JPS62163108A - Original point resetting system for numerical controller - Google Patents

Abstract

PURPOSE:To reset a numerical controller to its original point at a high speed and with high accuracy by using a monitor means to monitor a commanded position as well as the difference between said commanded position and the actual position and calculating the position of the original point from the data of the monitor means. CONSTITUTION:An original point reset monitor part 15 stores the droop at the relevant time point and the commanded position (cumulative value of shift commands 1a) and at the same time delivers a marker reception signal 15a to a main processor 1 to inform reception of a marker 6a. Thus the processor 1 can read the marker position information 15b when the signal 15a received. Then the remaining shift amount PR up to an original point P2 is defined as PR=PX-(PP-PM+D), where the droop 13 is referred to as D when the marker 6a is detected together with the commanded position as PM, the distance between the prescribed marker and the original point as PX, and the commanded position as PP at an instant when the processor 1 tries to perform this calculation. As a result, the remaining shift amount is calculated correctly and a numerical controller is reset to its original point with no deceleration.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a home return method for a numerical control device, and particularly relates to a home return method for a numerical control device using a position detector that outputs incremental pulses. .

[Conventional technology]

Conventionally, various position detectors have been used in numerical control devices, such as resolvers and inductosyns, but recently rotary encoders and linear pulse scales are often used. When these position detectors detect the entire minimum movement (or angle), the physiognomy correction B
In addition to phase pulses, all two-phase or marker pulses are output (see FIGS. 5(a) to 5(c)). This Z
The phase or marker is one pulse per revolution in the case of a rotary encoder, eJ total VCH total, x on a linear pulse scale.

Normally, one pulse is generated during each stroke. Next, FIG. 4 shows an example of a case where origin return is performed to determine the mechanical origin in a servo system using this type of position detector. FIG. 1 is a block diagram showing the origin return method of a conventional numerical control device.

In the figure, 1 is a main processor that outputs movement commands 1a 'e, 2 is a servo control section, 3 is a motor driven by this servo control section 2, and 4 is this motor 3.
A ball screw 5 is connected to the rotating shaft of the ball screw 4, and a mechanical part 6 is fixed to the mechanical part 5 and is moved as a decoy by the rotation of the ball screw 4.
Read the position of the linear pulse scale 6, 7U, and position feedback consisting of the A-phase signal and the B-phase signal shown in Figures 5(a) and 5(b). Signal 7a and 2 shown in Figure 5(c)
A reading head of a linear pulse scale 6 that outputs the entire marker detection signal Tb of the phase (marker), 8 is a deceleration dog for returning to the origin attached to the mechanical part 5, 9 is provided corresponding to this deceleration dog 8, A limit switch 10t outputs a deceleration detection signal 9a, and a flip-flop is set by the marker detection signal 7b. Next, refer to the time chart shown in FIG. 6 for the operation of the origin return method of the numerical control device with the above configuration. and explain. In order to move the main processor 1 in a determined direction at a determined speed, the main processor 1 outputs a movement command 1a'r indicated by a command position 11 in FIG. 6 to the servo control section 2. Then, this servo control m2 executes this movement command 1a'Th.
In response, the motor 3 is driven and controlled. Therefore, the ball screw 4 rotates and the mechanical part 5 moves to the actual position 12 shown in FIG. Then, when the mechanical part 5 moves and the deceleration dog 8 steps on the IJ mitsu switch 9 at time tl in FIG. 6, a deceleration detection signal 9a is output to the main processor 1 (note that at time t2 in FIG. The deceleration dog 8 passes the limit switch 9 and turns off). Therefore, the main processor 1 outputs the entire F deceleration command to the input of this deceleration detection signal 9a, and starts detecting the marker 6a. and,
When the reading head T detects the marker 6a k at time t3 in FIG. 6, the marker detection signal 7b''r flip-flop 1
Output to 0. Sorry, this flip-flop 10
is set based on this marker detection signal 7b, and notifies the main processor 1 that a marker has been detected. Therefore, the main processor 1 gives a movement command 1a'e to the servo control unit 2 by the distance (preset) from the marker 6a to the origin.

Then, when the mechanical section 5 stops at the origin position, the operation of returning to the origin is completed.

[Problem that the invention seeks to solve]

In the above-mentioned conventional return-to-origin method of numerical control equipment, the demand for higher speed and higher precision of numerically controlled machine tools has become stronger in recent years, but there is a deviation between the commanded position and the actual position.
The so-called True 113 (see Figure 6) is used. The reason why this droop 13 occurs is that there is a time delay in the servo control unit 2 in order to perform acceleration/deceleration control and to incorporate positional feedback looseness. This droop generally increases in proportion to the commanded speed. At the moment when the marker is detected when returning to the origin,
The main processor 1 calculates the entire origin position based on the commanded position at that time, but since the actual position differs by the droop, the origin position also shifts by the droop. Therefore, in order to increase the accuracy of the origin position, it is necessary to make the droop negligible, and it is necessary to reduce the speed when detecting the marker. Furthermore, the response time for the main processor to detect a marker and find out if it is a friend also causes deterioration in the accuracy of the origin position, and for this reason as well, it is necessary to slow down when detecting a marker. In this way, since it is necessary to sufficiently decelerate before detecting the marker, a problem often arises in that the return-to-origin operation is very slow.

