GB2241338A - Metrologically testing and self-correcting a measurement machine for geometric detection errors - Google Patents

Abstract

Dimensional checks are made on a measurement machine 1 and a sequence of re-calibration data is automatically collected by means of which calibration data in an electronic unit 2 are modified in such a way as to adjust this latter data automatically to any changed geometric characteristics of the measurement machine 1 and achieve an automatic correction to allow for any measurement errors caused by the changed metrological characteristics of the machine 1 itself. <IMAGE>

Description

- 1 A METHOD FOR METROLOGICAL TESTING AND SELF-CORRECTION OF GEOMETRIC

DETECTION ERRORS IN A MEASURING MACHINE The present invention relates to a method for metrological testing and for self-correction of geometric detection errors in a measuring machine, particularly a measuring machine having a rotating table.

Measuring machines which are provided with a table rotatable about at least one axis, on which workpieces to be measured are disposed are known. Such machines are provided with a movable feeler element which is adapted to touch the workpiece to be measured to detect its dimensions, and is movable in three orthogonal directions along the X, Y and Z axes. Such machines are also provided with a reference element, normally a cube or a sphere, serving as a reference and dimensional origin for the measurements effected in different angular positions of the rotatable table.

Such measuring machines are connected to an electronic control unit which includes a central processing unit for data provided by the feeler of the measuring machine, and a mass memory.

The electronic unit is provided with a sequence of calibration data.which contains the characteristic values 0 JMe190291 - 2 - of the machine itself, together with information regarding the angular positions of the rotating table, the reference element and the orthogonality of the X, Y and Z axes.

These data are characteristic for each machine produced, and are stored in the memory of the electronic unit during the calibration of the machine.

Since the geometric (and therefore also the metrologic) characteristics of the machine can change because of the action of external agents (heat, vibrations, knocks etc) the data present in the memory may no longer correspond and be adapted to the machine, and can give rise during measurement to measurement errors deriving from the changed geometric configuration of the measuring machine itself. This therefore makes it necessary to re-calibrate the machine which involves relatively long mechanical operations, which must be performed by specially qualified personnel.

An object of the present invention is that of providing a method for metrological testing and self-correction of geometric detection errors in a measurement machine, particularly a measurement machine provided with a rotating table, which permits the updating and correction of the calibration data of the measuring machine on the basis of the changed geometric characteristics, with h f JMe190291 relatively simple and rapid steps so as to obviate the above-indicated disadvantages.

According to the present invention there is provided a method for metrologic testing and self-correction of geometric detection errors in a measuring machinel particularly a measuring machine provided with a rotating table; the said measuring machine being provided with a feeler element and being controlled by an electronic unit; initial calibration data of the said measuring machine being resident in the said electronic unit; the said calibration data being characteristic of an initial geometric configuration of the said measuring machine, characterised by the fact that it comprises at least a first stage of checking the dimensions of the changed geometric configuration of the measuring machine and collecting a sequence of re- calibration data, and a second stage of self-adjusting and modification of the calibration data on the basis of the said re-calibration data whereby to adapt the said calibration data to the said changed geometric characteristics of the machine.

Implementation will now be described with particular reference to the attached drawings which illustrate a preferred, non-limitative, embodiment thereof, in which:

Figure 1 illustrates in perspective a measuring 0 JMe190291 - 4 - machine provided with a rotating table connected to an electrical control unit utilising a method hereof; Figure 2 is a perspective view of the rotating table provided with first abutment means utilised by a feeler of the measuring machine of Figu re 1 to effect a first dimensional test; Figures 3, 4, 5, 6 and 7 illustrate, in perspective, the rotating table provided with the first abutment means of Figure 2, and utilised by a feeler of the measuring machine of Figure 1 to effect, respectively, a second, third, fourth, fifth and sixth dimensional test; Figure 8 is a perspective view illustrating the rotating table provided with second abutment means and utilised by a feeler of the measuring machine of Figure 1 to effect a seventh dimensional test; Figures 9, 10 and 11 are perspective views illustrating the rotating table provided with the second abutment means of Figure 8, and utilised by a feeler of the measuring machine of Figure 1 to effect respectively an eighth, ninth and tenth dimensional test; Figure 12 is a flowchart illustrating a. method embodying the present invention; Figure 13 is a more detailed flowchart of two of the blocks forming part of the flowchart illustrated in Figure 12; and Figure 14 is a more detailed flowchart of two blocks 1 j 1 1 z 1 forming part of the diagram illustrated in Figure 12.

