IX Tri-National Conference of the North American Coordinate Metrology Association

Ponencias Confirmadas


Eugen Trapet

Trapet Precision Engineering

Ver info

Otto Jusko

Physikalisch-Technische Bundesanstalt

Ver info

Steven Phillips

National Institute of Standards and Technology

Ver info

Alessandro Balsamo

Istituto Nazionale di Ricerca Metrologica

Ver info

Joel Barbosa


Ver info

Maximino Cruz


Ver info

Héctor Aguilar

Soporte Metrology

Ramón Zeleny


Ver info

Rubén Galindo


Ver info

David Gale


Ver info

Octavio Icasio


Ver info

Edgar Arizmendi


Ver info


  • Otto Jusko

Investigations on high precision CMM form scanning

It may be very useful in industry to be able to measure form along with other specifications once a work-piece has been set on a Coordinate Measuring System. A second measurement on a dedicated Form Measuring Machine (FMM), requires additional time to set-up the work-piece again and perform these measurements. When scanning of a work-pieces was developed on CMS, its application was hampered by the limited speed of the scanning process as well as by the time and complexity required to process usually large number of data points. Additionally, it was not accurate enough for most applications. Nowadays, however, scanning on CMS has been optimized by CMS manufacturers. It can be performed at convenient speeds, data processing has dramatically improved and manufacturers claim accuracy has also suffered a significant improvement. Nevertheless, users question the achievable accuracy, as work performed when scanning had been introduced did not produce very good results at that time.

To qualify the accuracy of the form scanning process, we have just finished a project which investigates form scanning with a coordinate measuring instrument either with three or four active axes i. e. with and without a high precision rotary table. A selection of calibrated form artifacts either with simple (e. g. sphere) or complex profile geometries (e. g. multi wave standard) were measured under scanning mode on a 3D CMS (Zeiss PRISMO 7 ultra CMM). The influences of measurement parameters such as probing system (force, probing sphere diameter, geometry and material), scanning speed, as well as the determination of the rotary axis position were investigated. The results were compared with those obtained on form measurement instruments. Additionally, pure form parameters like roundness of a sphere and straightness, parallelism and squareness of cylinders were examined. The study shows that CMS form measurement results can have high reproducibility and reasonably low uncertainties can be achieved in special cases [1]. The current advantages and limitations of this technique are discussed. Major uncertainty influences were identified as thermal drifts, probe material and filter settings. The latter was specially investigated to determine the effective signal transmission bandwidth of the whole CMS. CMS rotary tables are not equipped with internal eccentricity and tilt compensation as FMM. The advantages and drawbacks of that fact were also investigated.

The results shall be used to improve the standardization of form measurements with CMS which is currently not very well defined.

  • Alessandro Balsamo

Uncertainty of measurement of calibrated test lengths realized by interferometry in ISO 10360-2 testing

ISO 10360-2 standard is concerned with testing CMM (Coordinate Measuring Machines) using alternative measuring calibrated lengths, realized either by material standards or by interferometry. Whatever the calibrated test lengths, this standard requires the evaluation of the test value uncertainty, and hence it’s several components; one of them is the uncertainty in realizing the calibrated test lengths.

ISO 10360 2 tests CMM vs. a stated MPE (Maximum Permissible Error). Whether a CMM passes or fails the test is decided according to the decision rule defined in the ISO 14253 1 standard, which is based on the measured values and their uncertainties. Consequently ISO 10360 2 requires the evaluation of the test uncertainty through ISO 14253 1 standard. To estimate this uncertainty there is a recommendation, ISO/TS 23165, which provides guidance. However, it was published when the ISO 10360 2:2001 was in force and it did not foresee interferometric measurements. The current version ISO 10360 2:2009 allows for the use of diverse calibrated test lengths, including those realized by interferometry, however, the uncertainty estimation of this technique is not covered in ISO/TS 23165.

This presentation addresses how to evaluate the uncertainty when interferometry is used in ISO 10360 2 standard.

Principles of achieving metrological traceability with CMM

A recognised strength of CMMs is their versatility and capability to adapt to very diverse geometries, a virtually infinite variety of possible measurands. On the other hand, the traceability of measurements done with CMMs is very difficult to document and prove: many uncertainty contributors are involved, propagated through possibly complex individual part programmes involving non trivial – and often undocumented – algorithms. ISO 15530 series of standards helps in evaluating the task-specific uncertainty of measurements taken with CMMs. This is usually deemed as sufficient for inspecting parts in industry, even though with limitations due to practical applicability. However, it may not suffice for calibrations performed by officially accredited laboratories.

This presentation addresses this topic and illustrates the approach used at the INRIM for calibrations done with a CMM. The method resorts as much as possible to comparison with auxiliary calibrated standards providing traceability. This is usually sufficient when simple geometries are involved, as it is the case for most standards under calibration; it requires additional investigation when the measurands are defined by complex specification operations, i.e. when the geometry involved is complex.

