An investigation into non-destructive testing strategies and in-situ surface finish improvement for direct metal printing with SS 17-4 PH : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Engineering at Massey University, Albany, New Zealand
Loading...
Date
2022
DOI
Open Access Location
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Massey University
Rights
The Author
Abstract
Additive Manufacturing (AM) technologies have the potential to create complex geometric parts that can be used in high-end product industries, aerospace, automotive, medical etc. However, the surface finish, part-to-part reliability, and machine-to-machine reliability has made it difficult to qualify the process for load dependent structures. The improvement of surface finish on metal printed parts, is a widely sought solution by these high-end industries and non-destructively characterizing the mechanical aptitude of metal printed parts, would pave the way for quality assessment strategies used to certify additively manufactured parts. This thesis examines the capability of laser polishing and non-destructive testing technologies and methods to address these difficulties.
This research study presents an investigation into quality management strategies for Direct Metal Printing (DMP) with powdered Stainless Steel 17-4 PH. The research aim is split into two key categories: to improve the surface finish of metal additive manufactured parts and to non-destructively characterize the impact of defects (metallurgical anomalies) on the mechanical properties of the printed part. To improve surface finish of a printed part, a novel methodology was tested to laser polish the Laser-Powder Bed Fusion (L-PBF) parts during print with the built-in laser. Numerous technologies for non-destructive testing techniques already exist, and in the duration of this doctoral study various technologies were explored. However, the final solution focuses on layer-wise capture with a versatile low-cost imaging system, retrofitted within the DMP machine, to capture each layer following the lasering process. In addition, the study also focuses on progressing the characterization of data (images), using a combination of image processing, 3D modelling and Finite-Element-Analysis to create a novel strategy for replicating the as-built specimen as a computer-aided design model and performing simulated fatigue failure analysis on the part.
This thesis begins with a broadened justification of the research need for the solutions described, followed by a review of literature defining existing techniques and methods pertaining to the solutions, with validation of the research gap identified to provide novel contribution to the metal additive manufacturing space. This is followed by the methodologies developed, to firstly, control the laser parameters within the DMP and examine the influence of these parameters using surface profilometry, scanning electron microscopy and mechanical hardness testing. The control variables in this methodology combines laser parameters (laser power, scan speed and polishing iterations) and print orientation (polished surface angled at 0º, 20º, 40º, 60º, 80º and 90º degree increments from the laser), using several Taguchi designs of experiments and statistical analysis to characterize the experimental results.
The second methodology describes the retrofitted imaging system, image processing techniques and analysis methods used to reconstruct the 3D model of a standard square shaped part and one with synthesized defects. The method explores various 2D to 3D extrusion-based techniques using a combination of code-based image processing (Python 3, OpenCV and MATLAB image processing toolbox) and ready-made software tools (Solidworks, InkTrace, ImageJ and more).
Finally, the new research findings are presented, including the results of the laser polishing study demonstrating the successful improvement of surface finish. The discussion surrounding these results, highlights the most effective part orientation for laser polishing the outline of an AM part and the most effective laser parameter combination resulting in the most significant improvement to surface finish (roughness and profile height variation). Summarily, the best improvement in surface roughness was achieved with the <80 angled surface with the laser speed, laser power and polishing iterations set to 500mm/s, 30W, 3 respectively. The sample set total average measured a 16.7% decrease in Ra. NDT digital imaging, thermal imaging and acoustic technologies were considered for defect capture in metal AM parts. The solution presented is primarily focused on the expansion of research to process digital images of each part layer and examine strategies to move the research from a data capture stage to a data processing strategy with quantitative measurement (FEA analysis) of the printed part’s mechanical properties. In addition, the results discuss a method to create feedback to the DMP to selectively melt problematic areas, by re-creating the sliced part layers but removing the well-melted areas from the laser scanning pattern.
The methods and technological solutions developed in this research study, have presented novel data to further research these methods in the pursuit of quality assurance for AM parts. The work done has paved the way for more the research opportunities and alternative methods to be explored that complement the methods detailed here. For example, using a combination of in-situ laser polishing, followed by post-processing the AM specimens in an acid-based chemical bath. Alternatively, further exploring acoustic NDT techniques to create an in-built acoustic-based imaging device within the AM machine. Finally, this thesis cross-examines the work done to answer the research questions established at the start of the thesis and verify the hypotheses stated in the methods chapter.
Description
Figure 1.1 is re-used under an Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) licence
Keywords
Additive manufacturing, Metals, Finishing, Nondestructive testing