With particular reference to Figure 1, the reference numeral 1 generally indicates a measuring machine of known 5 type which is connected to an electronic control unit 2. The electronic unit 2 includes a central processing unit 3 which is provided with a mass memory, and is connected to a lopersonal computer" 4, to a video unit 5, to a keyboard 6P and finally to a printer 8. The electronic unit 2 is further provided with a disk drive unit 7 for reading and storing information on a magnetic support.

The measuring machine 1 includes a metal support structure 9 forwardly from which extends a bed 12 which supports a pair of horizontal rectilinear parallel guides 13 on which slides, parallel to the X axis, a slide 15 provided with a circular table 17 rotatable about a vertical axis 18 parallel to the Z axis.

The structure 9 is provided with a pair of horizontal rectilinear guides 20 parallel to the Y axis and disposed above the bed 12, along which, parallel to the Y axis, slides a slide 22 provided with a rectilinear vertical guide 24.

Along the guide 24, parallel to the Z axis, slides a slide 26 forwardly from which extends an arm 28 to the end of 6 which is mounted an appropriate feeler 40 of known type, which comprises a rod 41 connected to one end of the arm 28 and provided with a forward free end from which extends, perpendicularly of the rod 41, a second rod 42 5 screwed to the rod 41 itself.

From the lower end of the rod 42 extends a rod 44 coaxial with the rod 42, of smaller diameter than the rod 42 and terminating with a ruby sphere 46 fixedly secured to the rod 44. When this rod 44 is displaced from its equilibrium position by mechanical contact on the ball 46 it is adapted to open a circuit to provide a corresponding electrical signal.

Laterally from the slide 15, moreover, extends a tubular projection 30 from the free end of which extends a vertical rod 31 which carries at its upper end a cube 32 serving as a reference element. The cube 32 is therefore fixed to the slide 15 and serves as a reference for the determination of the angular position of the table 17 with respect to a reference radius.

The machine 1 is finally provided with a plurality of sensors (of known type and not shown) which make it possible to know and measure the instantaneous position of the feeler 14 and therefore of the end ball 46, and which send the measurements of this position to the electronic i i t 1 1 0 unit 2.

The electronic unit 2 of the machine 1 is provided with a sequence of calibration data which contains the characteristic values of the machine 1 itself, with information relating to the angular positions of the table 17 and the reference element 32 with respect to the or igin of the X, Y and Z axes constituting the frame of reference of the machine, and with the orthogonal values along the axes X, Y and Z so that the detected position values determined with the feeler 40 are corrected for the inaccuracies existing in the structure of the measuring machine 1.

As has been previously mentioned, these data are characteristic for each machine produced, and are stored in the memory of the unit 2 during the calibration of the machine.

Since the geometric characteristics (and therefore also the metrological characteristics) of the machine can change because of the action of external agents (heat, vibrations, shocks etc), according to the method forming the subject of the present invention the calibration data are easily checked, and possibly modified, to obtain selfcorrection of the measurement errors deriving from the changed geometric disposition of the machine 1.

( 1 - a In particular, the method forming the subject of the present invention comprises a series of dimensional checks by means of which a set of data is detected with which the original calibration data can be modified and updated.

In particular, according to the method the subject of the present invention, the orthogonality of the set of three cartesian axes X, Y and Z of the machine 1 is tested and possibly corrected by means of a first series of dimensional checks, and the position of the cube 32 with respect to the centre of the rotating table is tested and corrected by means of a second series of dimensional checks. During these dimensional checks, moreover, a series of points are measured from which re-calibration data characterising the modified geometrical disposition of the machine 1 are processed; such data are subsequently utilised to modify and correct the calibration data originally present in the memory of the electronic unit 2.

To effect these first and second series of dimensional checks a circular plate 50 is mounted on the table 17 and fixed to this plate 50 there are tools which function as abutment and reference elements for the detection of the reference points.

