JPH0542601B2 - - Google Patents

Description

Detailed Description of the Invention Technical Field The present invention provides a multi-rotation absolute detector capable of detecting the absolute amount of rotation of a rotating body rotating one or more revolutions from its original position without actually rotating the rotating body from its original position. It is related to.

Prior Art A multi-rotation absolute detector is used, for example, to absolutely detect the distance from the origin of a linearly moving member fed by a ball screw. By detecting the absolute amount of rotation of the ball screw from its original position,
The position of the linearly moving member can be determined from the detection result and the pitch of the ball screw.

Optical absolute encoders are currently used as a means to absolutely detect the absolute amount of rotation from the original position of a rotating body that is rotated multiple times, such as the ball screw mentioned above, but these are generally expensive and difficult to detect. Those that obtain a large number of rotations are particularly expensive.

Purpose of the Invention In view of the above-mentioned circumstances, the present invention has been made with the object of providing an inexpensive multi-rotation absolute detector that can absolutely detect the absolute amount of rotation from the original position of a rotating body that rotates many times. It is what was done.

Structure of the Invention In order to achieve this object, the multi-rotation absolute detector according to the present invention is (1) a one-rotation absolute detector that detects the absolute rotation amount within one rotation as a digital value, and a sub-rotation detector connected and rotated thereby; (2)
(3) a one-rotation absolute detector similar to the lower rotation detector, which is connected to the lower rotation detector and reduces the rotation of the lower rotation detector at a constant reduction ratio; From the upper rotation detector connected to the device, (4) the output value a ₁ of the lower rotation detector and the output value a ₂ of the upper rotation detector, the following formula a ₂ =Z + nY / N + a ₁ /N However, , Z: Value representing the deviation of the origin of the upper rotation detector from the origin of the lower rotation detector in output units of the upper rotation detector Y: Output value corresponding to one rotation of the lower rotation detector 1/N: Deceleration An integer n that satisfies the ratio within the set tolerance.
Find the absolute rotation amount of the rotating body using the integer
and a processing device that calculates nY+a ₁ .

Effect As described above, if the lower rotation detector is connected to the rotating body whose rotation amount is to be detected, and the upper rotation detector is connected to the lower rotation detector via the reduction gear, the upper rotation detector The motor rotates by an amount obtained by multiplying the rotation amount of the lower rotation detector by the reduction ratio of the reduction gear. Conversely, by dividing the rotation amount of the upper rotation detector by the reduction ratio of the reduction gear, the number of rotations of the lower rotation detector can be obtained, and the rotation amount and lower rotation of the lower rotation detector corresponding to the number of rotations can be obtained. The sum with the current output value of the detector is the absolute amount of rotation of the rotating body from its original position.

However, in reality, there may be a synchronization difference between the carry of the lower rotation detector and the change in the output value of the upper rotation detector (a difference in the reading timing of the output value between the lower rotation detector and the upper rotation detector, or a backlash of the reduction gear). Errors occur due to rotational deviations between the lower rotation detector and the upper rotation detector caused by the become. In the vicinity of the position where the lower rotation detector makes exactly one revolution (hereinafter referred to as the boundary area), the detection error becomes extremely large due to the above-mentioned synchronization deviation, whereas in other areas (hereinafter referred to as the general area) Since the detection error is small,
The above (1)
The purpose is to find an integer n that satisfies the formula. If the detection error exceeds the tolerance, it is assumed that a large detection error due to synchronization has occurred, and the integer n is increased or decreased by 1 to find an integer n that makes the detection error less than the tolerance. Once n is determined, the absolute rotation amount nY+a ₁ of the rotating body can be determined by calculation.

Effects of the Invention That is, the multi-rotation absolute detector of the present invention can detect the absolute amount of rotation from the original position of a rotating body that rotates one or more revolutions using two single-rotation absolute detectors, for example. It becomes possible to use an inexpensive one-turn absolute detector such as a resolver, a relatively inexpensive one-turn absolute encoder even if it is an optical type, and an inexpensive multi-turn absolute detector can be obtained.

