The application of artefacts and lasers to performance verification of co-ordinate measuring machines

Transcription

1 The application of artefacts and lasers to performance verification of co-ordinate measuring machines A.D. ~ o ~ & ed.m.s. l ~lackshaw' l Faculty of Technology, Southampton Institute, UK 2 Intelligent Information Technology Ltd, UK Abstract In order to meet the ever increasing demands on product quality it is essential that the accuracy of Co-ordinate Measuring Machmes (CMMs) are regularly monitored and their stated performance levels are verified against accepted standards. Performance verification testing is currently carried out according to a number of international standards, of which the most widely used in Europe is IS , Part 2. This Standard specifies a wide range of mechanical artefacts for CMM performance verification and this can cause some confusion amongst CMM users. Also it is difficult to verify the performance of CMMs with working volumes in excess of one cubic metre using existing artefacts. In the machine tool industry performance verification testing using a laser interferometer is widely accepted and included in the appropriate standard (IS ). Also all CMM manufacturers now use the laser interferometer for performance verification, mainly during the build and on-site calibration of their machines. However, the laser interferometer has yet to be fully accepted for CMM performance verification and is not included in IS Hence the laser is not yet widely used by CMM users. The aim of this work was to define simplified performance verification procedures for CMMs of all sizes, types and configurations, using a laser interferometer and a new universal artefact. 1 Introduction to CMM testing methods Since the beginning of CMM technology, the question that has constantly arisen is "how can the user of a CMM test the accuracy specified and guaranteed by the supplier after the installation of the machine and later, during the years of

2 operation" [l]. Standardised testing methods are needed to examine CMM performance and the development of accurate and efficient techniques for checking CMMs has become a priority in recent years. There are a very large number of CMM configurations available and the working volume may also vary greatly from less than 0.5 cubic metres for small machines to as large as 100 cubic metres or more for large gantry machines. 1.1 Standards for CMM Verification In the early days of using co-ordinate measuring techniques, users and suppliers agreed on the application of machine tool test procedures in an adapted form for CMMs. Later some of the most important CMM manufacturers founded the CMMA (Co-ordinate Measuring Machine Manufacturers Association) and established the CMMA Accuracy Specifications, published in In recent years, several national standards have been introduced for inspecting CMMs. The British Standard Technical Committee was formed in 1984 and initially particular emphasis was placed on producing a standard for the performance verification of CMMs. The first two parts of BS 6808 [3] were published in 1987: Part 1, Co-ordinate Measuring Machines, Glossary of Terms and Part 2, Methods for Verifying Performance. Part 3, Code of Practice was published later in The ANSIiASME B M 141 is the American Standard, first published in 1985, with a more recent version appearing in Other standards used in Europe are the German, VDIIVDE 2617, Parts 1 to 5 ( ) [S] and the French, NF El (1986) [6]. The Japanese standard, JIS B7440 (71, was published in The International Standards Organisation (ISO), published it's standard in 1994, IS , Parts 1 and 2 for acceptance tests and interim machine checking of CMMs [g]. There are some differences between these published standards. In 1991 Knapp et a1 [l] reported that the standards vary as far as the importance of the various aspects of the test, including the environment. Peggs 191 in 1989 stated that the differences are mainly related to the extent of testing, i e. how many positions need to be chosen for the length standards and how many repeat measurements are required. These standards examine the volumetric performance of a CMM and all of them emphasise the measurement of artefacts located in specific positions within the CMM measuring zone [1,9]. A publication relating to some of the most popular CMM standards was published in 1996 by Abackerli et al. [10], which includes a comparison and a discussion about the differences between them. 1.2 Acceptance Tests, Interim Checks and Periodic Re-verification CMM perforn~ance verification testing may be classified according to the time scale on which they should be completed, e.g. acceptance tests. interim checks or periodic re-verification. In principle, an acceptance test is a matter of a contract between the CMM supplier and the user, Kunzmann et al., According to Trapet in , the acceptance test is usually carried out upon delivery, on the basis of a