In particular, with reference to Figure 1. the plate 50 is shown not yet mounted on the rotating table 17. and is j i A 0 C. 1 provided with five threaded holes the first four holes of which are indicated 100, 101, 102 and 103, and are disposed at the corners of a square, and a fifth hole 105 is disposed at the point of intersection of the diagonals 5 of the square, exactly at the centre of the plate 50.

To prepare the machine 1 for the performance of the first series of dimensional checks four tools such as those illustrated in Figures from 2 to 7 are screwed into the four holes 100-103 and securely fixed to the plate 50.

In particular the lower threaded end of a rectilinear rod 110 is screwed into the hole 100 and extends perpendicularly of the plate 50, terminating at its upper end with a ball 111 fixedly connected to the rod 110 itself.

Into the hole 101 there is screwed the lower end of a rectilinear metal support 115 which extends perpendicularly of the plate 50 and is provided at its upper end with a projecting metal plate 116 which is securely fixed to the support 115 and is parallel to the plate 50. The metal plate 116 carries a short rectilinear rod 117 which extends perpendicularly of the plate 116, terminating at its upper end with a ball 118 f ixedly connected to the rod 117 itself.

1 Into the hole 102 is screwed the threaded lower end of a rod 120 which is identical to the rod 110 and which extends perpendicularly of the plate 50, terminating at its upper end with a ball 121 fixedly connected to the rod 5 120 itself.

Finally, into the hole 103 is screwed the threaded lower end of a rod 125 which has a height which is very much less than the rods 110 and 120 and which extends perpendicularly of the plate 50, terminating at its upper end with a ball 126 fixedly connected to the rod 125 itself. To perform the first series of checks no tool is screwed into the hole 104.

To effect this first series of checks the ball 46 of the feeler 40 is able to touch, in a manner which will be described in more detail hereinbelow, the balls 111, 118, 121 and 126 in such a way as to gather data to be utilised to check the orthogonality of the axes of the machine.

To perform the second series of dimensional checks the lower end of a rod 130 is screwed into the hole 104, which rod has the same dimensions as the rod 125, and extends perpendicularly with respect to the plate 50, terminating at its upper end with a ball 131 fixedly connected to the rod 130 itself. To perform this second series of checks no tools are screwed into the holes 100-103.

4 1 1 - 11 The ball 46 of the feeler 40 is able to touch the ball 131 in a manner which will be described more clearly hereinbelow in such a way as to detect an assembly of data utilised by the method of the present invention to verify the position of the reference block 32 with respect to the centre of the rotating table 17.

The six dimensional checks of the first series will now be described with particular reference to Figures from 2 to 7.

In Figure 2 the first dimensional check is shown in which the feeler 40 touching in succession the balls 111 and 118 detects the X, Y and Z coordinates of the centres of these I'S balls. In this dimensional checking stage the table 17 is rotated through 00 with respect to the reference cube 32.

In Figure 3 the second dimensional check is illustrated in which the table 17 is rotated through 1800 with respect to the position of Figure 2, and the feeler 40 by successively touching the balls 111 and 118 detects the X. Y and Z coordinates of the centres of these balls.

In this way, by means of the points obtained in the first and second dimensional checks described above, a first and a second distance between the ball 111 and the ball 118 is calculated in the angular positions illustrated in Figures c 2 and 3 respectively. Therefore two distance vectors which extend between the ball 111 and the ball 118 and through which the X, Z plane passes are virtually defined. If the geometric configuration of the measurement machine has not been modified in any way these distance vectors will have the same length.

The third dimensional cheek is illustrated in Figure 4, in which the feeler 40 by touching in succession the balls 111 and 121 detects the X, Y and Z-coordinates of the centres of the balls. In this dimensional check stage the table 17 is rotated through 00 with respect to the reference cube 32 and therefore is turned to the same relative position as that illustrated in Figure 2.

The fourth dimensional check is illustrated in Figure 5, in which the feeler 40 by touching the balls 111 and 121 detects the X, Y and Z coordinates of the centres of these balls when the table 17 is rotated through 900 with respect to the reference cube 32.

In this way, by means of the points obtained in the third and fourth dimensional check described above,. a first and a second distance between the ball 111 and the ball 121 is calculated in the angular positions illustrated in Figure 4 and Figure 5 respectively. In this way two distance vectors which extend between the ball ill and the ball 121 1 1 1 - 13 are virtually defined, through which the X, Y plane passes. If the geometric configuration of the measurement machine has not been modified in any way these distance vectors will be orthogonal and will have the same length.

