JPS60120403A - Detection for origin - Google Patents

Abstract

PURPOSE:To detect an origin position in a high speed with a high precision inexpensively by performing the adjustment for origin positioning electrically with software. CONSTITUTION:A length L between a changeover point between the high level and the low level of an edge signal P4 and an encoder origin pulse P2 is measured. For the purpose of measuring this length L, a table 16 is driven in the direction determined by the high level or the low level of said signal P4, and counting of encoder output pulses P5 is started by a counter when an edge 30 is detected. This table 16 is driven by a motor 10 through a ball screw 12, and said counting is continued until the encoder origin pulse P2 is generated. This counted value of the counter indicates the length L and is stored in a non- volatile memory. When the origin is positioned after power is supplied, an edge signal is searched by said method, and a window is set with software on a basis of the value of the stored length L.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field of the Invention The present invention relates to a method for detecting the origin position required in robots, numerical control devices, and the like.

Prior Art and Problems In numerical control (NC) devices, as shown in FIG. 1, a device that converts the rotational motion of a motor 10 into linear motion using a ball screw 12 and a nut 14 to drive a table 16 left and right is often used. Ru. In this type of device, the origin position of the table 16 is determined in order to control its position, but the simplest and most direct method for detecting this origin position is to attach a position detection device to the table 16. . The position detection device includes a mechanical limit switch, but as a non-contact detection means, a slit 18 and a light emitter/receiver 20, 22 as shown are widely used.

That is, a slit plate is attached to one end of the table, and a floodlight 20 is installed.
The light receivers 22 are mounted above and below this slit plate and at desired positions in the transverse direction of the table 16.
The table position is detected by the light from the light emitter 20 passing through the slit 18 of the slit plate and entering the light receiver 22. If the light emitter and receiver are positioned so that the light from the light emitter passes through the slit and enters the light receiver when the table 16 is at the origin, the table origin position can be detected from the output of the light receiver.

However, the slit 18 has a certain width, so that the receiver output pulse has a width W as shown in FIG. 2+C). In systems where this is insufficient in accuracy, a low-return encoder 24 is attached to the ball screw 12 and its origin pulse is utilized. As is well known, a rotary encoder has slits that divide 360" into n equal parts and generates n pulses per rotation.
In order to be able to detect the direction of rotation, a two-phase A and B pulse train is used, and one origin pulse is also generated per rotation.

Figure 2 (al is one phase of the A and B two-phase pulses, f
b) shows the origin pulse. Therefore, the output pulse P1 of the optical receiver 22 and the encoder origin pulse P2 (!: (')
By calculating the logical product, the output pulse P3 shown in FIG. 2 1dl is obtained, which allows the origin position to be aligned with high precision. That is, the output pulse P3 in FIG. 2 (0) is
If the initial setting is made so as to obtain the origin at No. 6, it becomes possible to detect the table origin position with extremely high accuracy from then on.

However, this method is difficult to adjust. That is, table 16
If the origin position is precisely specified, the position of the slit 18 and the encoder origin pulse output position must be set so that the output pulse P3 of FIG. 2 1dl is obtained at the origin position. On the other hand, in a method in which the origin position of the table 16 is set by the slit 18 and its reproducibility is improved, the encoder origin pulse may be generated within the output pulse width of the light receiver 22. However, adjustment in this case is also not easy. That is, since the output pulse PI of the photoreceiver 22 has jitter (indicated by diagonal lines) as shown in FIG. , If the origin pulse P2 is shifted by δ from the center as shown in Figure 3 (al), depending on the size of δ, it will be outside the receiver output pulse, and the high precision origin pulse P3 in Figure 2 (dl) cannot be obtained. A situation (that does not occur) occurs.

Note that jitter included in the output of the light receiver 22 occurs for the following reasons. That is, since the output of the light receiver 22 has a width in the slit 18 and the light beam from the emitter 20 also has a width, the output rises when the light beam begins to pass through the slit, saturates eventually, and then falls. . This is amplified and rectangularized using a threshold value, as shown in Figure 2 (C1), but the rise and fall timings of the rectangular pulse may vary due to gain fluctuations, threshold fluctuations, noise contamination, etc. of the circuit that performs this rectangularization processing (shaping). Some fluctuations, or jitters, occur.Figure 3(b)
j indicates the amplitude of jitter.

