We describe the application of several surface profile filters as an enabling tool in the dimensional measurements of an engineering artifact, namely, a fuel cell plate. We recently reported work on the development of a non-contact system for dimensional metrology of bipolar fuel cell plates. That system is comprised of two laser spot triangulation probes that acquire profile data across a plate. While the non-contact system provides rapid measurements (measurement speed of 100 mm/s to 500 mm/s), the data is noisy and cannot be used directly to obtain features of interest such as channel depth and width. In this paper, we show how different surface profile filters such as the spline, morphological, and robust filters, can be employed to identify and suppress outliers and to produce a mean line that serves as a substitute geometry from which we can determine features of interest. Further, we compare the non-contact probe data against contact probe measurements made using a coordinate measuring machine (CMM). Surface profile filters are again useful in correcting the reference data for tip size and also in removing any free form deformation in both data sets prior to parameter evaluation and comparison.
Citation: Measurement Science and Technology
Pub Type: Journals
Dimensional metrology, Filtering, Fuel cell, Laser triangulation, Non-contact