[Means for solving problems]

The origin return method of the numerical control device according to the present invention is provided with a monitoring means for monitoring a commanded position and the difference between the commanded position and the actual position, and the origin position is calculated and controlled using data held by this monitoring means. It's a friend.

[For production]

The present invention can eliminate the influence of droop that exists in the servo control section, and can also achieve high-speed and highly accurate return to origin even if there is rarely a delay in the processing of the main processor.

[LOL example]

FIG. 1 is a block diagram showing an embodiment of the origin return method for a numerical control device according to the present invention. In the same figure, 14
When the marker detection signal 7b is input, the droop signal 14a at that time is outputted to the origin return monitoring section described below. 15 is the droop signal 14a that is input from this servo control section 14. A return-to-origin monitoring section monitors the movement command signal 1a input from 1 and outputs a marker reception signal 15a and completed marker position information 15b to the main processor 1.

Note that the droop and command position at the time of marker detection are collectively referred to as marker position information.

Next, the operation of the origin return method of the numerical control device according to the above 411# will be explained with reference to FIGS. 2 and 3. First, the main processor 1 determines to move in a determined direction at a determined speed, and outputs a movement command 1a k indicated by a command position 1116 to the servo control unit 2. Therefore, the servo control unit 2 receives the movement command 1ai and motor 3
to drive and control.

Therefore, the ball screw 4 rotates, and the mechanical part 5 moves to the position shown by the police position 17 in FIG. Then, the linear pulse scaler 6 moves and the reading head 7 moves to the marker 6a.
When detected at (marker position P+ in Fig. 2)',
The marker detection signal 7b' is output to the origin return monitoring section 15. Therefore, this home return monitoring unit 15 stores the droop and command position at that moment (this is the cumulative value of the movement command 1a), and also stores the marker reception signal 15a.
f Output to the main processor 1 to notify that the marker 6a has been received. Therefore, when the marker reception signal 15a is input, the main processor 1 receives the marker position information 1.
5b'e can be read. Here, as shown in FIG. 2, droop 13'! at the time of marker detection! rD, command position kPM + distance 'ePX from a predetermined marker to the origin, and PP to the command position at the moment when the main processor 1 attempts to perform this sum 'N, then the origin Pg
The remaining movement amount pR up to (see FIG. 2) can be expressed by the following equation.

PR=Px-(PP-PM+D) Therefore, no matter how fast the marker detection speed is, the time T(
See Figure 2], no matter how large the PPIPMID
Since the amount of movement is accurately known, the remaining travel distance can be calculated correctly and the return to origin can be performed with high accuracy without deceleration.
can be done. Note that at the processing time tx of the main processor 1 (see Fig. 2), the command position does not change during processing. Of course I can do what I can. If it is difficult to synchronize the process of returning to the origin with the process of giving commands to the servo control unit 2, it will be necessary to temporarily stop all movement commands, but the main processor 1 will not be able to calculate all of the above equations. Normally, it has almost no effect on the movement of the mechanical parts. [When a rotary encoder is used as a position detector, as shown in Figure 3, each rotation of the encoder (for example, if the lead of the ball screw is 10 mm, this corresponds to 10 mm of movement of the mechanical part). To,
A marker is generated and the marker position is pla, Pxb
+ Indicated by Ptc.

Therefore, in order to determine which marker to use as a reference to determine the origin P* vi- during the entire stroke of the mechanical part,
Dog, limit switch and droop are markers 1
A certain degree of deceleration is necessary so as not to exceed the cycle.9, the deceleration dog is turned on at time tl, the deceleration dog is turned off at time t2, the marker is detected at time ts, and the main processor 1 is activated at time t4. Indicates the processing point. In this way,
Return to origin can be performed quickly and with high precision.

〔Effect of the invention〕

As explained in detail above, the origin return method of the numerical control device according to the present invention eliminates the influence of droop.
Moreover, even if there is a delay in the processing of the main processor, the advantage is that the origin return monitoring unit can be operated at high speed and with high precision.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of the origin return method of a numerical control device according to the present invention, and FIGS.
Figure 4 is a block diagram showing the origin return method of a conventional numerical control device. Figures 5(a) to 5(C) are diagrams showing the relationship between command positions and police positions. 6 is a time chart showing the output waveform of the linear pulse scale shown in FIG. 1... Main processor, 3... Motor, 4...
...Ball screw, 5..Mechanical stand, 6..Linear pulse scale, T-shape...Reading head, 14..
... Servo control unit, 15... Origin return monitoring unit.

Claims (1)

[Claims] In a numerical control device that performs servo control using a position detector that outputs a pulse that discriminates the minimum amount of movement and a pulse that is output once per period or complete stroke, the position detector detects a marker and generates a marker detection signal. When outputting
Droop D when the marker is detected, commanded position P_M, distance P_X from the predetermined marker to the origin, commanded position P_P at the moment when the main processor tries to perform the calculation.
From, the origin position P_R is calculated as P_R=P_X-(P_P-
A return-to-origin method for a numerical control device, characterized in that the return-to-origin method is calculated using P_M+D).