RU2641939C2 - Device of determination of geometric errors in motion trajectory of milling-machine table with computer numerical control - Google Patents

Abstract

FIELD: measuring equipment.

SUBSTANCE: device contains a mandrel with a tapered end that a measuring rod contacts with. The measuring rod is installed on the antifriction guideways of the bed. The mandrel is designed to be installed in the machine spindle. In addition, the device contains a cylindrical hinged joint with a support, fixed on the machine table, with the joint being installed on the said hinged joint to rotate the body and carriage, on which the bed is installed, and with the second measurement device, measuring shifts of the body in relation to the support of the cylindrical hinged joint. Using the program, the machine table is set to move in circular trajectory with a center, associated with the spindle axis. The geometric errors of the table motion trajectory are determined by a measuring device, while they are reduced by the shift of the body in relation to the hinged joint, which is determined by the second measuring device. The obtained error of the circular trajectory is recalculated into errors for X and Y axes, which are used for the respective compensation in the Computer Numerical Control system.

EFFECT: application of the invention allows to increase the milling accuracy.

5 dwg

Description

The present invention relates to the field of technological processes, namely to metal cutting machines. The device is designed to determine the geometric errors of metal-cutting machines for the purpose of their subsequent software compensation and can be used on machines with numerical control of the milling group to increase the accuracy of processing workpieces on them.

A device for measuring the errors of a metal cutting machine [1], consisting of a rod connecting two ball bearings on the machine with the possibility of free rotation around these supports and measuring the distance between these supports. When using the device, one of the supports is programmed to move along a circular path centered in another support. The measured displacement between the supports is defined as the error of the circular path.

The disadvantage [1] is the inability to adjust the rod along the length and, as a consequence, the inability to measure on various machines with different stroke lengths, as well as the influence of support errors on the measurement accuracy.

The closest to the merits of the claimed invention, the prototype is a device for determining the accuracy of the tool path [2]. The device consists of a mandrel, which is attached to the end surface of the spindle, at the end of the mandrel there is a holder with a ball, which is contacted by one of the ends of the measuring sensor, the second end of which is mounted in a cylindrical hinge, which in turn is mounted in the working body of the machine. The spindle unit and the working body make coordinated movements according to the program, from which the resulting circular path is formed with a radius equal to the length of the measuring sensor. Machine errors affect the change in the radius of the specified path, which is measured by the sensor. The patent describing this device indicates the possibility of measuring displacement in the radial direction and recording this result, but does not indicate the possibility of determining and correcting errors along the coordinate axes.

The disadvantage of the prototype [2] is the inability to adjust the device along the length and, as a consequence, the inability to measure on various machines with different stroke lengths, as well as the influence of errors of the holder and the cylindrical hinge on the measurement accuracy. Also, the patent does not indicate the possibility of determining and correcting errors along coordinate axes.

The aim of the invention is the ability to measure and compensate for geometric errors of machines with different stroke lengths, as well as improving the manufacturability and accuracy of the device for measuring.

The technical result consists in the possibility of measuring the geometric errors of the machine, compensating for the measured errors and thereby increasing the accuracy of processing on the machine.

The goals are achieved in that the device comprises a measuring rod, making a circular motion along with the machine table and shifted in the presence of errors in the motion path, and the offset of the measuring rod is recorded by the measuring device. The position of the measuring device can be adjusted relative to the housing of the device, which allows measurement at different stroke lengths. The contact of the measuring rod with one support is made at a point, which reduces the influence of manufacturing errors; the error of the second support is compensated by a second measuring device. The design of the device uses guides and rolling bearings, which reduces friction and improves the accuracy of movement.

