Factors affecting the accuracy of areal surface texture data extraction from X-ray CT

Abstract

The ability to perform non-destructive areal surface analysis of the internal surfaces of additively manufactured (AM) components would be advantageous during product development, process control and product acceptance. Currently industrial X-ray computed tomography (XCT) is the only practical method for imaging the internal surfaces of AM components. A viable method of extracting useable areal surface texture data from XCT scans has now been developed and this paper reports on three measurement and data processing factors affecting the value of areal parameters per ISO 25178-2 generated from XCT volume data using this novel technique. X-ray, Metrology, Additive manufacturing 2.1. Measurement plates Individual rectangular plates, approximately 10 mm x 20 mm, were cut from a Rubert Microsurf 334 (casting) test panel. The casting panel was used as this surface was considered to most closely represent the surface of a powder bed fusion (PBF) metal AM component. The nominal surface Ra values for the plates used for this work were 50 µm and 25 µm as these approximate the as-Contents lists available at SciVerse ScienceDirect

Characterisation of additively manufactured metal surfaces by means of X-ray computed tomography and generalised surface texture parameters

“…Industrial X-ray computed tomography (CT), which is already successfully used for dimensional quality control of AM parts [7], has recently started to be considered also as a viable technique for AM surface topography evaluation [8]. In particular, currently available CT systems are capable of overcoming the main limitations of contact and optical measuring techniques, as they can be used to measure external as well as internal and difficultto-accesses AM surfaces, non-destructively and with adequate metrological structural resolution [9].…”

Introduction of a Surface Characterization Parameter Sdrprime for Analysis of Re-entrant Features

“…Measurement accuracy is dependent upon voxel size and surface texture roughness value. The resolution has to be sufficient to capture the required information at the required scales-of-interest [4]. It is expected that the accuracy and resolution limits will be similar for characterisation including the re-entrant features.…”

“…Characterisation of a re-entrant surface using line-of-sight measurement instrumentation and using height map analysis may, depending upon the surface texture parameter evaluated, produce significant errors. X-ray computed tomography (CT), used in this study, has no such line-of-sight restrictions and has been used successfully for the measurement of internal surfaces [4,5], dimensions [6,7] and porosity [8]. CT data is true 3D data, consisting of (x, y, z) co-ordinate information.…”

Automated Defect Inspection Algorithm Incorporated Spectral-Domain Optical Coherence Tomography for Optical Polarizing Thin Films

“…Visual inspection methods, machine vision with a charge-coupled device (CCD) camera-based inspection or electronic inspection have been frequently employed for the quality inspection in optical thin film industry [3,4]. Since these techniques provide information of top surface information, highly penetrating X-ray and computed tomographic techniques were considered for sub-surface, but none of them provides depth resolution at the micro-meter scale [5,6]. Since optical polarizing thin films based products are composed of several optical thin films, an optical inspection method with high depth-resolved resolution can be an ideal solution.…”