Scholarly article on topic 'Traceable Measurements on Machine Tools - Thermal Influences on Machine Tool Structure and Measurement Uncertainty'

Traceable Measurements on Machine Tools - Thermal Influences on Machine Tool Structure and Measurement Uncertainty Academic research paper on "Mechanical engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia CIRP
OECD Field of science
Keywords
{"metrology systems" / CMM / "process monitoring" / "machine tool" / traceability / temperature / "large volume engineering" / "thermal effects"}

Abstract of research paper on Mechanical engineering, author of scientific article — Robert Schmitt, Martin Peterek

Abstract Flexible manufacturing processes for high quality products at low costs are one of the main research objectives in the field of production technology. The quality inspection of large or complex workpieces manufactured on machine tools often takes place beside the production line. To assure the traceability of the quality inspection the features are measured on a CMM. The manufacturing process is interrupted and transportation, handling and the loss of the original manufacturing setup influence the workpiece quality. The high invest and the mentioned aspects show the need for an machine integrated traceable measuring process. Integrated touch probes can be found in many new machine tools and shall provide the possibility of testing product characteristics right after the manufacturing process in the original setup on the machine tool. Due to disturbances like machine defects or temperature fluctuations of the surrounding, the measurement process is not traceable and the data cannot be used for a process improvement or process control. New Approaches at WZL are aiming to assure the traceability of the inspection processes on machine tools. The fusion of appropriate methods for the traceability of CMM's and innovative calibration methods for machine tools allow the determination of a measurement uncertainty for the measurement system “touch probe and machine tool”. Different methods known from CMMs have been examined and validated for machine tools. As the main influence for deviations of the workpiece quality seems to be the temperature, the impacts of the surrounding temperature on the machine tool structure and the achievable measurement uncertainty was examined in a temperature controlled surrounding. The thermal effects on the tool structure shown by repetitive calibrations can now be directly linked to the measurement uncertainty. The objectives of the research activities are part of the Cluster of Excellence ‘Integrative Production Technology for high Wage Countries’ at the RWTH Aachen.

Academic research paper on topic "Traceable Measurements on Machine Tools - Thermal Influences on Machine Tool Structure and Measurement Uncertainty"

Available online at www.sciencedirect.com

ScienceDirect

Procedia CIRP 33 (2015) 576 - 580

9th CIRP Conference on Intelligent Computation in Manufacturing Engineering - CIRP ICME '14

TRACEABLE MEASUREMENTS ON MACHINE TOOLS - THERMAL INFLUENCES ON MACHINE TOOL STRUCTURE AND MEASUREMENT UNCERTAINTY

Schmitt, Robert; Peterek, Martin*

a+b Laboratory for Machine Tools and Production Engineering (WZL) of RWTH Aachen University * Corresponding author. Tel.: +49 241 80-271 ; fax: +49 241 80-22293 E-mail address: M.Peterek@wzl.rwth-aachen.de

Abstract

Flexible manufacturing processes for high quality products at low costs are one of the main research objectives in the field of production technology. The quality inspection of large or complex workpieces manufactured on machine tools often takes place beside the production line. To assure the traceability of the quality inspection the features are measured on a CMM. The manufacturing process is interrupted and transportation, handling and the loss of the original manufacturing setup influence the workpiece quality. The high invest and the mentioned aspects show the need for an machine integrated traceable measuring process.

Integrated touch probes can be found in many new machine tools and shall provide the possibility of testing product characteristics right after the manufacturing process in the original setup on the machine tool. Due to disturbances like machine defects or temperature fluctuations of the surrounding, the measurement process is not traceable and the data cannot be used for a process improvement or process control. New Approaches at WZL are aiming to assure the traceability of the inspection processes on machine tools. The fusion of appropriate methods for the traceability of CMM's and innovative calibration methods for machine tools allow the determination of a measurement u ncertainty for the measurement system "touch probe and machine tool". Different methods known from CMMs have been examined and validated for machine tools. As the main influence for deviations of the workpiece quality seems to be the temperature, the impacts of the surrounding temperature on the machine tool structure and the achievable measurement uncertainty was examined in a temperature controlled surrounding. The thermal effects on the tool structure shown by repetitive calibrations can now be directly linked to the measurement uncertainty.

The objectives of the research activities are part of the Cluster of Excellence 'Integrative Production Technology for high Wage Countries' at the RWTH Aachen.

