GB2253056A

GB2253056A - Method of determining the geometric dimensions of a testpiece by means of gauging with mechanical contact

Info

Publication number: GB2253056A
Application number: GB9201652A
Authority: GB
Inventors: Weiss Wolfgang
Original assignee: Carl Zeiss Jena GmbH
Current assignee: Jenoptik AG
Priority date: 1991-02-01
Filing date: 1992-01-27
Publication date: 1992-08-26
Anticipated expiration: 2012-01-27
Also published as: JPH0587501A; DE4103060A1; GB2253056B; DE4103060C2; CH683371A5; GB9201652D0

Abstract

Method of determining the geometric dimensions of a testpiece by means of gauging with mechanical contact to facilitate the comprehensive compensation of non-linear and linear elastic deformations in gauging with mechanical contact, so that an increased accuracy can be achieved in the determination of geometric dimensions of a testpiece. Two length measurement values are determined at the particular testpiece measurement point one after the other with two different measuring forces and including a calibration value, the actual testpiece measurement value being determined from the difference between these length measurement values after multiplication with suitable factors.

Description

2253056

DESCRIPTION

Method of determining the Scometric dimensions of a testpiece by means of SauSing with mechanical contact The invention serves for precision gauging with mechanical contact in high-resolution distance measurement. It can be used wherever precision measurements of length require compensation for elastic deformations. This applies in particular to instruments for fine measurement or coordinate measuring technology. Possible applications arise for example from the measuring tasks of ultraprecision processing in determining the dimensions of flat, spherical, cylindrical and other surfaces. Other advantageous applications may be seen Inter alla in the final measurement with regard to dimension calibration and monitoring of long-term stability as well as in the measurement of the diameter of precision balls.

The elastic deformations which occur during gauging with mechanical contact hinder the requirements for precision in high-resolution distance measurement to an increasing extent. Elastic deformations resulting from non-linear Hertzian ellipticity operations or linear gauge pressure loads inevitably lead to disruptive dimensional discrepancies. The cause of the deformation is the measuring force as regards it magnitude andconstancy. Depending upon the task, even a reduction of force to a few ponds can be associated with unacceptably hilgh residual deformations. On the other hand, it is also to be expected that a measuring force approaching zero is not al-ways advantageous for reasons of construction. function-'and application. Therefore there has for a long time been a requirement for suitable solutions as regards the reductio-nof such deformation influences.

Thus for example in the determination of dimensions of workpiece surfaces produced by ultraprecision processing peak requirements as regards measurement are predetermined for resolution and accuracy In the range between 1 lim and 1 nm which In turn only permit a few hundredths tm uncertainty for the actual gauging operation (Feingeratetechulk, Vol. 38, 1989, issue 1, p. 2 and Feingerc%tetechnik, Vol. 39, 1990, Issue 7, p. 302-304). With the choice of contact gauging In this case or with the setting of similar tasks, countermeasures regarding the disruptive deformation Influences are necessary in order to meet the high requirements for accuracy. In this case measures against the flattening deformations occurring at the gauging point are rendered particularly difficult by the variety of testpieces which are generally present, varying as regards shapes, dimensions and materials.

In order to reduce the deformation influence. the following methods which define the prior art present themselves:

Measurement against a standard for comparison:

In order thoroughly to compensate for deformations, refer- ence to standards for comparison is necessary. Problems remain here in the costly preparation of a number of comparison normals depending upon shape, d. Imension and material, normal dimensional discrepancies. the uncertainty of the measurement connection which always has to be freshly carried out, and measurement errors resulting from differences in hardness of technological origin or material fluctuations, also Inhomogeneities, between the normal and the testpiece. Reference is made by way of example to the final measurement calibration or monitoring of long-term stability by means of a final measurement test position (Jenaer Rundschau, Vol. 30, 1985. issue 2. p. 84r-87>. In this final measurement comparison process. in spite of high electronic measurement value resolution. the stated shortcomings and measurement errors. resulting Inter alla from differences In hardness in the normal comparison. cannot be 1 entirely excluded.

Computer correction of elastic deformations:

The computer correction of deformations caused by flattenj ing is basically possible using the flattening formulae resulting from the Hertzian ellipticity theory (H. Zill. Messen und Lehren im Maschinenbau und In der Feingeratetechnik, VEB Verlag Technik Berlin, 2nd edition, Berlin 1972, p. 68-71). These give the complicated relations of measuring force. surface geometry, material and flattening. In the case of contact "]ball with the diameter d against cylinder.with the diameter W, the flattening at a measuring force F Is for example:

116- 7 F' Of particular importance here are the elastic material constants E', W, m'. W' of the gauging element and the testpiece. In accurate investigation It is clear that with precision measurements of the highest accuracy a computer correction of the flattening is very time-consuming, encumbered by possible erroneous inputs and constants and other characteristic values and is only possible to a limited extent because of deviations of the material constants as regards their accuracy of determination or material fluctuations. In this connection it should also be observed that the possibility-for measurement or correction In the case of new materials is only given after the material constants have been determined and this can give rise to shutdown periods which reduce productivity. Furthermore, the influence of material inhomogeneities Is particularly critical for the correction process, since It cannot be determined by computer.

1 - t - Measuring force extrapolation to zero: This method takes place for correction of gauge or workpiece deformations. By switching in different measuring forces F 1 and F 2 of a measuring gauging system a bending characteristic for the gauge or the workpiece is determined which Is used for calculation to the measurement value m 0 with the measuring force zero. This produces:

- 6WZ 2. - M.0 = in which m 1 and M2 respectively are the measurement values determined with the measuring forces F 1 and F 2 respectively. The route of measuring force extrapolation to zero which is used In special cases does indeed basically offer a reduction In error of elastic deformations occurring, but because of the linear back-calculation to the measurement value with the measuring force zero It precludes a complete coifipensation for non-linear Hertzian flattening defornkations. Therefore in a consideration of the flattening A 1 and A 2 where errors are critical there remains a systematic proportion of error d mo of the magnitude (iq - P ef 0 z 1) In which A 1 and A 2 respectively are the flattenings caused by the forces F 1 and F 2 respectively. The disadvantage of this reduction method is to be seen in the fact that the elastic deformations are not determined in their entirety.

One example of an application of this method is the determination of dimension of plastic parts which are not inherently stable with the aid of multi-co-ordinate measuring 1 machines (Kunststoffe, Vol. 75, 1985. issue 11. p. 824828). Since here bending of the workpiece caused by the measuring force is dominant, the flattening influence which is still present can often remain unconsidered after testcritical decision. By contrast, precision measurements of the highest accuracy require the comprehensive inclusion of operating deformation components and therefore require new solutions for compensation for the deformation when alternatives to the prior art are called for in the case of peak requirements.

It Is the aim of the invention to overcome the aforementioned shortcomings of the solutions which are known from the prior art and to achieve a compensation which covers non-linear and linear elastic deformations in contact gauging in the ultraprecision range.

The object of the Invention Is that In a method for contact gauging, simultaneously with the compensation of linear gauge-related elastic deformations. a complete compensation of non-linear flattening deformations should be achieved with great accuracy, so that it should be possible to carry out the method for testpieces of differing geometry and material composition.

According to the invention this object is achieved with a method for the determination of geometric dimensions of a testpiece by means of gauging with mechanical contact, wherein by means of a calibration operation preceding the measurement of the testpiece non-linear Hertzian deformations associated with flattening and linear gauge-related elastic deformations are compensated, and In this operation the calibration object is brought onto a measuring table reference surface and, for calibration, two different measuring forces F,. F 2 are used one after the other and, -by means of a distance measuring system coupled to a gauge, f two length measurement values M,, M 2 are determined at the same calibration point on the calibration object, the length measurement value M 1 being determined with the measuring force F 1 and the length measurement value M 2 with measuring force F 2' in which the deformation calibration measurement K 2 Is formed from the length measurement values M,, M 2 and the ratio of the measuring forces F, . F 2' according to the equation K ------ with a F > F 2 (a. - -1 l _) 2 1 in which the calibration object is gauged at the same cal Ibration point with the measuring forCe F, and the distance measurement system is calibrated so that Instead of the length measurement value M 1 it displays the length measurement value K=K 1 + K 2, in which K 1 Is a defined length calibration measurement which corresponds to the geometric dimension of the calibration object at the calibration point and K 2 is the deformation calibration measurement.

in which the testplece is brought onto the measuring table reference surface and for determination of the geometric dimensions of the testpiece at the respective testpiece measuring point i using In succession two different measurIng forces F11. F'9 with the same measuring force ratio as in the calibration operation, two length measurewnt values M 21+l' M 21+2 are determined, and in which thereafter the actual testpiece measurement value x 1 is formed in each case from the determined length measurement values M 21+l' M,,+, according to the equation t 7 - X wi th i = 1, 2. 3, In which the measurement values M 21+1 have been determined with the measuring force F' 1 and the measurement values M 21+2 are determined with the measuring force F' 2 With the calibration operation which is preferred for the testplece measurement and sets the distance measuring SystEffi, the calibration value is formed in such a way that in the measurement of the testpiece It leads, simultaneously and irrespective of material, to complete compensation of the deformation components referred to in the object of the Invention. In this case the calibration operation does not have to be repeated before each testpiece measurement, but the distance measuring system used is to be set once with the calibration to deformation compensation. This calibration will be carried out when the instruments of the distance measuring system are switched on and If necessary will be repeated periodically to ensure measurements during a long period of use or in the case of extreme requirements.

A particularly advantageous processing of the measurement values Is produced for the equation / - % Z Itt's.t In which the corresponding measuring force ratio I- - F = 2.8 fl 1-, 1 reliably meets the requirement for highly precise deformation compensation with a margin of the measuring force ratio of:tO.l to:tO.2.

It is advantageous if the length calibration measurement KjO is coordinated with the deformation calibration measurement K 2 formed on the calibration object within the calibration value K for testpiece comparative measurements to be determined according to the equation K = K 1 + K 2 It is also advantageous If the deformation calibration measurement K 2 of the calibration value K to be determined according to the equation K = K 1 + K 2 is formed in the absence of a defined calibration object with the length calibration measurement Kl=0 on the measur Ing table reference surface which serves as the calibration object.

It is also advantageous that the deformation calibration measurement K 2 of the calibration value K to be determined according to the equation is formed in the case of specially testpiece-related coordinate measurements with the length calibration measurement K1=O on a testpiece reference surface which serves as the calibration object.

Furthermore, favourable conditions are produced in the calibration operation if the zeroing of the distance measuring system which is necessary for forming the testpiece measurement value is combined with the consideration of the deformation in such a way that at the same time a disruptive shift in the measurement value, which can be contained arbitrarily in the distance measuring system t before the beginning of all measurements, is absent.

Further significant advantages are:

highest basic accuracy of the gauging by maximum elimination of elastic deformation effects even in testpieces made from plastic where the material is critical, independence of the method from the knowledge of elastic material constants, negligible influence on the method of material inhomogeneities, particular suitability of the method for precision test parts made from compound materials, since due to the method there Is no dependence upon dimension in the material differences, when a ball gauging element is used, independence of the deformation compensation from the inclination of the testpiece gauging surface, i.e. from the difference between gauging and force direction, the preparation of a number of standards for comparison Is not necessary, and thus an extensive monitoring of standards is also unnecessary, so that the method can be carried out rationally as regards costs and time.

high precision is ensured when a number of standards for comparison is dispensed with by not putting in standard deviations and critical producibility of connection, and the increased measuring forces permitted by the method make possib-le a better adaptation to disruptive limiting conditions of gauging, such as for example vibrations. shocks, cleanliness and air cushion, and/or to the influence of the testpiece surface. such as residual roughness, as regards possible contact disruptions.

The essence of the invention will be explained in greater detail with the aid of an embodiment which is shown in the drawings, by way of example only, and has vertical gauging and F 1 1 and F 2 - F' 2' The Possible applications are not limited to this example and also include horizontal gauging.

In the drawings:

Figure I shows schematically the calibration operation which takes place before testpiece height measurement with vertical gauging, and Figure 2 shows schematically a testplece height measurement with vertical gauging.

The basic measurement set-up for the calibration operation which precedes a testpiece height measurement with vertical gauging is shown in Figure 1. It comprises a measuring table 1 with a measuring table reference surface 2, a calibration object 3 standing on the measuring table reference surface 2 and having a defined length calibration measurement K at the calibration point 7, and a measuring gauge 4 fixed to the frame of the measuring table 1 and having the appertaining gauge pin 5 and its gauge ball tip 6. The calibration operation is performed in such a way that in a first step the gauge pin 5 which is placed on the calibration point 7 of the calibration object 3 is loaded with a measuring force F 1 Elastic deformations at the actual f 1 - 11 contact point between the calibration object 3 and the gauge ball tip 6. that is to say the calibration point 7, and for example on the gauge pin 5 are associated with the loading operation. At the calibration point 7 these deformations are non-linear Hertzian flattening deformations and on the gauge pin 5 they are linear gauge pin deformations. Consequently a measurement value M 1 which is composed of the defined length calibration measurement K,, an arbitrary measurement value shift Ah, the flattening deformation - A 1 - and the gauge pin deformation - L according to M 1 =K 1 + Ah - A 1 - L is passed from the measuring gauge 4 to a distance measuring system which is not shown in Figure 1.

In a second step, retaining the co-ordination of the gauge pin 5 with the calibration point 7 of the calibration object 3, the measuring force F is switched over to a measuring force F 2 Both measuring forces stand In the ratio to one another F 2: F, = b with F 2 > F 1 Now a measurement value F4 which Is composed of the defined length calibration measurement K,, the unchanged measurement value shift Ah, the flattening deformation A1 and the gauge pin deformation - 1. L accord- 1r, ing to 2 = K, + &h is passed from the measuring gauge 4 to the distance measuring system..

Both measurement values M 1 and M 2 are falsified, apart from thedisruptive shift Ah in the measurement value. by elastic deformation of differing order.

In a third step a deformation calibration measurement K 2 is cal cul ated according to 1, - P11f K2 = VW -7 ú 2 -,1) 4,7 4 and combined with the defined length calibration measure ment K 1 to a calibration value K according to K = K 1 + K 2 In a fourth step this calibration value K with K = K., 4- N, - 114 Z - -I) Is exchanged for the measurement value M 1 newly appearing in the distance measuring system after withdrawal of the measuring force F 2 and switching back In of the measuring force F,, with unchanged co-ordination of the gauge pin 5 with the calibration point 7 of the calibration object 3. Thus a deformation amount of the magnitude Vz 1) - 2- is obtained which contradicts the actual gauging situation and inevitably stays with the following testpiece measurements. At the same time the measurement value shift Ah still contained in the measurement values M 1 and M 2 drops out of the further measurement value formation and no longer influences the measuringof testpieces.

The actual measuring of the testpiece which Is shown in Figure 2 is performed in the same way as in the first and second calibration steps described above, but with the preconditions that the distance measuring system for testpiece measurements is set with deformation compensation and the gauging operation takes place on the top surface 10 of the testpiece 9 which determines the height x of the testpiece. With the same measuring force ratio b the flattening deformations - A and 2 f(T11- - X- occur Irrespective of the testpiece material at the actual contact point 8 between the testpiece 9 and the gauge ball tip 6. Consequently. with the measuring force F 1 a measurement value M 3 composed of - testpiece height x, - the flattening deformation -A 2 - the gauge pin deformation -L, - and the deformation amount superimposed by the calibration L ( V(r- n-rll- to M3 5 - A2 - --- (- - - 1) - L X z n ( y21-74 1) is obtained in the distance measuring system.

By contrast. with the measuring force F 2 val ue M 4 composed of testpiece height X.