The fifth dimensional check is illustrated in Figure 6, in which the feeler 40 by touching the balls 118 and 126 detects the X, Y and Z coordinates of the centres of these balls when the table 17 is rotated through 450 with respect to the reference cube 32.

In Figure 7 the sixth dimensional check is illustrated in which the feeler 40 by touching the balls 118 and 126 detects the X, Y and Z coordinates of the centres of these balls when the table 17 is rotated through 2250 with respect to the reference cube 32.

In this way, by means of the points obtained in the fifth and sixth dimensional check described above, a first and a second distance between the ball 118 and the ball 126 is calculated in the angular position illustrated in Figure 6 and in Figure 7 respectively. By this means two distance vectors which extend between the ball 118 and the ball 126 are virtually defined, through which the YZ plane passes.

If the geometric configuration of the measurement machine has not been modified in any way these distance vectors will have the same length.

With particular reference to Figures from 8 to 11 the four dimensional checks of the second series will now be described.

In Figure 8 the seventh dimensional check is illustrated, in which the feeler 40 by touching the ball 131 detects the X, Y and Z coordinates of the centre of the ball when the table 17 is rotated through 00 with respect to the reference cube 32 in a position defined by a notch 80.

In Figure 9 the eighth dimensional check is illustrated, in which the feeler 40 detects the X, Y and Z coordinates of the centre of the ball 131 when the table 17 is rotated through 900 with respect to the reference cube 32.

Figure 10 illustrates the ninth dimensional check in which the feeler 40 by touching the ball 131 detects the X, Y and Z coordinates of the centre of the ball when the table 17 is rotated through 1800 with respect to the reference cube 32.

Figure 11 illustrates the tenth dimensional check in which the feeler 40 by touching the ball 131 detects the X, Y and Z coordinates of the centre of the ball when the table 17 is rotated through 2700 with respect to the reference cube 32. If the geometric configuration of the measurement machine has not changed with respect to its 1 1 i i 1 z j 1 c 1 initial calibration configuration the four groups of X, Y and Z coordinates acquired in the four above-mentioned dimensional checks will be equal to one another.

In Figure 12 there is shown a flow chart of the method according to the present invention by means of which the said checking operations in the measurement machine 1 are physically achieved, and by which the recalibration data are detected and consequent modifications to the original calibration data are made if necessary. This method further includes a series of control signals which are, for example, conveniently memorised on a magnetic support and can be read by the unit 7 and loaded into the memory of the electronic unit 2 thereby controlling and making possible all the checking stages described above for achieving self- correction of the calibration data.

Initially the process starts at a block 200 in which the operator is asked, by displaying on the video 5, if he wishes to check and update the calibration data of the system 1. The operator responds to this request utilising the keyboard 6, and if he does not wish to proceedwith the check the system exits from the programme from block 200, whilst if he does wish to proceed the system moves from block 200 to a block 201.

In block 201 the operator is asked if he wishes to 1 16 - commence the first series of dimensional checks to detect the orthogonality of the component parts along the three cartesian axes X, Y and Z of the machine 1 and to proceed to possible self-correction.

In the affirmative case the s ystem passes from block 201 to block 204, and in the negative case it passes to block 208. In block 204 the operator is asked, and the request is presented on the video 5, if the recalibration data obtained by means of the first series of dimensional checks described above are already present in memory. In the affirmative case it passes from block 204 to a block 210 and in the negative case it passes to a block 205, which will be described in more detail hereinbelow, which imparts to the machine 1 the commands to perform in the order and in the manner described above. with reference to Figures from 2 to 7, and in an automatic way, the six dimensional checking operations for testing the orthogonality of the axes and proceed to the reception and saving of the points X, Y and Z of the centres of the balls, and definition of the distance vectors between the balls detected in the various angular positions of the table 17.

From the block 205 it passes to a block 210 which provides for recall from the memory of the X, Y and Z recalibration points and the old calibration data, and i 1 j 17 - reconstruction of a new sequence of calibration data in which the data relating to the orthogonality of the three cartesian axes X, Y and Z are modified and corrected on the basis of the points measured in the block 205.

