = iFix [A ₀ / λ] λ + i ₀ × λ / 2000 is calculated. Here, iFix [A ₀ / λ] means conversion of A ₀ / λ into an integer. Therefore, here, the integral value iFix [A ₀ / λ] of the period λ of the linear inductosyn 71 included in the position data A ₀ from the absolute position detector 53 is multiplied by λ, and the product iFix [A _The absolute position X ₀ is calculated by adding the movement amount i ₀ × λ / 2000 based on the position data from the inductosyn 71 to ₀ / λ] λ.

Next, in STP2, it is checked whether the difference obtained by subtracting the product iFix [A ₀ / λ] λ from the position data A ₀ from the absolute position detector 53 is λ / 4 or less. If the difference is λ / 4 or less, move to STP3 and position data i ₀ is 1500 or more (3λ
/ 4 or more) is checked. Position data i ₀
When it is 1500 or more (3λ / 4 or more), assuming that the error in the absolute position detector 53 is within ± λ / 4, it is within one cycle before the cycle of the position data i _0. It contains the position data A ₀ . That is, position data
With respect to the accurate absolute value based on i ₀ , STP1 calculates the absolute position X ₀ by using the integer value iFix [A ₀ / λ] that is one cycle longer by λ, which causes an error. It will be. Therefore, in this case, the process moves to STP4, one cycle λ is subtracted from the absolute position ₀ obtained in STP1, and the process moves to STP8. The position data i ₀ is 15
If it is not 00 or more (3λ / 4 or more), the position data i ₀
Since the position data A ₀ is included in the same cycle of, the process directly moves to STP8 without correction.

On the other hand, in STP3, when the difference is not λ / 4 or less,
Go to STP5 and check whether the difference is 3λ / 4 or more. When the difference is 3λ / 4 or more, the process proceeds to STP6 and it is checked whether or not the position data i ₀ is 500 or less (λ / 4 or less). When the position data i ₀ is 500 or less (λ / 4 or less), assuming that the error in the absolute position detector 53 is within the range of ± λ / 4, one from the cycle of the position data i _0. The position data A ₀ is included in the previous cycle. That is, in STP1, an integer value iFix that is one cycle λ less than the accurate absolute value based on the position data i ₀
Since the absolute position X ₀ is calculated by using [A ₀ / λ], this means that an error is generated. Therefore, in this case, the process moves to STP7, one cycle λ is added to the absolute position X ₀ obtained in STP1, and the process moves to STP8.
When the difference is not λ / 4 or less and is not 3λ / 4 or more, or the difference is 3λ / 4 or more and the position data is
When i ₀ is not 500 or less (λ / 4 or less), the position data A ₀ is included in the same cycle of the position data i ₀ , and therefore the process directly proceeds to STP8 without correction.

In STP8, X _{0 is set} to X ₀ ^* , and an accurate absolute position is calculated.

Calculation of absolute position (2) In the calculation of absolute position (1), the absolute position detector 53 and the linear inductometer 71 are set to 0.
The absolute position has been calculated assuming that the points match each other. However, in practice, it is difficult to adjust the 0 points of the absolute position detector 53 and the linear inductosyn 71 so as to coincide with each other, and it is more common to calculate the absolute position without adjustment.

Now, the position data from the linear type induct synth 71 obtained at the zero point of the absolute position detector 53 is i ₁ , and the state when the table 51 is moved in the X-axis direction from this state is shown in FIG. . In FIG. 4, when the position data PA (A) = 0 from the absolute position detector 53, the position data Pi (i) = i ₁ from the linear inductosyn 71 is set, and when the motor 50 rotates arbitrarily, PA (A) = A ₀ , Pi
It is assumed that (i) = i ₀ is obtained.

At this time, if the ball screw 63 has a twist or backlash, the obtained position data A ₀ itself cannot represent an accurate absolute position. Therefore, the absolute position ₀ ^* is accurately calculated with the accuracy of λ / 2000 [μm / pulse] using the value of the position data i ₀ according to the process of the flowchart shown in FIG. Note that σ is an offset amount obtained from the linear inductosyn 71. Further, it is assumed that the error in the absolute position detector 53 is within the range of ± λ / 4.

In FIG. 5, in STP1, the position data i ₁ is used to calculate the offset amount with respect to the absolute position, and this is set as σ. That is, σ = i ₁ × λ / 2000 is calculated. Then, in STP2, the absolute position is calculated using the position data A ₀ , i ₀ and the offset amount σ, and this is set as X ₀ . That is, X ₀ = iFix [(A ₀ + σ) / λ] λ + i ₀ × λ / 2000 is calculated.

Next, in STP3, the difference obtained by subtracting the product iFix [(A ₀ + σ) / λ] λ from the value obtained by adding the offset amount σ to the position data A ₀ from the absolute position detector 53 is λ / 4 or less. Check if there is. If the difference is λ / 4 or less, ST
Moving to P4, it is checked whether the position data i ₀ is 1500 or more (3λ / 4 or more). Position data i ₀ is 1500 or more (3
λ / 4) or more), assuming that the error in the absolute position detector 53 is within ± λ / 4, the position data A ₀ within one cycle before the cycle of the position data i ₀ It means that ₀ is entered. That is, in STP2, the absolute position X ₀ is calculated using the integer value iFix [(A ₀ + σ) / λ] that is one cycle λ higher than the accurate absolute value based on the position data i ₀ . This causes an error. Therefore, in this case, the process moves to STP5, and one cycle λ from the absolute position X ₀ obtained in STP2.
Subtract minutes and move to STP9. The position data i ₀ is 1500
If it is not above (3λ / 4 or more), the position data A ₀ is included in the same cycle of the position data i ₀ , and therefore the process moves directly to STP9 without correction.

