A more efficient method for calibrating discrete element method parameters for simulations of metallic powder used in additive manufacturing

Published

October 31, 2018

Author(s)

Stephen R. Geer, Michelle Bernhardt, Edward Garboczi, Justin Whiting, Alkan Donmez

Abstract

Laser powder bed fusion (LPBF) is an additive manufacturing (AM) process that uses a high-power laser to selectively melt metal powder that has been spread, layer-by-layer, to create parts with highly complex features. Because of the strong influence that the powder spreading process has on the final part quality, a better understanding of the powder behavior and its interactions with existing powder layers and the solidified surfaces during this process is needed. Discrete element method simulations (DEM) provide a particle-scale approach capable of examining these mechanisms. While proper calibration of these simulations provides confidence in the quantitative results, the usual calibration process (hopper method) is computationally time intensive. Because of the wide variety of powder material used in LPBF, which will affect the powder spreading process, and because of the large numbers of particles present in only a small volume of powder, a more efficient calibration process was necessary. Use of a new cloud method was shown to make calibrations more tractable and reduce simulation times by up to 89% when compared to a typical hopper method. Similar to previous studies shown in the literature, a reduction in the particle material's Young's modulus was also used to reduce simulation times by up to 62%. Both sliding and rolling friction were needed for the DEM angles of repose to match the empirical data because of the non-sphericity of some of the powder particles, which has been quantified. The calibrated parameters resulted in determination of a powder density within 10% and an angle of repose within 1% of the targeted experimental values. These parameters will be used in extensive future DEM simulations of the LPBF powder spreading process. While AM processes were the main motivation for this work, the calibration procedures summarized herein can be extended to other gas-atomized powders used in injection molding and other metallurgical fields, as well as many other granular materials.

Citation

Granular Matter

Volume

20

Pub Type

Journals

Download Paper

https://doi.org/10.1007/s10035-018-0848-4

Local Download

Keywords

additive manufacturing (AM), laser powder bed fusion (LPBF), metal powder, discrete element method (DEM), numerical modeling

Citation

Geer, S. , Bernhardt, M. , Garboczi, E. , Whiting, J. and Donmez, A. (2018), A more efficient method for calibrating discrete element method parameters for simulations of metallic powder used in additive manufacturing, Granular Matter, [online], https://doi.org/10.1007/s10035-018-0848-4, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=921752 (Accessed December 7, 2025)