From block 210 it passes to a block 212 in which it asks if the operator wishes to proceed to a second selfcorrection stage. In the negative case it exits from the programme from block 212 and in the affirmative case from 10 block 212 it passes to block 208.

In block 208 the operator is asked if he wishes to commence the second series of dimensional checks to detect if the position of the reference block 32 has varied with respect to its initial position, and if he wishes to proceed to possible self-correction of the respective calibration data.

In the affirmative case, from block 208 it passes to a block 220 whilst in the negative case it exits from the programme.

In block 220 there is requested, and this request is presented on the video 5 to the operator, if the re- calibration data obtained by means of the second series of dimensional checks described above with reference to Figures from 8 to 11, are already present in memory. In 1 C - is - the affirmative case it passes from block 220 to a block 222 whilst in the negative case it passes to a block 221 (which will be described in more detail hereinbelow) which imparts to the machine 1 the commands to effect, automatically in the order and in the manner described above, the four operations for dimensional checking of the position of the cube 32, and provides for the reception and saving of the points X, Y and Z defining the centre of the ball 131 detected in these checking operations and constituting the re-calibration data.

From block 221 it passes to a block 222 which provides for recall from the memory of the re-calibration points X, Y, Z, and the old calibration data, and reconstructs a new sequence of calibration data in which the data concerning the position of the cube 32 are modified and corrected on the basis of the points measured in the block 221.

Prom the block 222 it passes to a block 230 in which it asks if the operator desires to proceed to a second selfcorrection stage or if self-correction is to be interrupted. In the negative case it exits from the programme from block 230 and in the affirmative case it returns from block 230 to block 201.

The block 205 of Figure 12 is now described in detail with reference to Figure 13. In particular, the block 205 4 1 1 C. - 19 - comprises a block 250 which controls the performance of the first, second, third, fourth, fifth and sixth measurement stages in the manner described above with reference to Figures from 2 to 7, and provides for the acquisition of the cartesian coordinates (X, Y, Z) of the centres of the balls Ill, 118, 121 and 126 and for calculation of the said distance vectors for each of the five measurement positions corresponding to the angular positions of the plate 50 at 00, 450, 9001 18001 22501 lo with respect to the reference cube 32.

From the block 250 it passes to a block 253 which provides for memorisation of the points detected by means of the block 250.

From block 253 it passes to a block 256 which arranges for presentation on the video 5 and sends to the printer 8 the points detected by means of the block 250.

In Figure 13 there is also shown the block 210 of Figure 12 which comprises a block 260 which provides for searching for the old calibration data and loading of this into the central unit 3.

From block 260 it passes to a block 263 in which the points detected by means of block 250 are treated with mathematical algorithms and are compared with the "old" - 20 calibration data for the production of a new sequence of calibration data as described hereinbelow.

The block 263 effects calculations by means of the coordinates of points (in this case the centres of the balls) in two different reference systems: a first reference system in which the three axes X, Y and Z li e at an angle of 900 to one another, and a second reference system in which the three axes V, Y' and V form an angle different from 900. Therefore the respective axes X and Xl, Y and V, and Z and V are offset from one another by three angles respectively called alpha, beta and gamma.

The two different reference systems are therefore linked by a vector equation of the type:

xl X Y1 Matrix X Y Z1 1 1 1 1 z where the matrix contains functions of the unknown values of the angles alpha, beta and gamma.

By means of the knowledge of the distance vectors defined in the block 205 and by means of the equation (1) it is possible to implement a method which makes it possible to get back to the angles alpha, beta and gamma and therefore to know in a complete way the relationship (1). In this 1 1 i 1 way, once the relationship (1) is known with precision the passage from one reference system to the other is uniquely determined.

From the block 263 it passes to a block 266 which provides for recording the new sequence of calibration data in the memory of the electronic unit 2 and possibly on a magn etic support.

The block 221 of Figure 12 will now be described in detail with reference to Figure 14. In particular, the block 221 comprises a block 309 which controls the performance of the seventh, eighth, ninth and tenth measurement phase in the manner described above, and which provides for the acquisition of the cartesian coordinates X, Y and Z of the centre of the ball 131 for each of the four measurement phases corresponding to the angular positions of the plate 50 at 00, 900, 1800 and 2700 with respect to the reference element 32. From the block 309 it passes to a block 310 which provides for memorisation of the points detected by means of the block 309. From the block 310 it passes to a block 311 which arranges for presentation on the video 5 and sends to the printer 8 the points detected by means of the block 309.

In Figure 14 there is also illustrated in detail the block 222 which includes a block 316 which provides for C..

searching for the old calibration data and loading of this into the central unit 3.

From block 316 it passes to a block 320 in which the points detected by means of the block 309 are treated with mathematical algorithms and are compared with the "old" calibration data for the production of a new sequence of calibration data as described hereinbelow.

In particular the position of the table 17 with respect to the reference cube 32 is determined by a vector V which expresses the distance from the centre of the table 17 to the centre of the cube 32. This vector V is determined during the calibration phase and varies if the position of the cube 32 varies, transforming into a vector V'.

During the second series of dimensional checks there is detected an assembly of data which are a function of how V has varied with respect to V', that is to say a function of the type:

f(V'-V) (2) It is therefore possible to seek an application g which is the inverse of f which, applied to it, will give the vector variation "delta" that is to say:

g(f(V'-V)) = delta(vector) (3) where:

V' = V + delta(vector) (4) 4 In this way it is possible to get back to the vector V', initially unknown, and therefore to know in a precise manner the new position of the reference cube 32. The knowledge of equation (4) finally makes it possible to refer back to the initial reference system data which has been measured in a machine in which the cube 32 has changed position.

Given that the distance between the origin of the reference system and the centre of the table 17 is expressible by means of the vector sum of a distance vector from the axis origin to the cube 32 (calculated in an automatic manner) and the cube 32-table centre 17 vector, the knowledge of equation (4) also makes it possible to know in a precise manner theposition of the table 17 with respect to the origin of the frame of reference.

From block 320 it then passes to a block 323 which provides for storing the new sequence of calibration data in the memory of the electronic unit 2, and possibly on a magnetic support.

From what has been described above it is evident that the method of the present invention provides for the recalibration of the machine in an automatic manner and without the use of specialist personnel.

1 - 24 Finally, it is clear that modifications and variations can be introduced to the method of the present invention without departing from the protective ambit thereof. In particular, the method of the present invention can be advantageously utilised for the self-correction of calibration data for measurement machines having rotating tables of different types from that described. The method can further include a series of dimensional checks achieved in a different manner from that described above, jo possibly controlled by an operator rather than being effected automatically, and achieved with abutment means having a different form and dimensions from those described.

1 1 Finally, the measurement machine 1 can have a different structure from that described, including parts having different functions and forms. In particular, the reference cube 32 can be replaced by a ball having the same function. 20 i 1 ( ' JMel90291 - 25 -

Claims (18)

1. A method for the metrological checking and selfcorrection of geometric detection errors in a measurement machine (1), particularly a measurement machine provided with a rotating table (17); the said measuring machine (1) being provided with a feeler element and being control led by an electronic unit (2); initial calibration data of the said measuring machine (1) being resident in the said electronic unit; the said calibration data being characteristic of an initial geometric configuration of the said measurement machine (1), characterised by the fact that it comprises at least one first dimensional checking stage for determining if the geometric configuration of the measurement machine (1) has changed and for collection of a sequence of re-calibration data, and a second phase of self-adjustment and modification of the calibration data on the basis of the said recalibration data to adapt the said calibration data to the changed geometric characteristics of the said machine (1).

2. A method according to Claim 1, in which the said feeler element (40)is movable in three orthogonal directions defining a set of three cartesian axesi characterised by the fact that the said first phase includes at least a first step for checking the orthogonality of the said cartesian axes, and for i i i l j i i ^190291 collection of a first sequence of re-calibration data for the said set of three axes, and in that the said second stage includes a first self- adjustment and modification stage in which the said initial calibration data is adapted on the basis of the said first sequence, and the orthogonality between the said axes is ensured for each measurement with the said machine (1).

3. A method according to Claim 1 or Claim 2, in which the said measurement machine includes a reference element (32) for the said rotating table (17), characterised in that the said first stage includes at least a second position verification step for determining the position of the said reference element (32) with respect to the said rotating table (17), in which a second sequence of re-calibration data relating to the position of the said reference element is collected, and in that the said second stage includes a second self-adjustment and initial calibration data modification step in which the said calibration data are modified on the basis of the said second sequence.