Further, in the present invention, since it is possible to use a speed reduction device having an arbitrary speed reduction ratio, it is very advantageous in actually manufacturing the device. For example, if a speed reduction device with an arbitrary speed reduction ratio can be used, a general commercially available speed reduction device can be used, and manufacturing costs can be reduced. Furthermore, when a dedicated speed reduction device is used, the capabilities of the upper and lower rotation detectors can be utilized to the fullest. When the maximum detection values of the upper and lower rotation detectors are determined, the maximum value that can be detected by both rotation detectors can be changed arbitrarily by appropriately selecting the reduction ratio of the reduction gear. Therefore, the capabilities of both rotation detectors can be utilized to the fullest according to the maximum amount of rotation to be detected.

Additionally, the upper rotation detector is provided to determine how many rotations the lower rotation detector has made before reaching the current rotational position, and the detection accuracy of the absolute rotation amount of the rotating body is determined by the accuracy of the lower rotation detector. The upper rotation detector does not affect the detection accuracy. Therefore, it is not necessary to align the origin of the upper rotation detector with respect to the lower rotation detector so strictly, and high precision is not required for the holding device and deceleration device of the upper rotation detector, which also contributes to cost reduction. Effects can be obtained.

Although the multi-rotation absolute detector according to the present invention can be manufactured at low cost as described above, it can enjoy all the inherent advantages of an absolute detector. In other words, there is no need to perform the troublesome return-to-origin operation every time the device is started up, as is the case with incremental rotation detectors, and there is no need for switches, control equipment, etc. for this operation, which also reduces trouble. . In addition, there is no need to back up the various memories in the processing device using a battery, and even in the event of a power outage, it is possible to start the system from where it stopped without having to return to its home position, allowing it to control the position of machine tool slides, etc. It can be widely used for controlling loader/unloader devices, robots, etc.

Examples Hereinafter, a case where the multi-rotation absolute detector of the present invention is used in a linearly moving member position detection device for absolutely detecting the absolute position of a linearly moving member moved by a ball screw, that is, the distance from the origin will be described as an example. This will be explained in detail based on the figures.

FIG. 1 is a perspective view schematically showing a mechanical part of a linearly moving member position detection device, and in the figure, reference numeral 1 is
0 indicates a ball screw. The ball screw 10 is rotatably but immovably attached to the main body of the machine tool by a bearing member (not shown), and is connected to a drive source at the right end in the figure, and is moved by being rotationally driven. The member 12 is moved back and forth in a linear direction.

A first resolver 14 is directly connected to the left end of the ball screw 10 and rotates at the same speed as the ball screw 10. In the first resolver 14, a rotor having a rotor coil is rotatably disposed within a stator having a stator coil, and when the rotor is rotated in an excited state of the rotor coil, a terminal of the stator coil is connected to the rotor. It generates a resolver signal, which is an electrical signal corresponding to the rotational position. This first resolver 14
As shown in FIG. 2, the resolver signal switch 1
6 to a resolver signal digital converter (hereinafter abbreviated as an R/D converter) 18. The R/D converter 18 emits a digital signal corresponding to the rotational position of the rotor of the first resolver 14 based on the resolver signal from the first resolver 14, and in this embodiment, the digital signal corresponds to the rotational position of the rotor of the first resolver 14. ²¹² digital signals are emitted, and together with the first resolver 14, it constitutes a one-rotation absolute detector which is a lower-order rotation detector.

A second resolver 22 is further connected to the ball screw 10 via a reduction gear device 20. In FIG. 1, the reduction gear 20 is shown for ease of understanding.
is shown as a simple gear train, but in reality, this reduction gear 20 requires a large reduction ratio, so for example, a planetary gear reduction device commercially available under the trade name Harmonic Drive is used. suitable. The second resolver 22 has the same structure as the first resolver 14, but as is clear from FIG. has been done. The resolver signal switch 16 selectively supplies the resolver signal from the first resolver 14 and the resolver signal from the second resolver 22 to the R/D converter 18 based on a command from the microcomputer 24. , these second resolvers 22 and R/D
The converter 18 constitutes an upper rotation detector.
The computer 24 uses the first resolver 1 when necessary.
Digital signals corresponding to the resolver signals from the first resolver 14 and the second resolver 22 (hereinafter referred to as digital output values of the first resolver 14 and the second resolver 22, respectively) can be read from the R/D converter 18. ing. Note that 26 is an excitation circuit for exciting the rotor coils of both the resolvers 14 and 22.

