Comparability and uncertainty of shape measurements with white-light interferometers

Abstract

We discuss how the results obtained from a white-light interferometer can be compared to tactile measurements. The core idea to achieve comparability is to determine a short cut-off wavelength up to which the spatial frequency components of the surface topography are measured with less than 3 dB attenuation. We demonstrate for different interferometers that the data has to be filtered to achieve a linear transfer characteristic which allows to define the short cut-off wavelength. In addition, we demonstrate investigations of the error sources in shape measurements that we have identified. Results of our work are influencing a VDI/VDE calibration guideline for shape measurements which is currently under development. We show in this paper how the procedure developed for the guideline can be employed to real measurement devices. Uncertainty contributions to the error budget are also discussed and measurements on shape standards are presented.

Principles of interference microscopy for the measurement of surface topography

“…concepts of resolution in optical microscopy as applied to 3D quantitative measurement of surface topography [84,85]. The working consensus defines topographic spatial resolution as the ability to distinguish closely spaced surface features in a surface topography map [86][87][88].…”

“…Given the importance of the noise specification, it is important to consider technical means to improve it. Surface filtering is a legitimate postprocessing step in areal surface topography and may even be required for correlation to other measurement techniques that may have a different spatial frequency response [85]. Lateral smoothing, or the use of the S filter as described in ISO 25178-3, can considerably improve the measurement noise, as illustrated in Fig.…”

Metrological characteristics for the calibration of surface topography measuring instruments: a review

“…one where the amplitude transmission is assumed to be 100%) can be used as the reference. However, the response to a rectangular structure with a certain size does not necessarily predict the response to other sizes or shapes, so care should be taken when using rectangular cross-section topographies with optical instruments, especially when approaching an instrument's linear operating range (see section 2.7 and [27,78,79]).…”