- the flattening deformation the gauge pin deformation a measurement L 1 and the superimposed deformation amount -"I L L to (yntr7-1) ---7 M4 =Xrp, r :rl- - % A2 - -,) is obtained in the distance measuring system.

The actual testplece measurement value x is determined by computer evaluation of the two measurement values M3 and M from the equation -Hill X from which the deformation values contained in the individual measurement values M 3 and M 4 drop out. Further testpiece measurements are possible with renewed calibration and lead, irrespective of material and irrespective of technologically necessitated hardness values and of material inhomogeneities, to measurement results which are correspondingly free of deformation.

In the illustrated calibration operation the zeroing of the distance measuring system which is necessary for forming the testpiece measurement value is included in such a way that the actual length measurement value zero for immediate testpiece measurements is co-ordinated with the measuring table reference surface 2. BY calibration variation with K 1 =0 in the calculation of the calibration value K, a virtual reference surface running through the calibration point 7 parallel to the measuring table reference surface z 1 in the embodiment can be formed with the actual length measurement value zero, which is particularly suitable for measurements of differences with extreme requirements of accuracy.

Y

Claims

1. Method for the determination of geometric dimensions of a testpiece by means of gauging with mechanical contact,,whereint by means of a calibration operation preceding the measurement of the testpiece, nonlinear Hertzian deform&tions associated with flattening and linear gaugerelated elastic deformations are compensated, and in this operation the calibration object is brought onto a measuring table reference surface and, for calibration, two different measuring forces F,, F

2 are used one after the other and, by means of a distance measuring system coupled to a gauge, two length measurement values M1. M 2 are determined at the same calibration point on the calibration object, the length measurement value M 1 being determined with the measuring force F I and the length measurement value M 2 with measuring force F 2" in which the deformation calibration measurement K 2 is formed from the length measurement values M,, M 2 and the ratio of the measuring forces F, F.. according to the equation K 2 = -11-- fil- (a- -.1) with a F 2 > F 1 in which the calibration object is gauged at the same cal ibration point with the measuring force F 1 and the distance measurement system is calibrated so that instead of the length measurement value M 1 it displays the length measurement value K=K 1 + %. in which K 1 is a defined length calibration measurement which corresponds to the geometric dimension of the calibration object at the calibration point and K 2 is the deformation calibration measurement.

in which the testpiece is brought onto the measuring table reference surface and for determination of the geometric dimensions of the testpiece at the respective testpiece measuring point i using in succession two different measurIng forces F',, F' 2 with the same measuring force ratio as in the calibration operation two length measurement values m 2i+l' m 21+2 are determined, and in which thereafter the actual testpiece measurement value x 1 is formed in each case from the determined length measurement values M 21+l' m 21+2 according to the equation A4 X wi th 1 = 1, 2, 3,...

in which the measurement values M 21+1 have been determined with the measuring force F 1 and the measurement values m 21+2 are determined with the measuring force F' 2 2. Method as claimed in claim 1, Wherein the measuring forces F,. F 2' F' 1 and F' 2 are dimensioned according to the equation (L _) 21 . 1 (-1-) 2. = 8 TI

3. Method as claimed in claim 1 or claims 1 and 2.

wherein the length calibration measurement K, =0 is co-ordinated with the defonliation calibration measurement (K) formed on the calibration object within the calibration value (K) to be determined according to the equation k K = K 1 + K2

4. Method as claimed in claim 1 or claims 1 and 2.

wherein the deformation calibration measurement (K 2) of the calibration value (K) to be determined accordlug to the equation K = K 1 + K 2 Is formed with.the length calibration measurement Kl=0 on the measuring tabl c ref erence surface which serves as the calibration object.

5. Method as claimed in claim 1 or claims 1 and 2.

wherein the deformation calibration measurement (K 2) of the calibration value (K) to be determined according to the equation K = K 1 + K 2 is formed with the length calibration measurement Kl=0 on a testpiece reference surface which serves as the calibration object.

6. A method for the determination of geometric dimensions of a test piece, substantially as hereinbefore described with reference to the accompanying drawings.