On the other hand, in STP3, when the difference is not λ / 4 or less,
Move to STP6 and check whether the difference is 3λ / 4 or more. When the difference is 3λ / 4 or more, the process proceeds to STP7 and it is checked whether or not the position data i ₀ is 500 or less (λ / 4 or less). When the position data i ₀ is 500 or less (λ / 4 or less), assuming that the error in the absolute position detector 53 is within the range of ± λ /, one period before the cycle of the position data i ₀ Position data A ₀ is included in the cycle of. That is, in STP2, an integer value iFix that is one cycle λ less than the accurate absolute value based on the position data i ₀
Since the absolute position X ₀ is calculated by using [(A ₀ + σ) / λ], it means that an error is caused thereby. Therefore, in this case, move to STP8, add 1 cycle λ to the absolute position X ₀ obtained in STP2, and add STP9
Move to. The difference is not less than λ / 4, and 3λ / 4
When not more than, or the difference is 3 [lambda] / 4 or more and, when the position data i ₀ is not more than 500 (lambda / 4 or less), contains position data A ₀ in the same period of the position data i ₀ Because there is, it moves to STP9 without correction.

In STP9, the offset amount σ which has been added up to now is subtracted from X _0, to calculate the exact absolute position X ₀ ^*.
In addition, in the calculation (2) of the absolute position described above,
Position data PA (A) from absolute position detector 53
= 0, position data from linear inductosyn 71
It is necessary to store Pi (i) = i ₁ in a non-volatile memory so that the absolute position can always be calculated according to the flow shown in FIG.

In the above description, in FIG. 1, the linear detector (inductosyn) was used as the first detection means.
Also in this case, it is not limited to the inductosyn, and may be a magnescale or the like. In short, the rotary type, the linear type and the like may be used as long as they have a fixed period and the absolute amount can be measured within one period.

Further, in FIG. 1, a rotary absolute position detector is used as the second detecting means, but a linear type may also be used for this.

Furthermore, although one method of processing the measurement data is shown in FIGS. 3 and 5, the present invention is not limited to this method.

〔The invention's effect〕

As described above, according to the present invention, even if the drive shaft is twisted or played, the current position of the measured member such as the linear table or the rotary table can be accurately adjusted as an absolute amount with a relatively simple configuration. Can be detected.

[Brief description of drawings]

1 to 5 show an embodiment of the present invention.
The figure is a diagram for explaining the arrangement and hybrid (hybrid) measurement circuit configuration in the case of using an absolute position detector and a linear inductousin, and FIG. 2 is a diagram showing the absolute position hybridized by the device of FIG. FIG. 3 is a waveform diagram for explaining the process for detecting by the (hybrid) method, and FIG. 3 is a process for calculating the absolute position by the hybrid (hybrid) method from the waveform diagram shown in FIG. 2 by the CPU of FIG. Fig. 4 is the first flowchart
FIG. 5 is a waveform diagram for explaining a process for detecting an absolute position by a hybrid system when the absolute position detector and the linear inductosyn are not adjusted to each other in the apparatus shown in FIG. 6 is a flowchart for explaining a process of calculating an absolute position by the hybrid system in the CPU of FIG. 1 from the waveform diagram shown in FIG. 50 …… Motor, 51 …… Table (member to be measured), 52 ……
Driver / converter circuit, 53 ... Absolute position detector (second detection means), 57 ... CPU (calculation means), 6
3 ... Ball screw (drive shaft), 71 ... Linear inductosyn (first detection means).

Front Page Continuation (72) Inventor Kimio Suzuki 3-2068 Ooka, Numazu City, Shizuoka Prefecture Numazu Works, Toshiba Machine Co., Ltd. (56) References JP-A-57-171207 (JP, A) JP-A-56-135102 (JP) , A) JP-A-50-48951 (JP, A) JP-A-55-10520 (JP, A) JP-A-58-9003 (JP, A) Noriaki Suda "Electromagnetic rangefinder" 1980-2-5 , Morichi Publishing, P. 108-118

Claims (3)

[Claims] 1. A device for detecting an amount of movement of a member to be measured, which is moved by rotation of a drive shaft, from a predetermined reference state position as an absolute value, wherein the member to be measured is moved. A first detection unit that is provided along the direction and that generates a periodic electric signal corresponding to the movement amount of the measured member; and an absolute value that is connected to the drive shaft and that matches the movement amount of the measured member. Second detecting means that can be detected, and an error due to twisting or playing of the drive shaft is within a range of ± λ / 4 of the period λ of the first detecting means; and the first detecting means. And a calculating means for obtaining the absolute position of the member to be measured based on the output signal from the second detecting means, wherein the calculating means is the first detecting means included in the output signal of the second detecting means. The integer value of the period The product of the cycle of the first detection means is the first
The first means for obtaining the absolute value of the member to be measured by adding the movement amount based on the output signal of the detection means, and the value obtained by subtracting the product from the output signal of the second detection means is λ / 4 or less. And the output signal of the first detection means is 3λ / 4
When the above value or the difference is 3λ / 4 or more and the output signal of the first detecting means is λ / 4 or less, one cycle λ of the first detecting means with respect to the absolute value. And a second means for correcting the absolute position of the absolute position. 2. The method according to claim 1, wherein
The detection means of (1) is composed of a linear position detector in which an electric signal generated corresponding to the amount of movement with respect to the member to be measured changes by one phase when moving by one grid, and an absolute position detecting device. 3. The method according to claim 1, wherein
The detecting device of (1) is configured so that the absolute value that matches the amount of movement with respect to the member to be measured is a value that can cover the entire stroke length of the member to be measured.