A device for determining geometric errors in the trajectory of the table of CNC milling machines contains a mandrel with the possibility of placement in the spindle of the machine and with a conical end, which is contacted by a measuring rod mounted on the guide rails of the rack. The offset of the measuring rod is controlled by the measuring device. The stand has the ability to adjust the displacement along the body, which is mounted on a cylindrical hinge with the possibility of placement on the machine table. According to the program, the machine table is set to move along a circular path with a center corresponding to the axis of the spindle. The radius of this circular path materializes in the form of a measuring rod located between the top of the sending cone and the measuring device. Geometric errors, including positioning errors, of the machine are manifested in the form of distortion of a circular path, in particular its radius, which is measured by a measuring device. After passing a certain polar angle, the radius error determined by the measuring device, the X and Y coordinates of the current point of the circular path are recorded as the measurement results. From the measured error of the radius, the displacement of the plate relative to the hinge, determined by the second measuring device, is subtracted. The obtained error of the circular path is converted into errors on the X and Y axes. The calculated axial errors are entered into the compensation program of the numerical control system.

The essence of the claimed invention is illustrated in FIG. 1, FIG. 2, FIG. 3, FIG. 4 and FIG. 5. In FIG. 1 shows a general view of the device in isometric view. In FIG. 2 shows a longitudinal section of the device along the midline. In FIG. 3 shows a graph of errors obtained by processing the measurement results with compensation parameters. In FIG. 4 representations the table in which the results of measurements using the device are entered. In FIG. 5 illustrates a method for determining the amount of play.

The device consists of a support 1 (Fig. 1) with the possibility of rigid attachment to the table 2 of the machine, the housing 3, which can be rotated relative to the spike 4 of the support 1 on the bearing 5 (Fig. 2), fixed by a cover 6. The bearing 5 and the spike 4 form a cylindrical hinge. A measuring device 7 is fixed in the housing 3, which is designed to control the displacement of the housing 3 relative to the support 1. A carriage 8 is attached to the housing 3 with the possibility of adjusting movement, to which, in turn, a rack 9 is mounted, inside which an adjustment sleeve 10 with a nut 11 is installed. the sleeve 10 is based in a rack 9 on a conical surface. A separator 12 is attached to the adjusting sleeve 10, inside of which balls 13 are installed. The balls 13 are in contact with the measuring rod 14. The measuring rod 14, in turn, abuts against the conical surface 15 of the mandrel 16, which can be placed in the machine spindle. At the end of the measuring rod 14, the pin of the measuring device 17 abuts, which is fixed in the bracket 18 mounted in the carriage 8. On the mandrel 16 there is an inner ring of the bearing 19, the outer ring of which is connected to the clamp 20. The clamp 20 is connected to the stand 9. by means of the leash 21 22 mandrel 16 is mounted in the spindle of the machine. The initial position of the carriage 8 sets the stop 23. The adjustment size can be set by the block of end measures 24.

The methodology of using the device is based on measuring the deviation of the radius-vector of a circular path relative to a fixed center. When the table moves along a circular path, various geometric errors of the machine cause the table to deviate from the ideal circle. These deviations can lead to a deviation of the radius vector of this circle. Deviations of the radius vector materialize in the form of the displacement of the measuring rod 14, which is recorded by the measuring device 17. Thus, a number of geometric errors of the machine can be detected by the displacement of the measuring rod in contact with the center when the table moves along a circular path.

The fixed center is materialized by the top of the cone 15 of the mandrel 16. To eliminate the runout of the cone 15 when the machine spindle rotates, the cone must be processed directly on the machine itself. To perform the test, the table 2 of the machine is informed of the movement of circular interpolation (plane motion along a circular path) with the center lying at the intersection of the table plane and the axis of the machine spindle. Together with the machine table 2, the support 1 moves, while the housing 3 rotates relative to the spike 4 of the support 1, so that the axis of the measuring rod 14 always intersects the spindle axis (was located strictly in radius relative to the center of the circular path). This is facilitated by the design of the clamp system. The leash 21 is fixed in the collar 20, so that it is always directed perpendicular to the axis of the spindle. The leash 21 is connected with the stand 9 by means of a kinematic pair with only a translational degree of freedom. Therefore, for any displacement of the table, the lead 21 together with the clamp 20 rotates on the bearing 19, respectively, turning the housing 3 with the stand 9 relative to the spike 4, thereby orienting the measuring rod 14 strictly radially relative to the center of the circular path. The measuring rod 14 is constantly pressed against the conical surface 15 of the mandrel 14 using the internal spring of the measuring device 17. The measurement accuracy is affected by friction and gaps when moving the measuring rod relative to the guide. To reduce these factors, the rolling guides are used between the measuring rod 14 and the stand 9. The rolling guides are realized in the form of balls 13, which are arranged in three rows in the separator 12. Clearance of the gaps in the guide is achieved by tightening the nut 11, while the adjusting sleeve 10 is pulled inward into the conical holes of the rack 9, is deformed in the radial direction and thereby eliminates the gap between the inner diameter of the adjusting sleeve 10, the balls 13 and the measuring rod 14.

