Carnegie Mellon University
Browse
Wu_cmu_0041E_10756.pdf (16.02 MB)

Investigation and Mitigation of Powder Anomalies Induced Porosity in Powder Bed Additive Manufacturing

Download (16.02 MB)
thesis
posted on 2022-11-16, 22:23 authored by Ziheng WuZiheng Wu

Defect content is one of the major obstacles to the wider adoption of additive manufacturing (AM) as the field is still actively learning how to control it, developing standards to quantify it, and building up knowledge of its formation and impacts. On the other hand, there is a growing interest in reducing AM fabrication cost by using recycled materials and economically produced powder. State-of-the-art powder-based AM processes typically accept gas-atomized spherical powder with low entrapped gas porosity. However, using non-standard powder feedstock, e.g., the non-spherical hydride-dehydride (HDH) Ti-6Al-4V powder and the highly porous 17-4 PH stainless steel powder, can be more cost-efficient. This work presents two successful applications of the non-standard feedstocks through process optimization by measuring the process windows for the fully dense components and achieving comparable mechanical properties as the standard AM counterparts. Additionally, the author used synchrotron based high-speed imaging technique to visualize and quantify the porosity formation processes induced by the anomalies of the non-standard powders, i.e., irregular morphology and powder porosity. By coupling them with pore shape analysis and powder packing analysis, three powder induced porosity formation mechanisms were proposed. Melt pool dynamics, powder packing characteristics, and powder-laser interactions are believed to be the key factors for powder induced porosity formation. The optimization guideline and the better understanding of the porosity formation can certainly be generalized for the application of other non-standard feedstocks which could benefit the AM community.

History

Date

2021-11-28

Degree Type

  • Dissertation

Department

  • Materials Science and Engineering

Degree Name

  • Doctor of Philosophy (PhD)

Advisor(s)

Anthony Rollett

Usage metrics

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC