Simulation of transport phenomena and melt pool shape for multiple layer additive manufacturing

Yousub Lee, Dave F. Farson

    Research output: Contribution to journalArticlepeer-review

    73 Scopus citations

    Abstract

    A three-dimensional (3D) numerical simulation of transport phenomena using volume of fluid method is conducted for multiple-layer single track laser additive manufacturing (LAM). It predicts the temperature and fluid flow velocity distributions, transient variation of the melt pool fluid boundary shape and remelting, and solidified build geometry during deposition of successive LAM layers. The prediction showed reasonable accuracy in predicting peak temperature and deposit geometry. The prediction error of peak temperature is less than 2.5%, and prediction error of deposit height and width are less than 12%. Correlations between dimensionless process/material parameters (Pe, Pr, and Ma) and melt pool 3D fluid flow patterns and liquid boundary shape were studied. The analysis showed the hemispherical melt pool free surface in LAM causes the mechanisms that determine melt pool liquid-solid boundary shape to be different from melt pools formed on a flat surface. Although the maximum surface velocity decreased from 8.59 cm s-1 on the first layer to 5.06 cm s-1 on the fifth layer, the outward Marangoni flow is redirected from outward to downward due to the increase of surface curvature from 39.5° on the first layer to 75.9° on the fifth layer. Consequently, the penetration into the solid substrate at the outward edges becomes deeper and the pool bottom becomes more convex. These detailed physical insights provided by process simulations facilitate prediction of localized dimensional variations in LAM builds.

    Original languageEnglish
    Article number012006
    JournalJournal of Laser Applications
    Volume28
    Issue number1
    DOIs
    StatePublished - Feb 1 2016

    Fingerprint

    Dive into the research topics of 'Simulation of transport phenomena and melt pool shape for multiple layer additive manufacturing'. Together they form a unique fingerprint.

    Cite this