Surface topography datasets acquired using different measurement systems: a) confocal microscopy; b) coherence scanning interferometry; c) focus variation microscopy; d) X-ray computed tomography.
Additive manufacturing (AM) is beginning to mature. In the past few years, we have seen technologies, originally conceived as methods to quickly prototype new products, become genuinely viable for the manufacture of high-value products, for serious application in the medical, aerospace and automotive sectors.
In order to provide quality control for functional surfaces, parts must be inspected in some way for compliance to a specification. Further to the need for quality control, however, the information present on surfaces can be used at the process development stage. Particularly the surfaces produced by novel processes such as AM can be used to make inferences about the physical interactions taking place during the process, using the features on the surfaces as a ‘process fingerprint’ to elucidate information about the process itself.
For both of these applications, surface measurement and characterisation can be performed. However, surface measurement of AM parts is complex.
AM surfaces exhibit features at a large range of measurement scales, thereby requiring measurements of large areas acquired at high resolutions. Features often include non-uniform surface properties, high slopes, undercuts and step-like transitions that cause unpredictable behaviour when measured using conventional technologies.
These issues generally cause problems for established measurement systems, with conventional contact and optical technologies often providing erroneous results when applied to the AM case. To further complicate the issue, AM parts often contain surfaces of functional importance that are difficult to access, thereby preventing measurement by conventional technologies entirely.
Work performed at the University of Nottingham and at other research institutions has begun to address these issues. Through this work, we have formed a deep understanding particularly of metal powder bed fusion (MPBF) surfaces, examining the features present and the ways in which measurement instruments interact with the surfaces.
Recently, we have presented measurement good practice for MPBF surfaces, providing solutions for a variety of existing technologies. For the measurement of internal and otherwise difficult-to-access surfaces, X-ray computed tomography has recently been pioneered as a viable method of surface measurement.
Aside from the issues relating to data capture, surface characterisation of MPBF surfaces is complex – the characterisation methods and parameters generally used were originally developed for the random surfaces that result from variations in conventional machining processes, but are often not suitable for characterising the unique complexities that exist on an MPBF surface.
Surface topography datasets acquired using different measurement systems: a) confocal microscopy; b) coherence scanning interferometry; c) focus variation microscopy; d) X-ray computed tomography.
In order to address this issue, novel methods of characterisation have been developed, using algorithmic techniques of feature extraction to separate specific features from their surroundings, and facilitate dimensional measurements of these features. Such analyses allow both researchers and quality engineers to acquire new information for iterative process improvement and functional performance testing.
With these advances, the difficulties experienced in measuring and characterising AM surfaces are gradually being overcome, and the challenges are being addressed. With the development of good practice, researchers and industrialists can have increased confidence in their data, and the novel methods of characterisation developed provide new tools for verification and process development.
Despite these advances, the book on AM surface measurement remains open; significant further research requirements have emerged from the work already reported. Particularly, good practice development continues – while a variety of instrument-surface interactions are now better understood, a number of instruments remain unexplored. The use of XCT for surface measurement also requires further experimentation, as, because of its novelty, XCT still has an array of unsolved measurement issues.
Additionally, while MPBF surfaces have now been well studied, polymer powder bed fusion (PPBF) surfaces remain largely unexamined, as do surfaces produced by a number of the other metal AM processes. On the PPBF front, work to provide an understanding of these surfaces is just beginning, but early data implies that polymer surfaces exhibit all of the aforementioned complexities of their metal counterparts, with the additional complication of material translucency. Such translucency is likely to affect optical technologies, and so polymer surfaces represent another mountain to climb in conquering the AM surface.