If the width W of the receiver output pulse P1 is increased, the encoder origin pulse P2 will be less likely to deviate from the line width W, and there will be no need to accurately align it to the center, but of course the line width W will be larger than the period of the origin pulse P2. It must not be large (if it is large, the next origin pulse P2 will generate the origin pulse P3).

Therefore, in the conventional method, the position of the slit indicating the table origin must be adjusted while watching the encoder origin pulse, which is troublesome, and the slit position and/or the encoder origin position may shift due to aging, parts replacement, or some other trouble. The problem is that similar adjustments must be made when

Purpose of the Invention The present invention has been developed in view of the fact that functionally adjusting the table slit position and the encoder origin pulse generation position as in the conventional method requires a large amount of man-hours and cannot be adjusted to the best of its ability. The purpose of this invention is to provide a method that can optimally detect the origin at high speed and at low cost by performing adjustment electrically and by software without using mechanical methods.

Structure of the Invention The present invention provides a device for driving a low-speed rotational or linear motion system by a high-speed rotational motion system via a deceleration mechanism. In a method for detecting an accurate origin position of a low-speed motion system by providing a detection device that outputs an origin position signal automatically, and using the output of these detection devices, the origin position is determined by the detection device of the low-speed motion system at an initial setting stage. output, move in a certain direction from this, calculate the distance until the detection device of the high-speed motion system generates an output, store this in memory, and in the operation stage, combine this memorized distance with the high-speed and low-speed motion The present invention is characterized in that the accurate origin position of the low-speed motion system is detected using the output of the system, and this will be explained next with reference to examples.

Embodiment of the invention FIG. 4 shows an embodiment of the invention. It is identical to FIG. 1, and like parts are given the same reference numerals throughout the figures. 26 and 28 are bearings that support the pole screw 12. Reference numeral 30 corresponds to the slit 18 in FIG. 1, but in this embodiment it is simply an edge (a tongue piece that acts as an edge). The light beam from the projector 20 reaches the edge 30
When the light is blocked, the output of the light receiver 22 becomes L (low) or off, and when it is not blocked, the output becomes H (high) or on.
The output P4 is as shown in FIG. If an edge is used, there is no need to pay attention to the width W as with a slit, and it is cheaper than using a slit. Also, the output is as shown in the figure <H
, L, it is clear which direction to move in order to approach the origin (H, L switching point) (if L, move backward to the left; if H, move forward to the right), so this output is It also shows polarity.

The slit in Figure 1 has no polarity and only detects coincidence or mismatch, so although it is not shown in Figure 1, this method also produces an output that shows polarity as shown in Figure 5 (C). is necessary.

In the present invention, first, the H level of the edge signal P4 shown in FIG. 5(C) is
, the distance between one point of L rotation and the encoder origin pulse P2 is measured. For this, a counter is used to generate the encoder output pulse P.
A method of measuring 5 is used. Specifically, to detect an edge, that is, to drive the table 16 in the direction determined by the H and L of the edge signal P4 until the H and L turning point of tc+ in FIG. Start counting, advance the table 16 to the right or left (this direction is fixed), and therefore rotate the motor 0 and the ball screw 12 in that direction.
The above counting is continued until the encoder origin pulse P2 is generated. In this way, the counter count value when the pulse P2 is generated indicates the distance. This count value is stored in a non-volatile memory such as a bubble memory or a floppy disk.

This operation needs to be performed once after assembling the NC device, robot, etc., but compared to conventional mechanical adjustment work, it can be performed almost automatically and in a shorter time, and moreover, the distance of 11 tL can be adjusted with high precision. can be measured.

During operation, when aligning the origin after power is turned on, edge signals are searched for using the same method as above. When an edge is detected, a window P6 of L±ΔL as shown in FIG. When the origin pulse P2 is generated, it is accepted (taken in), and the counting of the encoder output pulse P5 is started by edge detection, and when the count reaches (L-ΔL), the above acceptance is permitted, This is regressed until the count value becomes (L + ΔL). In this example, the origin pulse P2 is generated at time 0 when the count value approaches L, and this becomes the accurate origin pulse P3.