The procedure for using the device during operation is as follows.

1. Set the table of the machine 2 (Fig. 1, Fig. 2) in the middle position, ie so that the center of the table is in the middle of the stroke along each axis. This position is a local zero for the measurement program, that is, the coordinates of this table position should be entered as zero offsets at the X and Y coordinates, for example, under the address G54. This method allows you to program the movement along a circular path with a radius l _izm .

2. Insert the shank 22 of the mandrel 16 into the bore of the machine spindle.

3. Move table 2 along the X coordinate by l _izm

where l _x is the stroke length along the shortest coordinate, s is the margin, depends on the possibility of installing the support.

The use of such a value l _izm as the length of the radius vector allows measurements with the largest possible radius, which makes it possible to detect errors over the largest possible stroke length.

4. Install the device on the machine table. The support 1 must be based on the middle groove of the table 2. The end face of the carriage 8 must be brought to a stop 23. Insert the lead 21 into the hole of the clamp 20. Bring the measuring rod 14 in contact with the conical surface 15 of the mandrel 16 by moving the supports 2 on the table 2. Make the initial zeroing the measuring device 17 with an interference fit of about 1 mm.

5. Set the device to the required measurement length l _izm . To adjust the size, holding the carriage 8 in place, move the housing 3 together with the support 1 by a distance

where l ₀ is the initial displacement due to the design of the device. The size can be set using the block of end measures 24. At the end of the adjustment, the carriage 8 is fixed with a latch, the support 1 is fixed on the table 2 of the machine. Set measuring device 7 to zero.

6. In the MDI mode, dial the circular interpolation program clockwise (G02), recording the readings of the measuring device at equal intervals of the trajectory. The radius with which the movement is made must be equal to l _izm , the center of the circle will be at the point whose X and Y coordinates are fixed at address G54. The program job block is different for different CNC systems. In the SINUMERIK 840D CNC system, a block of such a movement may look like this

G03G54G91AR = 5RP = 50F100.

In this case, a circle movement with a radius of 50 mm will be performed with a normal working feed of about 100 mm / min, with a stop every 5 °.

In some CNC systems, it may not be possible to program arc motion in polar coordinates in increments. In this case, the measurement program can be represented as a cycle with displacements in increments and a rotation function of the coordinate system. Program code for SINUMERIK802S might look like this

(R10 is the rotation angle counter, R11 is the final coordinate along X, R12 is the final coordinate along Y, R13 = 5 ° is the rotation angle; the values of the variables can be set in advance in the corresponding control area);

G17 G54 (indicating the address of the coordinate offset);

R10 = 0 (zeroing of the counter of an angle of rotation);

MARK2: (setting the label of the transition that starts the cycle);

IFR10> 360 GOTOFMARK1 (comparison of the angle of rotation counter with a full revolution, when it reaches the end of the cycle);

R12 = 50 * SIN (R13) (calculation of the final X coordinate);

R11 = 50 * COS (R13) (calculation of the final Y coordinate);

G03 X = R11 Y = R12 CR = 50 F100 (assignment of the circular interpolation command);

M00 (stop the program for fixing the results);

G259 RPL = R13 (rotation of the coordinate system);

R10 = R10 + R13 (increase in the counter of the angle of rotation);

GOTOBMARK2 (transition to the beginning of the cycle);

MARK1: M30 (end of program);

7. Turn on the development of the program. After stopping movement, lock the deviation δr _edited by the measuring device 17; deviation δr _pogr caused by device errors by measuring device 7. The final value of the radius vector deviation, taking into account the compensation of device errors:

8. Enter the data in MicrosoftExcel in the form of a table shown in FIG. 4, where θ _e is the absolute current angle of rotation.