© 2014TheAuthors.PublishedbyElsevierB.V.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Selection and peer-review under responsibility of the International Scientific Committee of "9th CIRP ICME Conference" Keywords: metrology systems, CMM; process monitoring; machine tool; traceability; temperature; large volume engineering; thermal effects

1. Motivation

The measurement of large-scale devices is absolute vital for the manufacturing and alignment of many products which modern life depends on. Large Volume Metrology (LVM) is concerned with the measurement techniques and methods for structures, objects and assemblies of a few meters up to tens of meters. These techniques are necessary because the items demand special requirements for the measurement process and the quality control. Structures or objects are too large to fit into conventional measuring machines or to be transported to a calibration laboratory. They have to be measured in process or

in situ. The trade off between increasing work piece dimensions and constant or even decreasing tolerances (for example in the field of large gears for wind power industries) and the measurement in uncontrolled environments sustainably complicates an accurate and traceable metrology. The regulation pressures in many industries request a metrology that is able to keep pace with these demands.[1][2][3]There are different approaches for a traceable LVM in the different fields of manufacturing and assembly of large scale parts. The integration of the measurement process into the machine tool (MT) can improve the process quality and lead to reduced machining waste material, a better conformance with the

2212-8271 © 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Selection and peer-review under responsibility of the International Scientific Committee of "9th CIRP ICME Conference" doi: 10.1016/j .procir.2015.06.087

tolerances required by in directives or even standards and even a reduction of the production time by prevention of tempering time. Especially the possibility of one clamping set up allows re-work processes in the same coordinate system and consequently improves the product quality.

2. Traceability of On-Machine Measurements

The objective to integrate the measurement process into the MT initially seems to be an engineering problem especially as the majority of newer machine are equipped with touch probe systems that can be automatically loaded. But to ensure the comparability of the measurement values the traceability of the measuring process is necessary. The closed calibration chain is the main scientific objective within this approach. The adaption of known methods for the traceability of Coordinate Measuring Machines (CMM) is adapted to the challenges of a MT. The scientific objective is the determination of the uncertainty for the measurement process on the MT.[4][5][6]

There are some main differences between a MT and a CMM. The geometric errors of the MT kinematic for example influence the manufacturing result and prevent if not known a traceable measurement on the MT. The knowledge of these errors is the basis for the uncertainty determination. Beneath these systematic errors, the environment of the shop floor means the biggest challenge to a traceable measurement as the manufacturing of large-scale devices in most instances cannot take place in expensive controlled environments. The dominant uncertainty source for the measurement of large scale devices are in time and space varying thermal effects of the environment and the gravitational distortion of both measuring instrument and measured part. Issues such as gravitational sag, thermal expansion, thermal diffusity and thermal effects on instruments and parts have to be tackled by using multidisciplinary approaches involving dimensional and thermal metrology and state-of-the-art modeling.[7][8][9] The presented concept can be seen as a first step in the chain for an process integrated quality inspection for large scale parts. The model based approach for the user friendly determination of the measurement uncertainty combines methods from the coordinate metrology field with research results in the field of MTs and precision engineering.

3. Measurement uncertainty determination

Different methods are investigated to be adapted for a determination of the measurement uncertainty for the measurement system. The methods to establish the uncertainty of measurement for a CMM are described in ISO 15530 and in VDI 2617. The general guidance on calculating uncertainty in measurement is given in the ISO Guide 98-3:2008 to the Expression of Uncertainty in Measurement (GUM).[10][11][12]

The GUM will also be used to determine the uncertainty as described in DIN ISO 15530 part 3 by using the method of substitution. This experimental approach aims to simplify the uncertainty evaluation by using a calibrated work piece or a referenced standard of a similar geometry and dimension (similarity standards described in clause 5.2).[12]

The second method is the establishment of the measurement uncertainty by the use of uncertainty evaluation software (UES). The software uses Monte Carlo methods as mathematical algorithms to compute the results. In joint projects with the Physikalisch Technischen Budesanstalt in Braunschweig (PTB) the virtual Coordinate Measuring Machine (vCMM), a UES developed by the PTB and the National Physical Laboratory (NPL) in London, shall be further developed for the uncertainty establishment for MTs.

The third approach is linked to the guideline VDI 2617 of the german engineering association (VDI). The concept aims to set up an uncertainty budget for the measurement process. The budget consists of listing each uncertainty source, its magnitude, effect on the measurement result and the correlation with the other sources.[11][13] The theoretical approach was developed for CMMs. The research activities implemented a software bases concept that adapts the methods to the challenges of an on-machine measurement.