4. A method according to Claim 3, characterised in that the said reference element (32) is a cube.

5. A method according to Claim 3, characterised in that the said reference element is a ball.

7 i i i i 4 1. 1 # C ' iMe190291 - 27

6. A method according to any of Claims from 3 to 51 characterised in 1 that in the said second checking step the relative position of the said reference element (32) with respect to a set of three cartesian axes is tested; the said second sequence of calibration data modifying the said initial calibration data and ensuring for each measurement the exact knowledge of the position of the said set of three axes with respect to the said reference element.

7. A method according to any preceding Claim, characterised in that it includes a body (110, 120, 125, 117, 130) with reference surfaces (111, 121, 126l 118, 131) adapted to cooperate with the said feeler element (40) for performing the said first dimensional checking step.

8. A method according to Claim 7. characterised in that the said body (110, 120, 125, 117, 130) is stably connected in use to a plate (50) fixed to the said rotating table (17).

9. A method according to Claim 8, characterised in that the said reference surface of the said body cooperates with the said feeler element (40) at successive instants providing a plurality of dimensional checks; each dimensional check being defined by a determined angular a t ( 3 JMe190291 - 28 - position of the said plate (50) with respect to the said reference element (32).

10. A method according to any preceding Claim, characterised in that it includes several bodies (110, 120, 125, 117, 130) with reference surfaces (111, 121, 126, 118, 131) constituting a tool; the said bodies being adapted to cooperate with the said feeler element (40) for the performance of the said dimensional checking step.

11. A method according to Claim 10, characterised in that the said tool comprises a series of rods (110, 117, 120, 125, 130) adapted to be secured at their lower ends to the said plate (50); each of the said rods (110, 117, 120, 125, 130) terminating at its upper end with a respective ball (111, 121, 118, 126, 131) adapted to cooperate with the said feeler element (40) for the performance of the said dimensional checking step.

12. A method according to Claim 11, characterised in that the said tool includes at least one spacer element (115) adapted to be interposed between one (116) of the said rods and the said table (50).

13. A method according to Claim 11 or Claim 12, characterised in. that the said balls (111, 121, 1 -. ' JMe190291 - 29 118, 1261 131) cooperate with the said feeler element (40) in succession in performing a plurality of dimensional cheeks; each dimensional check being defined by a determined ahgular position of the said plate (50) with 5 respect to the said reference cube (32); the said balls (1111 121, 1181 126, 131) being disposed respectively in different positions on the said plate (50) and at different heights to define together directions along planes defined by the said three axes in the said first orthogonality check on the said cartesian axes, and on the centre of the said table (17) to define a checking position in the said second checking step in which the relative position of the said reference element (32) with respect to the said rotating table (17) is Lested.

is

14. A method according to any preceding Claim, characterised in that the said first checking step and the said second self-adjusting and calibration data modification step are controlled and achieved automatically by means of the said electronic unit (2).

15. A measurement machine of the rotating table type, characterised in that it includes a system for metrological testing and self -correction of -geometric detection errors according to the netbod of any preceding ClaIrn.

16. Equipment for metrological testing and self- A (1 JMe190291 correction of geometric detection errors of a measurement machine, particularly a measurement machine provided with a rotating table, according to any of Claims from 1 to 14, characterised in that it comprises at least one body (110, 120, 125, 117, 130) with a reference surface (111, 121, 126, 118, 131) adapted to cooperate with a feeler element (40) of the said measurement machine (1) for the performance of the said first dimensional checking step.

17. A method for metrological testing and self-correction of geometric detection errors in a measurement machine, substantially as herein described with reference to the accompanying drawings.

18. A measurement machine arranged and adapted to operate substantially as described with reference to and as shown in the accompanying drawings.

Published 1991 at The Patent Office. Concept House. Cardiff Road. Newport. Gwent NP9 IRH. Further copies may be obtained from Sales Branch. Unit 6. Nine Mile Point. Cumfelinfach. Cross Keys. Newport. NP1 7HZ. Printed by Multiplex techniques ltd. St Mary Cray. Kent.

1 i j 1 1 i j i 1