  • Eugen Trapet

Interlaboratory Comparison for geometric error compensation Parameter measurement

For the geometrical error compensations there are different error parameter measurement methods are on the market since about 1983, when the first full geometric error correction was implemented on a CMM.

To our knowledge there has never been an attempt to compare the performance of such methods, at least not known to a wide public. Therefore, ISM3D, that already collaborated on a European research project (SOMMACT) advocated to making machine tool correction more efficient, included this performance test different measurement techniques a, instruments and methods.

ISM3D invited some instrument manufacturers and calibration service providers with the most advanced measurement systems to participate in this exercise in the facilities of ISM3D in Gijón, Spain, in order to perform the measurements under exactly the same conditions on a CMM of 2 m3 measuring volume.

Every participant measured the uncorrected CMM with its own method under the same conditions. The results were to be presented in such a way that they could be compared. This meant that the so-called 21 kinematic error model was used, with 3 translational errors and 3 rotational errors per axis, all as a function of axis position.

The values of that participant with the results best-placed in the entire bunch of results from all participants were finally used to correct the CMM.

With the machine so corrected, an ISO 10360-2 verification was performed by ISM3D.

Participants and their used methods were:

  • Participant 1 (Spain) with
    • Laser interferometer of RENISHAW classic
    • Ball bar 3D
    • Ball Bar 2D with camera probe
    • Ball Plate 2D
  • Participant 2 (UK, Italy)
    • Laser Tracer
  • Participant 3 (Germany) with
    • Laser Tracer
  • Participant 4 (UK) with
    • 6dof (6 degrees of freedom measuring system)
  • Participant 5 (Spain)
    • Laser Tracer
  • Participant 6 (Spain)
    • Laser Tracer

    In this paper, first the fundamentals of geometric error correction and the alternative methods are explained. Then the comparison measurements are discussed as well as the observations made during the machine calibrations by the different teams of experts. Finally the results are compared with respect to their ease of handling, the time required and the differences in the results. One of the most important observations was that for all methods much more in-depth knowledge and practical experience is required than is generally assumed to be necessary. There was no “winner” in accuracy, because most methods yielded very similar results, but there was one system which was particularly convincing with its speed and reliability (“right at the first try”). Note that other systems might be favorable when other criteria is taken in account, such as universality or the applicability for rotary axes.

    • Steven Phillips

    Measurement Sampling Strategies

    Measurement sampling strategies describe the number and location of the measurement points on the work-piece surface. All Coordinate Measurement Systems (CMSs) involve sampling strategies. The sampling strategy can significantly affect the accuracy of the measurement result and involves how the measurement point locations interact with the form error of the work-piece (i.e., imperfect geometry), the errors from the CMS, and the fitting algorithm. This talk will describe these effects and make recommendations for sampling strategies that will improve the accuracy of measurement results. Also the issue of partial features, e.g. partial arcs, will be presented; these features can produce extremely large errors (if not properly addressed) due to the restricted nature of the sampling strategy.

    • Ramón Zeleny

    Tolerancias Dimensionales y Geométricas (GD&T) – Especificaciones Geométricas de Producto (GPS) y Máquinas de Medición por Coordenadas (CMM)

    Se considera frecuentemente que la Norma ASME Y14.5-2005 sobre Tolerancias Geométricas (GD&T) y las normas ISO GPS concuerdan bastante bien, sin embargo, la realidad es que alrededor de 1985 más o menos coincidían pero ISO se ha ido alejando de ASME en las diferentes normas publicadas entre ese año y el actual. El desarrollo de las normas ISO se ha enfocado en la medición con máquinas de medición por coordenadas (CMM) mientras que ASME en el uso de patrones funcionales aunque ninguna de ellas es una norma de inspección. Se comentan algunas diferencias y la importancia de seleccionar apropiadamente los criterios de asociación al efectuar mediciones y establecer referencias (datum).

    La realidad es que el uso de CMM ha ido ganando terreno para la medición de partes toleradas geométricamente especialmente en el caso de superficies de forma libre (tolerancia de perfil) debido, entre otras cosas, al incremento de la cantidad de puntos adquiridos para hacer una medición pasando del palpado discreto, al continuo y luego a nubes de puntos.

    Las especificaciones han ido cambiando del dibujo 2D al modelo 3D lo que está facilitando la manufactura e inspección de partes hasta llegar en la actualidad a poder tener la posibilidad de generación automática de programas de parte en muy pocos minutos a partir de la información de producto y manufactura (PMI), reduciendo así la variación de criterios entre programadores de CMM al medir dimensiones y tolerancias geométricas.