KR950014594B1 - Method for compensating error of multiaxis in dnc system - Google Patents

Abstract

The multi-axial error compensation method is prepared for generating an NC program compensating inner mechanical errors such as a straightness error and a holing error in DNC system of a NC lathe in order to perform precise machining. The method comprises a step of computing reference values based on detection; a step of generating coordinate values of a work piece according to the values of error elements; a step of generating the NC program based on the coordinate values of a work-piece.

Description

Multi-axis error compensation method of DNC system

1 is a configuration diagram of a DNC system.

2 is a schematic diagram showing the operation of the NC lathe.

3 is a flow chart of the multi-axis error compensation method of the present invention.

* Explanation of symbols for the main parts of the drawings

1: Spindle head 2: Slide in Z direction

3: Slide in the X direction

The present invention relates to a multi-axis error compensation method for generating a corrected NC program in consideration of the internal error of the machine, that is, the straightness error and the mounting error in the NC lathe.

The NC lathe uses an NC controller to create the required shape by moving the tool while rotating the workpiece. NC controller is operated by a series of commands. NC program is used. The NC program is executed using commands for various preparation functions, interpolation commands, and special functions added by the user.

As an example of NC program, referring to linear interpolation, in the form of "G01 × 100.000 Z 50.000 F l00", G01 is the command related to linear interpolation, × 100.000 Z 50.000 is the position to move, and F00.

Eventually, the tool moves according to the above left unfolded. At this time, the coordinate value is determined by the shape of the workpiece, and the other feed speed, spindle speed, etc. are determined by the cutting conditions.

The use of a CAD / CAM system in creating such an NC program results in the DNC system shown in FIG. To automatically generate an NC program based on the shape of the workpiece, shape data and machining data are required. Shape data can be constructed in a CAD system and represent the external shape of the workpiece, primarily as face and solid data. Machining data shows machining conditions, tool paths, etc., and tool shape and machine structure are also considered.

The basic program generated by the shape data and the machining data is a CL file. The CL file is not a command system that can be directly recognized by the NC controller, so it goes through a post processor.

The post processor uses a CL file to convert it into an NC program that can be recognized by a particular NC controller. NC program is transferred to NC controller using LAN. A LAN is a network between computers that exchanges information with each other. Upper LAN and lower LAN have nothing special except communication range.

The NC program sent to the CELL controller is downloaded to the specific NC lathe by the NC controller, whereby the NC lathe is processed.

The existing DNC system ignores various error components of the NC lathe itself. The error components include workpiece mounting errors, spindle errors, and slide feed errors. Workpiece mounting errors are errors caused by the chuck, etc., and appear as mounting errors on the reference plane. After all, the machining reference plane and the machining plane have that much error. The main shaft rotational error is a change in the instantaneous rotational axis of the workpiece with respect to the tool, which causes problems in power supply and surface roughness. In addition, the spindle rotation error is difficult to correct and if the problem is to increase the performance of the spindle.

The feed error of the slide includes positioning error in the feed direction, straight error in the right direction, rolling, yawing, and pitching. These errors reduce the dimensional accuracy and shape accuracy of the workpiece. .

These various errors are described in the test report during the inspection of the machine tool, and the user can use the data, so that the degree of deterioration may be a problem even when machining precision parts without considering such an error.

SUMMARY OF THE INVENTION An object of the present invention is to provide a multi-axis error compensation method capable of producing a corrected NC program in consideration of the error component of each element in relative motion in order to improve the generated NC program without considering the error of each part of the conventional NC lathe. The aim is to provide a multi-axis error compensation DNC system that improves machining accuracy and quality.

Referring to FIG. 2, an NC shelf is schematically shown, which is divided into a T-base type as shown in FIG. 2 (A) and a cross slide type as shown in FIG. 2 (B) according to the slide type. Lose.

In the T-base type, slides 2 and 3 in the Z and X directions are divided into each other and arranged in a T shape. In the cross slide type, two slides 2 and 3 are arranged overlapping each other.