3 contract betwcen the manufacturer and user which explicitly specifies the properties to be tested and the test method to be applied. CMM interim checks are the lowest level of test and are the sole responsibility of the owner. These tests should be easily and rapidly performed at short intervals. The purpose of the inspection performed by the user is to make a general statement on the metrological condition of the machine, which can be easily interpreted. The result is a passlfail decision, because the tests are not intended to analyse the error behaviour in detail. However, it is expected that every kind of error should affect the result in this 'overall test'. Kunzmann et al. Ill] suggested that the reference objects usually used for this type of test are step gauges, ball plates and space frames, which represent calibrated lengths. Periodic re-verification implies the determination of the complete volumetric error pattern for the CMM and requires the determination of 21 detailed parameters, which describe the errors of the co-ordinate system. This should be carried out on a regular basis, say annually. Peggs 1131 states that one procedure to estimate the magnitude of the geometrical errors may be to measure the departures from the correct geometry of all degrees of freedom of the machine using a laser interferometer, straight-edge, precision level, autocollimator, and other instruments. These methods are, however, very time consuming because many different experimental arrangements are needed to cover the principal error sources. Because these techniques are expensive and require skilled operators, Peggs 1131 considers the use of artefacts as an alternative calibration technique. A given artefact can be measured in a multiplicity of locations and orientations throughout the working volume of the machine to create an error map which provides detailed information on the variation of machine error in its working volume. CMMs may be calibrated using a wide range of mechanical artefacts which include gauge blocks, length bars, ball-ended plates, ball plates, hole plates, step gauges, straight edges, squares, space frames and many others. The wide range of mechanical artefacts currently in use were reviewed by Miguel 1141 in The current state of the art The most widely used performance verification standard in Europe is IS , which is based on artefact techniques and does not recognise the use of the laser interferometer. This standard specifies a wide range of acceptable artefacts, which can be confusing for the CMM user. Also CMMs with working volumes in excess of one cubic metre are not addressed in the standard and are difficult to calibrate using existing artefacts. In contrast parametric tests using a laser interferometer are widely used in the machine tool industry and all CMM users now use laser interferometers, mainly during the build and on-site servicing of their machines. 3 The project aims and objectives As a result of a feasibility study carried out by the authors for the European

4 Commission it was concluded that there was an urgent need to define simplified test procedures for performance verification of CMMs of all sizes, types and configurations. This should include recognition of the laser interferometer and the use of a new universal artefact for interim checks. The project objectives were as follows. To develop a general method of test for CMM performance verification using laser interferometer techniques and a novel universal artefact. To design and develop a low cost, novel, single, universal artefact for CMM interim checks, which will replace the wide range of artefacts currently specified. To design and develop enhanced software for use with the laser interferometer to calculate the volumetric accuracy for a range of CMM configurations. To apply the universal artefact as a "quick check" device for CMM performance verification in conjunction with the laser interferometer for complete parametric calibration. To address a wide range of CMM type and configuration. 4 Performance testing of conventional CMMs and articulated arm CMMs Extensive performance verification testing on a range of conventional and articulated arm type CMMs has been successfully completed. The performance of twenty different machines has been evaluated during the project. The performance verification tests were carried out using both laser and artefact techniques. The artefacts used were length bars, step gauges and the Renishaw Machine Checking Gauge and the laser tests were conducted according to the procedure detailed in the German standard, VDIIVDE The laser tests included linear positioning accuracy, angular data capture, straightness and squareness. In addition to the laser tests two mechanical tests were also completed, i.e. a roll check of the X and Y axes using an electronic or spirit level and a squareness test in XZ, YZ and XY axes using an engineer's square. The majority of machines tested were found to be outside the manufacturers specifications. An assessment of volumetric accuracy was made using both artefact techniques and the laser interferometer in conjunction with volumetric calibration (ESP) software. The ESP (Error Simulation Programme) software was originally developed for machine tool use by Postlethwaite and Ford [l51 and has been adapted for CMM use during this work. The software calculates the volumetric accuracy for each point in the machine working volume by taking the vector sum of the three axis errors. The value displayed is then the

5 largest absolute value that is calculated. Table 1 shows the errors obtained for eight different CMMs using both artefacts (according to IS ) and the laser interferometer together with the ESP software. Table 1: Volumetric error results for a selection of CMMs Machine Rolls Royce LIK S1 Type H H X axis mm Y axis mm Z axis mm Artefact L bars L bars IS Pm ESP Vol. Pin Eastman S1 B L bars St H - Horizontal arm B - Bridge G - Gantry A - Articulated arm L bars - Length bars St gauges - Step gauges Rmc - Renishaw machine checking gauge 5 The requirements for a new artefact Following the results obtained from the previous performance verification testing the requirements were identified for a new artefact to be used as an interim check device on CMMs of all sizes, types and configurations. The fundamental requirements for the artefact design were considered to be simple construction, low cost, ease of use, light weight, portable, low temperature coefficient, high rigidity and mechanically stable features. The specification for the artefact was agreed and is summarised in Table 2 below.