A carry borrow detector 28 is further connected to the R/D converter 18. This carry borrow detector 28 detects 1 when the content of the digital signal of the R/D converter 18 transitions from a full count to zero.
It generates two up clock signals, and conversely, one down clock signal when transitioning from zero to full count. An up-down counter 30 is connected to this carry-borrow detector 28, which is further connected to a microcomputer 24.
It is connected to the. The up-down counter 30 stores the numerical value supplied from the microcomputer 24, and increments the numerical value by 1 each time an up-clock signal is generated from the carry-borrow detector 28, and by 1 each time a down-clock signal is generated. The contents of this up-down counter 30 can be read into the microcomputer 24 when necessary.

The microcomputer 24 is further connected to an input device 32 equipped with various keys and operation buttons, and a display 34 for displaying instructions, inquiries, and position detection results from the microcomputer 24 to the operator. There is.

Further, the first resolver 14 and the second resolver 2
As shown in FIG. 3, the mounting flange 38 is fixed to the main body of the machine tool by a plurality of fixing members 40 and bolts 42, and bolt 42
Loosen the mounting flange 38 using the fixing member 40.
By releasing the tightening, the mounting angle position to the machine tool body can be easily adjusted.

Before explaining the operation of the position detection device configured as above, the operating principle of this device will be explained in the fourth section.
This will be explained based on the diagram.

In FIG. 4, a straight line L1 indicates a mechanical movement limit position at which the movable member 12 comes into contact with a stopper and is prevented from moving further, and a straight line L2 indicates the origin position of the position detection device. Further, the first digital value, that is, the digital output value of the first resolver 14 is represented on the horizontal straight line L3, and the second digital value, that is, the digital output value of the second resolver 22 is represented on the straight line L4. There is.

It is assumed that the origin of the first resolver 14 is aligned with the origin O of the movable member 12, the origin of the second resolver 22 is aligned with the mechanical movement limit position, and the current position of the movable member 12 is represented by point A. , the digital output value of the first resolver 14 will indicate the distance a ₁ and the digital output value of the second resolver 22 will indicate the distance a ₂ . Therefore, in the following description, for convenience, it is assumed that the output value of the first resolver 14 is a ₁ and the second output value is a ₂ . The digital output value of the second resolver 22 represents the current position of the moving member 12 (however, it is not a position based on the origin O, but a position based on the straight line L1), but its accuracy is low. . On the other hand, although the output value of the first resolver 14 has high accuracy, it is not the position from the origin O, but the point B.
It merely represents the position from. The origin O of this point B
It is known that the position based on is an integer n times the output value Y corresponding to one rotation of the first resolver 14;
cannot be known from the output value alone.

However, this integer n can be determined from the output value a ₁ of the first resolver 14 , the output value a ₂ of the second resolver 22 , and the reduction ratio 1/N of the reduction gear device 20 .
That is, since the amount of rotation of the second resolver 22 is 1/N of the amount of rotation of the first resolver 14, the following equation holds true as is clear from FIG.

a ₂ = Z + nY / N + a ₁ /N ... (1) Here, Z is a value that represents the deviation of the origin of the upper rotation detector from the origin of the lower rotation detector in units of output of the upper rotation detector. , this value should be selected appropriately according to the circumstances of the mechanical parts such as the ball screw 10, moving member 12, and stopper.Here, suppose that Z=3Y/2N...(2) If so, the above equation (1) becomes a ₂ =3Y/2N+nY/N+a ₁ /N (3).

In this equation (3), the output value Y corresponding to the reduction ratio N of the reduction gear 20 and one rotation of the first resolver 14
is known in advance, so by substituting the output value a ₁ of the first resolver 14 and the output value a ₂ of the second resolver 22 into this equation, the integer n is uniquely determined. Therefore, from the output value a ₁ of the first resolver 14 and the output value a ₂ of the second resolver 22, the position of the moving member 12, that is, the multiple rotations representing the accurate distance from the origin O to the point A representing the current position. The output value nY+a ₁ of the absolute detector can be accurately determined.

