Date of Award


Degree Name

Doctor of Philosophy


Mechanical Engineering


Amit Lopes


Laser powder bed fusion (LPBF) is one of the most widely used additive manufacturing (AM) methods for metal parts. Geometrical tolerances required for industry applications are determined by geometrical measuring methods and process capability calculations. However, geometry characterization presents challenges for LPBF because measuring uncertainty values are not defined and the standardized measurement framework Geometric Tolerancing and Dimension (GD&T) is not fully adopted.

Measurement uncertainty is influenced by this process's high surface roughness (10 µm < Ra ⩽ 80 µm). As an example, when using multiple probing points or scanning pathways on a LPBF surface with the most accurate (+/- 0.004 mm) and traceable measurement method, the coordinate measuring machine (CMM), the results can vary by up to 0.050 mm to 0.070 mm. This measurement uncertainty leads to an imprecise benchmark reference for comparison with optical or X-ray computed tomography (XCT). These measurement variations also make it difficult to accurately characterize parts for automotive, aerospace, medical and other high-tech industries, such as a turbine blade geometry used in aircraft engines with a blade height tolerance of +/- 0.15 mm, which required a measurement uncertainty less than 30% of the tolerances. Mechanical properties such as tensile testing assessments useful in comparisons of materials, alloy development, quality control, and design under certain circumstances are also affected by this imprecise geometry characterization. Overestimating the diameter of the ASTM E8 sample by 0.100 mm can make up to a difference in stress results of 7%.

This research aimed to evaluate various measurement approaches (contact and non-contact) for surface texture and geometrical size dimension characterization and provide information regarding measurement uncertainty and sampling procedures issues to accurately analyze the geometry and process capabilities needed for LPBF industrialization. Based on the research conducted, the LPBF's typical surface texture requires the definition of sampling procedures, and neither research nor industrial applications may directly use the manufacturer's declared repeatability and accuracy of a measurement method developed for the traditional manufacturing process.

To characterize a measurement system and improve the measurement process, statistical tools like the gage R&R (repeatability and reproducibility measurement assessment) are required. The first factor in choosing a measuring method is the tolerance range of interest. Fast and inexpensive measurement methods such as Micrometers and Calipers are an optimal solution when the measurement uncertainty range can be in a range of 0.100 mm. In contrast, Optical, and XCT measuring methods can achieve a medium measurement uncertainty range of 0.030 mm or less following a sampling procedure.

The results indicate that the surface of LPBF typically shows positive skewness and kurtosis values. This curve characterization can be used to estimate the percentages of geometrical deviations. A measurement framework based on a novel cross-sectional method, a combination of surface roughness measurements and size dimension measurements was proposed. The application of this work enables more precise geometric measurements used for research and industry assessments. Future research directions indicate the need for more measuring methods characterization and filtering approaches for asperities from attached particles, and partially melted particles to evaluate their contribution to mechanical behavior and geometrical form and fit properties.




Recieved from ProQuest

File Size

122 p.

File Format


Rights Holder


Available for download on Wednesday, September 18, 2024