6 122 Laser tletroloa and Ilachrne Performonre Table 2: Universal artefact specifications Specification Measurement Size Accuracy and repeatability Applied force Weight Material Measured features Alignment facilities Software Selling price No. reauired Type or Value Interim checks Large - 2 m max. Medium - l m max. Small-250 mm min. Overall accuracy not so important as repeatability. Probine forces 30 kg maximum Choice to be made according to stability. Measured distances in X, Y and Z. Derived squarenesses in XIY, X/Z and Y/Z To be considered at different levels in the measurement plane. Specialist software not required Large Euro Medium Euro Small Euro One per site Comments.Artefact to be easy to use Three possible artefact sizes to be considered. Since this artefact is to be used for interim checks only, then good repeatability is far more important Artefact to be strong enough Should be portable and easy to lift Temperature stability, mechanical stability and cost are important. Measurements must be completed within a short timescale. The artefact is to be used with CMM manufacturers' existing software Maximum target prices for the final commercial artefact For thermal stability reasons 6 Detailed artefact design Work has focused on the design of a new artefact to meet the design requirements identified in the previous section. A number of designs were proposed and initial prototypes developed and tested. After testing the prototype artefacts the advantages and disadvantages were considered and it was decided to proceed with the development of a granite square artefact. This artefact had many advantages, which made it ideal for interim checks on small to medium sized CMMs. As over 80% of all CMMs manufactured per annum have working volumes less than (750 mm13, it was considered by the SME partners in the project that this market represented by far the best potential for commercial exploitation. Furthermore one of the other proposed designs was far more suitable for larger machines and it was decided to develop this artefact specifically for larger CMMs at a later date.

7 7 The Granite Artefact This is basically a two dimensional artefact, although diagonal measurements (3-axis) are possible by placing the artefact at approximately 45 degrees to the machine axes, the probe direction being set to match. The prototype measured approximately 500 mm long by 310 mm wide, and was made from black granite, thereby ensuring stable results. This artefact was designed in response to three particular requirements: Reduced checking times The provision of one artefact per machine The need to maintain it's geometric and dimensional characteristics with time. A schematic diagram showing the calibrated dimensions of the granite artefact is shown in Figure 1 CALIBRATED DIMENSIONS Distance A to B mm Distance B to C mm Distance D to E mm Angle AB to BC degrees Angle BC to DE degrees Angle DE to AB degrees B Figure 1 : Schematic diagram of granite artefact The artefact is assembled in two parts. The first part is basically a square while the second comprises a straight edge. This straight edge is bonded onto the longer side as shown. Two large holes are machined in the body of the square to

8 124 Laser Iletrolon and Ilacl~rne Perjorrntrnce reduce the weight. All surfaces are lapped flat and particular attention has been paid to the parallel end faces, marked A to E. Three reference marks (+) have been made on the flat (triangular) face, for use when levelling a plane on the part. The artefact has two large holes to reduce its weight and for possible clamping and lifting. The three measured distances between each of the faces A to B, B to C and D to E, together with the three angles at the at the corners can be measured in a suitable laboratory before delivery to the operator. This provides reference dimensions for comparison with those measured during a quick "health check on the CMM. The following tests could be carried out on a CMM using this artefact depending on the orientation on the CMM bed. Length measurement in two planes Compound length measurement Up to three Angular measurements Straightness of an axis 7.1 The placement of the artefact on the table The artefact can be set on any face and levelled in any plane, provided the reference marks (+) are used in setting the level. The easiest orientation is in the XY plane, i.e. flat down on the working surface with the reference marks on the top face. For measurement in the YZ or ZX planes, or in the diagonal direction, the artefact must be mounted on blocks or parallels to give access for the probe to all three sides and the lower face. It can be set down on any of the sides AB, BC, or DE; for the best accuracy, with the longest side AB roughly parallel to the longest axis of the CMM. 7.2 Advantages of the granite artefact The main advantages of the granite artefact can be summarised as follows. Stability, by virtue of the fact that granite is used. Repeatability, providing measurements of definite lengths and angles. This artefact can complement the measurement of difficult components by providing readily available references, i.e. squares, lengths and angles. Additional features of straightness and hysteresis checks are available. It is a simple artefact to measure and understand. The main results comprise only measurement of lengths and angles.