The operation of the embodiment apparatus based on the above principle will be explained with reference to the flowcharts shown in FIGS. 5 to 7.

FIG. 5 is a flowchart of the main program that controls the overall operation of the apparatus of this embodiment. Upon power-on, normal initial reset, self-diagnosis, etc. are automatically executed in step S1.
Subsequently, in step S2, the original position alignment flag is cleared, which is provided for the purpose of aligning the original positions of the first resolver 14 and the second resolver 22. Next, a position detection section reset subroutine in step S3 is executed, and this subroutine will be described in detail later. After various steps have been executed, it is determined in step S4 whether or not the original position alignment button of the input device 32 has been operated, and if it has been operated, the program execution moves to step S5. However, after the original positioning flag is set, the original positioning subroutine of step S6 is executed.

This subroutine of step S6
2 is located at the origin, as shown in FIG. That is, in step S7, an instruction to select the first resolver 14 is issued to the resolver signal switch 16, and in step S8, the display 3
4, a display indicating that the first resolver 14 has been selected is displayed, and the digital output value of the first resolver 14 is read into the microcomputer 24, and the value is displayed on the display 34. Thereafter, in step S9, it is determined whether or not the set completion button of the input device 32 has been operated, and if it has not been operated, the program execution continues at step S9.
Returning to S8, steps S8 and S9 are repeated until the set completion button is operated.

While this execution is repeated, if the operator loosens the bolt 42 of the fixing member 40 that fixes the first resolver 14 and changes the mounting angle position of the first resolver 14 with respect to the machine tool body, the Since the displayed value on the display 34 changes, when this displayed value becomes 0, tighten the bolt 42 to fix the first resolver 14 and operate the set completion button, and the work of adjusting the original position of the first resolver 14 is completed. finish.

Thereafter, steps S10, S11, and S12 are similarly executed to adjust the original position of the second resolver 22. However, in this case, the display value on the display 34 is not set to 0, but as 3Y/2N as described above.
The mounting angle position of the second resolver 22 with respect to the machine tool body is adjusted so that.

In this way, the original position alignment of the first resolver 14 and the second resolver 22 can be performed extremely easily, and once this original position alignment is completed, the program execution returns to step S3 in the main routine of FIG. By executing the position detection unit reset subroutine in step S3, the current position of the movable member 12 (here, the origin position) is displayed on the display 34, and thereafter changes every moment as the movable member 12 moves. The absolute position is displayed on display 3.
4 will be displayed.

This position detection section reset subroutine is as shown in FIG. 7, and will be explained in detail below. However, this subroutine is executed independently of the above-mentioned original position alignment subroutine even when the power is turned on after the position detection device is put into use. , a case will be explained in which the moving member 12 is at the general position represented by point A in FIG. do.

The second resolver 22 is selected in step S13, and the output value _a2 of the second resolver 22 is stored in the _a2 memory in step S14. However, this step S14 is executed after a sufficient period of time has elapsed, for example, several milliseconds, after the resolver signal switch 16 is activated.

Subsequently, steps S15 and S16 are similarly executed, and the output value of the first resolver 14 is stored in the _a1 memory.

After that, steps S17 to S22 are executed,
It is possible to find n that satisfies the above formula (3). Note that the calculation result in step S19 should theoretically be 0, but in reality, an error of several counts occurs when the first resolver 14 and second resolver 22 are set to their original positions, so it may not be 0. It won't happen. Therefore, in step S20, it is not determined whether the error result of step S19 is 0, but whether it is within ±10, and if it is within ±10, n that satisfies equation (3) is determined. It has come to be judged as something that has been achieved. In this embodiment, this ±10 is the allowable error. In addition, to find n, n must be increased by 1 from 0 to find n that satisfies equation (3), but this n cannot be exceeded by the reduction ratio N of the reduction gear 20. (In such a case, the second resolver 22 will rotate more than one revolution, but in reality, the specifications are set so that it does not rotate more than one revolution, so this is not a problem.) In step S21, If it is discovered that such a situation has occurred, the program execution moves to step S23, the original position alignment error flag is set, this fact is displayed on the display 34, and the program execution continues as shown in FIG. Return to the main program. Therefore, the operator operates the original position adjustment button and performs steps S5 and S6.
The original position alignment work will be performed.