Resume the program, repeating these steps until the full circle.

9. To plot a deviation δr _e depending on the angle θ _e in the polar coordinate system

θ _{e is} calculated on an accrual basis from the values of the increments of the angle.

According to the measurement results, the following geometric errors of the machine can be determined.

1. Deviation from circularity of the trajectory of circular interpolation. As a result, the largest value of δr _{e is} taken.

2. Positioning errors along coordinate axes.

Errors along the coordinate axes are understood to be errors due to both inaccuracy of positioning along a given coordinate and the indirectness of the trajectory of movement in the horizontal direction along the perpendicular coordinate. The main objective of the measurement process is to determine the values of the errors necessary for subsequent software compensation, therefore, it is quite sufficient to present the resulting error, without dividing it into its components.

To enter compensation values, it is necessary that the coordinate values at which compensation values are entered are equally spaced. Therefore, compensation values need to be taken not at equal increments of the angle θ, but at angles corresponding to uniform coordinate values. For this, the obtained dependence δr _e = ƒ (θ _e ) must be represented as a spline

where r (θ) is the radius vector of the points of the curve specified by the spline; Bi (θ) - i-th basic spline; R _i is the radius vector of the i-th control point.

To approximate the experimental curve in the form of a spline, you can use the Mathcad program. To solve this problem, cubic splines are used. Cubic spline interpolation allows you to draw a curve through a set of points so that the first and second derivatives of the curve are continuous at each point. This curve is formed by creating a series of cubic polynomials passing through sets of three adjacent points. The cubic polynomials are then joined together to form a single curve. To implement the calculations, the Mathcadlspline function (θ _e , δr _e ) is used. This function generates a spline curve, which approaches a straight line at the boundary points, and a vector of coefficients of the second derivatives is formed, which is denoted by νs. The arguments are: θ _e - the values of the rotation angles in the measurement cycle, for the calculation a matrix is formed of a column of values from the column θ of the table (Fig. 4); δr _e - error values corresponding to θ _e ; a column matrix is also formed from the column δr of the table (Fig. 4).

The use of this particular function gives a better approximation to the initial dependence than the use of other interpolation functions.

For further calculation, it is necessary to form a new array of values of the rotation angles θ corresponding to uniform values of the X and Y coordinates. For this, for example, a range of values is determined along the X axis (-l _izm ; l _izm ). An array of coordinates {Xi} is formed inside the range with a certain step value tx.

Next, the rotation angles θ corresponding to the values of the array {Xi} are calculated

The values of θ are calculated taking into account the quadrant of the angle. Next, the error values are determined for new values of the angle θ

where interp (νs, θ _e , δr _e , θ) is a function that returns the spline value for values of in the initial vectors X and Y and in the spline coefficients νs.

Considering small increments of the polar angle θ _j , at each step j it is possible to determine the positioning errors along the X axis with a positive direction of movement

In the range from 180 to 0 °, positioning errors are determined with a negative direction of movement

Similarly, the errors are determined along the Y axis.

According to the results of calculations, you can determine the parameters that can be entered in the compensation program of the CNC system. For example, along the Y axis, the compensation program for Sinumerik CNC systems is as follows (the corresponding graph is shown in Fig. 3):

MD: MM_ENC_COMP_MAXPOINTS [0, Y] = 6 (the number of compensation reference points in the array {Yi} is set);

$ AA_ENC_COMP_STEP [0, Y] = 20 (the distance between the control points is set, equal to the step tx)

$ AA_ENC_COMP_M1N [0, Y] = - 50 (sets the starting point along the axis);

% _N_EECDAT_EEC_INI (compensation table is entered);

$ AA_ENCCOMP_MAX [0, Y] = 50 (sets the end point along the axis);