4. Uncertainty budget for OMM

The main objective of the approach is the determination of an suitable MPE value for the measuring MT. Regarding the standard DIN ISO 10360 for CMMs the MPE can be determined by the measuring of material standards in the standardized manner. These measurements identify the volumetric error of the measuring machine. The MPE can bee seen as the maximum sum of the volumetric errors within the working room of the CMM.[13]

To evaluate a value that is comparable to the MPE of the CMM an experimental set up described in the standard can be implemented on the MT. This value is calculated regarding the standard. The research activities focused on the question if the MPE as a value describing the volumetric error of the measuring machine can also be determined out of the calibration data of the MT. Therefor the volumetric errors are measured by the calibration of the MT with the ETALON LaserTracer. The knowledge of the mathematical errors and a mathematic model of the MT allow the simulation of the measurement setup for the determination of the MPE. The input for the mathematical model are the calibration data of the MT. The set point of the TCP is used as the grid point for the simulation of the actual value. The deviation between the set point and the actual value is included in the calculation of the MPE. Additional influences in the experimental determination are the uncertainties of the touch probe. This influence is also included in the simulative determination. The results can be validated by a measurement with the commercial software TracCheck that is originally used for CMMs (Figure 1).

Sim jlation --

(T alidatior racChec k)

50 100 150 200 250 300 350 400 450

Test Length[mm]

Figure 1: MPE simulation and validation with TracCheck.

The determination of the measurement uncertainty is calculated with the developed software tool regarding VDI 2617-11. Inputs are:

MPE of the machine tool Information about the measuring task Information about the touch probe Information about the workpiece material Detailed information about temperature: o Surrounding o MT structure o Workpiece.

The result is a task specific measurement uncertainty for a 95 % confidence level.

4. Test Set-up in a controlled environment

The demonstrator to validate the approach of the measuring system consists of:

machine tool 5- Axis Hermle B300U, touch probe Heidenahin TS440, special designed and calibrated workpiece, self developed evaluating software temperature controlled surrounding in a climate chamber (Figure 2).

The demonstrator was set up on a common and smaller machine tool to prove the feasibility of the approach. Accordingly it is adapted to larger MTs for the manufacturing of turbine housings (3x4x6m, 150t) for example.

The surrounding and the MT structure was equipped with a number of thermal sensors. The sensors visualize the thermal load of the surrounding and the MT structure during the measurements. To avoid gradients the structure has to be uniformly tempered. Evaluated was a warm up time of more than 24 hours to ensure a homogenous temperature (figure 3). The experiments were carried out at 20°C (reference temperature), 22°C and 25°C.

The MT is calibrated with a Etalon Laser Tracer at each temperature point to simulate the measurement uncertainty. To validate the results a calibrated master workpiece is measured with the touch probe on the MT. The measurement strategy has to be the same as during the calibration. For the test set up the calibrated workpiece is clamped on the MT, the strategy is set up and the measurement is carried out 25 times. The

Figure 2: Climate chamber and machine tool at WZL in Aachen.

ri'JIt v

-^ \ J f

X \ C uss Slides ( Bads) \

_ Slide (Front) Room

Time [h|

Figure 3: Example temperature curve in climate chamber.

uncertainty will be determined comparing the means of the measured values to the reference values from the calibration certificate.

The uncertainty for the whole system is then calculated regarding the DIN ISO 15530 and the "Guide to the expression of uncertainty in measurements" (GUM).[10][12]

5. Thermal Effects on the measurement process

The temperature induced geometric errors are expected to be the main influence on the achievable measurement uncertainty of the complete measuring system.[14] To evaluate this hypothesis a correlation between the geometric errors of the MT and the measurement data has to be identified. Figure 4 shows the nominal values of a repeat measurement of different measurement tasks at each temperature point.

Figure 4: Measurement of the calibrated workpiece (diameter 80mm).

# Measurements

The results are expectable. The values for the measured task (diameter of 80mm) are growing when temperature of surrounding and MT structure is growing. A further evaluation of the data visualizes the thermal effect on the tool structure and the measurement results in a better way. Determining the center of the workpiece in the working space by calculation the coordinates of the center seem to "shift" (Figure 5).

These deviations seem to visualize the thermal effects induced by the changing of the surrounding temperature. Therefore the deviations should be verifiably with the volumetric error parameters of the MT. Figure 6 shows the simulated difference between the volumetric errors at 22°C and 25°C. The orientation and also the nominal value of these errors are comparable to the deviations seen in the experimental results in Figure 5.

uncertainty of calibration of the calibrated workpiece (calibration certificate) .standard uncertainty of the measurement process that is determined with

y = - / yt

standard uncertainty influenced by the material of the work piece systematic deviation between the values yi indicated by the CMM and the calibration value xcal of the calibrated

workpiece (b = 9 ~ xcal) number of measurements coverage factor

22-25 'C ^ * ■ 20-22 C

20°C 22"C

** 25"C

25-20 C

388.385 338.39 388.395 308.4 388.405 38S.41 388 415 388.42 368.425

X [mm]

Figure 5: Calculated center of the workpiece in the working space of MT.