If only the reproducibility accuracy of the origin position is required or in a device, the pulse P3 generation point may be used as the table origin.

Further, if the table origin determined by the edge 30 has an important meaning, it is sufficient to move back by a distance from the generation position of the pulse P3 while counting the pulse P5, and set the point P as the table origin. Although this point P is near the edge detection point, it is characterized in that it contains no jitter and is more accurate than when determining this point P using an edge signal.

Edge signal P4 also has jitter due to noise etc.
Also, due to changes over time, it is expected that the window P6 will be slightly shifted from the initial position (01),
If ΔI is set appropriately so that the first pulse after edge detection can be reliably captured within this window, the table origin will be determined by the encoder 24 and extremely high accuracy will be achieved.

The window width 2ΔL is determined from the uncertainty of the edge signal P4 and the period of the encoder origin pulse P2. This is more flexible and accurate than conventional methods of fixing the mechanical slit width. Additionally, the edge allows the table origin, which is precisely set during assembly, to be reproduced for many years.

As described in detail, according to the present invention, the number of man-hours required for adjusting the origin detection system can be significantly reduced, and even if there is trouble such as parts replacement due to failure of the motor and/or encoder, the recovery work can be carried out quickly. The window center position and width can be set more accurately than the conventional mechanical adjustment of slit width and position, and the absolute origin position accuracy can be maintained semi-permanently, maximizing the margin against the uncertainty of the table origin signal P4. This provides advantages such as guaranteed performance.

In the embodiment, a system that converts rotational motion into linear motion is used, but the present invention can also be used in a system that converts high-speed rotation into low-speed rotation using a speed reducer.

That is, the motor 10 and pole screw 12 of the embodiment are used as a high-speed rotation system, the nut 14 and the table 16 are used as a low-speed linear or rotation system, and the ball screw 12 and nut 14 are used as a deceleration mechanism that connects these. Applicable to systems or mechanisms that

[Brief explanation of the drawing]

Figures 1 to 3 are explanatory diagrams of the conventional origin position detection method;
6 to 6 are explanatory diagrams of the origin detection method of the present invention. In the drawing, 10.12 is a high-speed rotational motion system, 24 is a detection device that outputs its origin position signal, 12.14 is a deceleration mechanism, 1
4.16 is a low-speed linear motion system, 18.20, 22.30
is the origin position detection device, and L is the distance to be stored in the memory. Applicant Fujitsu Ltd. Representative Patent Attorney Minoru Aoyagi 3 8-

Claims (3)

[Claims] (1) In a device that drives a low-speed rotational or linear motion system by a high-speed rotational motion system via a deceleration mechanism, the low-speed motion system is provided with a device that detects its origin position, and the high-speed motion system also periodically points to the origin. In a method of providing a detection device that outputs a position signal and detecting an accurate origin position of a low-speed motion system using the outputs of these detection devices, the detection device of the low-speed motion system is made to output the origin position at an initial setting stage. , calculate the distance until the detection device of the high-speed motion system generates an output by moving in a certain direction from this, and record this in the memory. 1. An origin detection method comprising detecting an accurate origin position of a low-speed motion system using the output of a detection device. (2) At the operation stage, the low-speed motion system is moved in the same direction as the initial setting from the position where the detection device generated the origin detection output, and a predetermined value is added to and subtracted from the distance stored in the memory. Claim 1, characterized in that the output position of the detection device of the high-speed motion system is determined within this range, and the output position is set as the accurate origin position of the low-speed motion system. Origin detection method described. (3) At the operation stage, move the low-speed motion system from the position where the detection device generated the origin detection device in the same direction as the initial setting by adding and subtracting a predetermined value to the distance stored in the memory. The apparatus is characterized in that the output of the detection device of the high-speed motion system is measured within this range, and the output position is returned by the distance described above to make that position the accurate origin position of the low-speed motion system. An origin detection method according to claim 1.