After the integer n (the number of rotations of the first resolver 14) that satisfies equation (3) is obtained as described above, steps S24 and S25 are executed, and the integer m that is 2 larger than n is stored in the up-down counter 30. is set. This means that when the value of the integer n itself is set in the up-down counter 30, the moving member 12
This is to avoid that when the up-down counter 30 moves in a negative direction from the origin O, the contents of the up-down counter 30 become negative, making data processing troublesome. Therefore, when the moving member 12 is at the origin position, the contents of the _a1 memory are 0, but the contents of the m memory are 2. In addition, the up-down counter 30
The content of is that the current position of the moving member 12 is the ball screw 1.
This is a number that indicates the number of rotations relative to 0 (actually, the value is n plus 2, so the number of rotations is 2 more than the actual number of rotations of the ball screw 10), and it indicates the position currently being explained. It functions as higher-order data of the 12-bit data of the first resolver 14 that is read every moment when the device of this embodiment detects the position after the detection section reset subroutine has been executed.

Therefore, as described above, the current position of the moving member 12 is now in a state where absolute detection is possible, but in reality, after the output value of the first resolver 14 is stored in the _a1 memory in step S16, The output value of the first resolver 14 may change between 0 and full count within a very short time until it is set in the up-down counter 30 in step S25. In such a case, an extremely large reading error will occur, so in this embodiment, steps S16 and S25 are
A check is made to see if such an error has occurred between the two, and if such a reading error has occurred, correction of such a reading error is automatically performed. Steps S26 to S29 are the checking process, and steps S30 to S32 are the correction process. The check is performed using extended data (D ₁ and D ₂ ) whose upper data is the output value D ₁ of the up-down counter 30 read in step S26 and whose lower data is the output value D ₂ of the first resolver 14 read in step S27. step
In step S24, the value m stored in the m memory is taken as the upper data, and in step S16, the output value _a1 of the first resolver 14 stored in the _a1 memory is taken as the _lower data. This is done by determining whether the That is, in step S28, the extended data ( _D1 .
D ₂ ) and the extended data (m・a ₁ ) are calculated,
In step S29, it is determined whether the calculation result is within ±100. If the calculation result is within ±100, it is determined that the above-mentioned inconvenience did not occur, but if it exceeds ±100, it is determined that the above-mentioned inconvenience has occurred. The correction is made in steps S30 to S32. That is, in step S30, it is determined whether or not the calculation result in step S28 is greater than 100. If it is greater than 100, it is determined that the current position of the moving member 12 has been excessively read due to a reading error, and step S31 is performed. At this point, the contents of the up-down counter 30 are decremented by one. If the calculation result in step S28 is less than minus 100, it is determined that the current position of the moving member 12 has been read too low due to a reading error, and the content of the up-down counter 30 is incremented by 1 in step S32. after that,
The program execution returns to step S26 and the same thing is executed, but in this case the determination result at step S29 should be YES, and the program execution moves to step S33.

In step S33, it is determined whether or not the original position alignment flag is set, but since this flag is not set in normal use, the original position alignment error flag is cleared in step S34. is executed (no change will occur if this flag is not set), and the execution of the position detector reset subroutine ends.

The case where the position detection unit reset subroutine is executed with the moving member 12 placed at a general position other than the home position has been described above.Next, when the home position adjustment button of the input device 32 is operated Now, a case will be explained in which the position detection unit reset subroutine is executed after the original position alignment subroutine is executed.