Figure 6: Deviations of the volumetric errors of MT at 22°C and 25°C.

The results are shown in Figure 7.

The measurement system seems to be quite suitable for a fixed surrounding temperature of 20°C. With a temperature change in the surrounding and in the MT structure, the achievable uncertainty is influenced negatively. The thermal effects on the positioning accuracy of the MT can be visualized and proven by a volumetric calibration at the respective temperature points. This correlation seems to be useful for the determination of the measurement uncertainty even at from 20°C differing temperatures. The knowledge of the geometrical errors furthermore allows the improvement of the achievable task specific uncertainty by an volumetric compensation of the known errors. This compensation can be a volumetric error

U = k* ¡Uçal + b2

ucal: standard uncertainty calculated from the

76.062

Figure 7: Measurement uncertainty at different temperature points.

definition of a "temperature window" with an assigned uncertainty. To specify this temperature range the thermal effects on the machine and the induced errors have to be determined with a complete calibration of the machine. Another possibility is an "interim" check of the positioning accuracy of the MT. The test has to be user-friendly and fast to allow the implementation in a production line. The development of a method is connected to an innovative measurement set up or technology. The research at WZL focus on absolute interferometry and a

further developed mathematical approach for the error determination. The method and set up is designed to be suitable for larger MTs (working space of more than 3x3x2m for example).

compensation software solution running on the MT control or an offline approach that will correct the measured data.

7. References

6. Outlook and Challenges in LVM

The experimental determination of the measurement uncertainty is adapted to a measurement system based on a common MT. The method can be used to ensure the traceability of the measurements run on the machine. The experimental approach requires a workpiece, that is similar to the manufactured and measured part. Therefor the solution is not very flexible and especially for larger parts laborious and expensive.

The model based uncertainty budget determination is developed and seems to be suitable to overestimate the uncertainty of the measuring system without the need of a calibrated workpiece. The approach presumes the detailed knowledge of the MT to set up a mathematical model. With information about the geometrical errors of the MT, the touch probe uncertainty, the measurement task, the temperature of the surrounding and the workpiece and the material of the workpiece the uncertainty can be overestimated.

The next objective is the validation of the results for different temperature points and even a guideline for instable temperature states. This will presume a better understanding of the thermal effects in the MT structure. The instable status causes gradients inside the structure and the induced deviations are hard to simulate or predict. The interaction between temperature and measurement uncertainty requires the

[1] [1] Estler, W.T. et alt., Large-Scale Metrology - An Update, CIRP Annals Manufacturing Technology 2002 (51)

[2] Peggs, G. N. et alt., Recent developments in large-scale dimensional metrology, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering

Manufacture, 2009

[3] Franceschini, F. et alt., Distributed Large-Scale Dimensional Metrology: New Insights, Springer 2012

[4] Trapet, E. et alt., Traceability of Coordinate Measurements According to the Method of the Virtual Measuring Machine, Final Project Report, 1999.

[5] van Dorp, B. et alt.: Traceability of CMM Measurements, ASPE 1999 Annual Meeting, Monterey (1999)

[6] A. Balsamo et alt., Evaluation of CMM Uncertainty Through Monte Carlo Simulations, CIRP Annals Manufacturing Technology 1999(48),

[7] Schwenke, H et alt., Geometric Error Measurement and Compensation of Machines- an Update, CIRP Annals, Manufacturing Technology, 2008

[8] Lewis, A., Large Volume Metrology in Industry, Joint Research Project in EMRP Programme

[9] Lin, Y.; Shen, Y.; Modelling of Five-Axis Machine Tool Metrology Models Using the Matrix Summation Approach, Washington D.C., 2003

[10] ISO/IEC Guide 98-3:2008: Uncertainty of measurement - Part 3: Guide to the expression of uncertainty in measurement, ISO, Genf 2008

[11] Hernla, M.: Uncertainty of coordinate measurements, QZ, 2010

[12] DIN ISO 15530, Geometrical product specifications (GPS) - Coordinate measuring machines (CMM): Technique for determining the uncertainty of measurement - Part 3: Use of calibrated workpieces or measurement standards, ISO 2011

[13] DIN ISO 10360, Geometrical product specifications (GPS) - Acceptance and reverification tests for coordinate measuring machines (CMM), ISO 2010

[14]Mayr, J. et alt., Thermal Issues in Machine Tools, CIRP

Annals Manufacturing Technology, 2012