Calibration is central to most measurement procedures. This is especially true in those cases where a large number of difficult-to-identify and difficult-to-control factors hinder the experimenters in their efforts to obtain reliable measurement results. Dimensional measurements of features on the micro- and nano-scales is one such case. A white-light interferometer (WLI) microscope can perform measurements of a variety of measurands over a broad dimensional range: from surface texture characterisations on the nano-scale to measurements of step heights of several millimetres. Calibration methods based on the hypothesis of a linear calibration curve can be inadequate to express the relationship between measurement results and traceable reference materials (RMs). A calibration procedure built into a commercially available WLI microscope is critically compared with methods presented in an international standard. This comparison is enabled by a cost-effective procedure for establishing traceable RMs in the micro-range. Advantages of calibration procedures based on more than one RM are then demonstrated within the ranges from 180.5 to 219.5 μm and from 1.5 to 501.5 μm. Calibration methods involving regression modelling of transformed measurement results are considered for these two intervals to overcome the highlighted weaknesses of the calibration procedure built in the examined WLI.
|Number of pages||11|
|Journal||International Journal of Advanced Manufacturing Technology|
|Publication status||Published - 10 Mar 2010|
- White light interferometry
- Regression analysis