$ AA_ENC_COMP [0,0, Y] = 0 (in subsequent frames, the compensation values at the reference points with the corresponding numbers are recorded, the compensation values are determined by the formulas)

$ AA_ENC_COMP [0,1, Y] = - 37.207

$ AA_ENC_COMP [0,2, Y] = - 10.627

$ AA_ENC_COMP [0.3, Y] = - 8.017

$ AA_ENC_COMP [0.4, Y] = - 12.079

$ AA_ENC_COMP [0.5, Y] = 93.407

M17 (end of the compensation table assignment routine);

3. Backlash in feed drives.

As can be seen from FIG. 5, the values of the errors at the same reference point are not the same for different directions of motion. This is due to the presence of play in the feed drive. Thus, according to the test results, the play along the axis can be defined as the difference between the values of axial errors, for example, for the X axis - ΔXj ↑ and ΔXj ↓, that is, the ordinates of the points of the positioning error graphs (Fig. 5) at the same reference points. On the X axis, the play values at the reference point j can be found

Processing the data obtained, you can find the maximum

and the average value of play along the axis

The average value can be stored in the memory of the CNC system for software error compensation. For Sinumerik CNC systems, this value is entered in the MD: BACKLASH system variable.

Similarly, the results are determined along the Y axis.

The implementation of the present invention gives the following results.

1. The ability to measure geometric errors of the machine with minimal labor by measuring the detected errors in one working pass along a circular path by evaluating the displacement of the measuring rod with a measuring device.

2. The ability to compensate for geometric errors of the machine by presenting the measurement results in a form that can be converted into values perceived by the program compensation program of the CNC machine.

3. Improving measurement accuracy:

a) due to the contact between the measuring rod and the mandrel at the point, which reduces the influence of their manufacturing errors on the measurement accuracy;

b) by compensating for errors in the manufacture of the cylindrical joint using a second measuring device;

c) through the use of rolling guides of the measuring rod and the rolling bearing in a cylindrical joint, which reduces friction in these joints and reduces the errors associated with friction.

4. The ability to effectively use this device on various machines with different stroke lengths by adjusting the position of the measuring rod with the measuring device relative to the cylindrical hinge, which makes it possible to carry out measurements with the optimal measurement radius for each machine, which, in turn, allows you to most fully identify errors on this machine.

Thus, this device allows you to measure the geometric errors of milling CNC machines with high accuracy and minimal labor for their subsequent software compensation.

The present invention satisfies the criteria of novelty, since when determining the level of technology, no means have been found that have characteristics that are identical (that is, matching the functions performed by them and the form in which these signs are performed) to all the signs listed in the claims, including the purpose of the application.

The present invention has an inventive step, because it has not been identified technical solutions having features that match the distinguishing features of this invention, and not known the influence of distinctive features on the specified technical result.

The claimed technical solution can be implemented in mechanical engineering, in technological processes of machining, to increase the accuracy of processing on metal cutting machines.

Used sources

1. US patent No. 5052115, 04.10.1991. Melvyn Burdekin. Accuracy testing device // US Patent US 5052115 A.

2. US Patent No. 386666, 06/03/1975. Charles N. Thompson, Fred W. Jones. Apparatus for measuring tool path accuracy // US Patent US 386666.

Claims (1)

A device for determining geometric errors in the trajectory of the table of the CNC milling machine, containing a mandrel made with the possibility of installation in the spindle of the machine, a measuring rod associated with the measuring device, and a cylindrical hinge with a support made with the possibility of fixing on the table of the machine, characterized in that it is provided with a housing mounted on said cylindrical hinge with a possibility of rotation with a carriage with a stand placed on it with the possibility of adjusting movement and the second measuring device, mounted in the said housing with the ability to measure the displacement of the housing relative to the support of the cylindrical hinge, while the mandrel is made with a conical working part, connected to the said rack with the possibility of rotation of the rack relative to the longitudinal axis of the mandrel and is installed with the possibility of point contact of its working part with the end said measuring rod made spherical, wherein the measuring rod is mounted in said rack on rolling guides with ozhnostyu measuring longitudinal displacement.

Images