In this case, the movable member 12 is positioned at the origin, and the first resolver 14 and the second resolver 22 have been temporarily aligned in their original positions by executing the original positioning subroutine, so step S18 is performed. After 0 is stored in the n memory, the result of the first judgment in step S20 is
It is normal for the answer to be YES. Therefore, from now on, steps S24 to S32 are executed with the integer n being 0, but in this case, since the original alignment flag was set in step S5, the determination result in step S33 is YES
Then, steps S35 and S36 are executed, and it is determined whether or not the original position alignment has been performed with sufficient accuracy. That is, if the determination result in step S20 is YES with the movable member 12 positioned at the origin, it means that the original position alignment has been performed with a certain degree of accuracy.
S20 is not originally a step for determining whether or not the original position alignment has been performed with sufficient accuracy, but is a step executed to know the general current position of the moving member 12, so the margin is ± It is widely taken as 10. On the other hand, it is desirable that original positioning be performed with even higher accuracy, so step
In S36, the accuracy of the original position alignment is checked with a narrow margin of ±5. If the determination result in step S36 is NO, the program execution proceeds to step S23, where the original position alignment error flag is set, and a message to that effect is displayed on the display 34, but the process proceeds to step S36. If the judgment result is YES, it means that the original position alignment was performed with sufficient accuracy, and you can proceed with the step.
In S34, the original position alignment abnormality flag is cleared, and the execution of the position detection section reset subroutine (step S3) associated with the original position alignment work is completed.

In the embodiment described in detail above, the one-rotation absolute detector constituting the lower rotation detector consists of the first resolver 14 and the R/D converter 18, and the one-rotation absolute detector constituting the upper rotation detector The absolute detector consists of a second resolver 22 and an R/D converter 18. In other words, two digital one-rotation absolute detectors are constructed by two inexpensive resolvers and one R/D converter commonly used for them, and this embodiment uses an optical type This has the unique advantage that a multi-rotation absolute detector can be constructed at a lower cost than when two absolute encoders are used.

However, even when using an optical absolute encoder as a digital single-rotation absolute detector, the equipment cost is still considerably lower than when using an optical multi-rotation absolute encoder. The effects of the present invention can also be obtained by using two rotary absolute encoders.

Furthermore, the present invention can of course be used not only to detect the absolute position of a moving member that moves linearly, but also to absolutely detect the rotational position itself of a rotating body that rotates many times.

It goes without saying that the present invention can be practiced with various modifications and improvements made to the mechanical parts such as the speed reducer, the electronic control circuit, the control program, etc. based on the knowledge of those skilled in the art.

[Brief explanation of drawings]

FIG. 1 is a perspective view showing the principle of a mechanical part of a linear moving member position detection device using a multi-rotation absolute detector, which is an embodiment of the present invention, and FIG. 2 is a perspective view showing the electronic control circuit of the device. It is a block diagram. FIG. 3 is a perspective view showing the resolver fixing means in the apparatus shown in FIG. 1. Figure 4 is the first
FIG. 4 is a diagram for explaining the operating principle of the device shown in FIGS. 5 to 7 are flowcharts showing only the portions necessary for understanding the present invention out of the control program stored in advance in the microcomputer of the apparatus of the above embodiment. 10... Ball screw, 12... Moving member, 14
...First resolver, 16... Resolver signal switch, 18... Resolver signal digital converter (R/D converter), 20... Speed reduction device, 22... Second resolver, 24... Microcomputer, 2
8...Carry borrow detector, 30...Up-down counter, 34...Display device, 38...Mounting flange, 40...Fixing member.

Claims (1)

[Scope of Claims] 1. A device for detecting the absolute amount of rotation from the original position of a rotating body rotating one or more revolutions, which is a one-turn absolute detector that detects the absolute amount of rotation within one revolution as a digital value. a lower rotation detector connected to the rotating body and rotated by the rotating body; a reduction gear device connected to the lower rotation detector and reducing the rotation of the lower rotation detector at a constant reduction ratio; A one-rotation absolute detector similar to the lower rotation detector, which is connected to the reduction gear, and an output value _a1 of the lower rotation detector and an output value _a2 of the upper rotation detector. From this, the following formula a ₂ = Z + nY / N + a ₁ /N where, Z: the deviation of the origin of the upper rotation detector from the origin of the lower rotation detector expressed in the output unit of the upper rotation detector Y: lower rotation Output value corresponding to one revolution of the detector 1/N: Integer n that satisfies the reduction ratio within the set tolerance range
Find the absolute rotation amount of the rotating body using the integer
A multi-rotation absolute detector comprising a processing device that calculates nY+a ₁ .