9 8 Conclusions Laser Iletrologv and.uuchrne Perfornzium 125 The performance verification tests have shown that good results can be achieved using the laser interferometer. The introduction of the laser interferometer into the standards is considered to be very important, particularly for complete volumetric analysis of larger machines where a large artefact would be impractical. In addition performance verification testing with a laser is now less time consuming due to the faster set-up times available with the latest laser interferometers. As a result of tests carried out during the course of this project a report has been prepared as a recommendation for the use of the laser interferometer in the performance verification of CMMs. Recommended procedures for the use of the laser in both acceptance and periodic reverification tests are detailed in this report. The artefacts quoted in IS are generally to be used for acceptance testing and periodic re-verification and the results of the project have highlighted the need for a universal artefact which can be used as an interim check device. As CMMs are now much more versatile, portable and workshop oriented, an interim check artefact is required, which is quick and easy to use by a wider range of workshop personnel, other than the qualified inspector. This interim check, or quick check verification, should be performed on a daily, or at least weekly, basis to re-assure the CMM user that the machine is still performing to the same standard as after the previous interim check. The granite artefact proposed is a simple checking device with a high degree of repeatability. The measurement routine is extremely simple, is quick to implement and can be easily introduced for both manual and motorised types of CMMs. A certificate of the true distances and angles, obtained from a recognised laboratory, would encourage the user that the results were satisfactory. The Error Simulation Programme (ESP), originally developed by Huddersfield University for machine tool use, has been tested on CMMs in the UK, Belgium and Ireland. The ESP software has been used to assess the volumetric accuracy of these machines and preliminary results were encouraging. A number of modifications have been made to the original software to change it from being primarily a machine tool programme to a CMM programme, catering for all types and configurations. Further testing of the modified software is required before it can be offered as a commercial package but it is believed that this software has an important role to play with its comprehensive display of all the geometric errors. References [l] Knapp W., Tschudi U. and Bucher A. A Comparison of Different Artefacts for Interim Co-ordinate Measuring Machine Checking: A Report from the Swiss Standards Committee. Precision Engineering, Vol. 13 No 4, pp , October CMMA. Accuracy Specification for Co-ordinate Measuring Machines.

10 126 Laser Iletrolo,q and \lac/nne Perjormance Co-ordinate Measuring Machine Manufacturers Association, BS6808. Part 1, 1987, Glossary of Terms. Part 2, 1987, Methods of Verifying Performance. Part 3, 1989, Code of Practice. British Standards Institution. ANSIIASME B M. Methods for Performance Evaluation of Coordinate Measuring Machines. The American Society for Mechanical Engineering, New York, VDIIVDE 2617 Parts 1-5. Accuracy of Co-ordinate Measuring Machines. Dusseldorf, NF El Accuracy of Co-ordinate Measuring Machines. France, JIS B Test Code for Accuracy of Co-ordinate Measuring Machines. Japanese Standards Association. IS Co-ordinate Metrology Parts 1 and 2. International Organisation for Standardisation, Switzerland, 1994 Peggs G.N. Developing Standards for CMMs. Quality Today, pp S1- S3, 1998 Abackerli A.J., Miguel P.A.C, and King T.G. CMM Traceability in Manufacturing Systems. 2nd Intl. Congress of Industrial Engineering, Piracicaba, Brazil, October Kunzrnann H., Trapet E., and Waeldele F. A Uniform Concept for Calibration, Acceptance Test, and Periodic Inspection of CMMs Using Reference Objects. Annals of CIRP, Vol. 39 No l, pp , Trapet E. Ensuring the Accuracy of Co-ordinate Measuring Machines. Tutorial of 6th Intl. Precision Engineering Seminar, Braunschweig, Germany, Peggs G.N. Creating a Standards Infrastructure for Co-ordinate Measurement Technology in the UK. Annals of CIRP, Vol. 38 No 1, pp , Miguel P.A.C. CMM Performance Verification, Considerations for a Virtual Artefact Approach. PhD Thesis. University of Birmingham. Postlethwaite S.R. and Ford D.G. Geometric error analysis software for CNC machine tools. Proceedings of 31d International Conference on Laser Metrology and Machine Performance (LAMDAMAP). Computational Mechanics Publications, Southampton and Boston, pp , 1997.