Considering the cross slide type first, various error components are as follows.

Error of the workpiece mounting device (chuck) in front of the spindle head 1, mounting error on the bed of the spindle head, mounting and feed error of the slide 2 to the bed, slide to the slide 2 (3) There may be a mounting error and a transportation error.

If the X, Y, Z Cartesian coordinate system is set for each element unit, and the geometric error is expressed as δ A, it is as follows.

[Delta] s is a microstrain component, mainly an error due to elastic deformation.

δxx, δyy, and δzz denote shrinkage and tension in the X, Y, and Z axis directions.

δr is the rotational error.

α, β, and γ are rotational errors of the X, Y, and Z axes, which are rolling, yawing and pitching.

delta t is associated with microtranslation and is positioning accuracy.

However, since the error due to elastic deformation is small, ignoring the sum of δr and δt becomes an error matrix, denoted by ε.

When N is the number of elements, the error vector Δro is as follows.

Where A is the coordinate transformation matrix, which defines the kinematic relationship between the elements. In other words, since the spindle head rotates with respect to the bed, it becomes a rotation transformation matrix. γN is a tool shape vector that is different depending on the contact point or the tangent side.

The error vector Δro of the Z-axis NC lathe thus calculated is as follows.

Here, the subscripts 0, 1, 2, 3 attached to δx, δy, δz, α, β, and γ denote each element unit. Ie o is the spindle head with workpiece, 1 is bed (slide (2) for T-base type). 2 is a slide 2 (bed in the case of the T-base type), 3 is a slide 3 including a tool.

After all, δy ₃ is the y-direction error of the slide 3.

In the NC program, the value of "X" in "X 100.000 Z 50.000" corresponds to the above α value and the "Z" value corresponds to the above z value.

By the way, since the lathe is a rotating body, the value calculated by calculating θ up to 0 ° 360 ° should be regarded as an error of the position.

In the end, the X and Z coordinate values of each interpolation command are substituted into the above equation, and θ is repeatedly calculated from 0 ° to 360 °, and the average value is replaced with the existing program coordinate value.

In the multi-axis error compensation method of the DNC system of the present invention, when the NC lathe is operated as shown in FIG. 3, the NC controller reads the NC program, calculates an error of the NC lathe, corrects the NC program, and generates a corrected NC program. The corrected NC program is downloaded to the NC lathe and the machine starts machining.

The multi-axis error compensation method of the present invention configured as described above is a NC program by predicting a machining error due to the relative error of each component of the NC lathe such as spindle head (1), slide (2), slide (3) and bed. By calibrating the workpiece, it improves precision and provides stability of quality.

Claims (4)

In the process of converting the shape data provided by the CAD / CAM system into an NC program that can be recognized by the NC controller of the machine tool, the NC program is corrected by considering the error components of each element of the NC machine that performs relative motion. In a multi-axis error compensation DNC system, in which the classification consists of a spindle head, a bed, and a slide, including a tool, the error component of each element includes a mounting error of the workpiece, a rotational error of the spindle, and a feed error of the slide part. Multi-axis error compensation method of DNC system The method of claim 1, wherein the correcting of the NC program comprises: calculating an error component of each element based on a value according to a precision test; calculating a workpiece coordinate value to be corrected according to the error component value; A multi-axis error compensation method of a DNC system comprising the steps of creating an NC program according to coordinate values. The method of claim 2, wherein the step of calculating the error component is obtained by converting the error values of all elements in the NC coordinate system and the feed error of the slide portion is represented by the following general formula as a geometric error on the Cartesian coordinate system Multi-axis error compensation method of system. (Δs is elastic transformation error, δr is rotation component error, δt is positioning error due to translational component error) The method of claim 3, wherein the obtaining of the coordinate values to be corrected is performed by substituting the NC coordinate values (X, Z) of each interpolation command into the following equation and calculating θ over the main power range to obtain the average value as the coordinate values of the NC program. Multi-axis error compensation method of the DNC system characterized in that the setting. (expression)