1 © 2020 by ASME
Proceedings IMECE2020
2020 ASME International Mechanical Engineering Congress and Exposition
USA
IMECE2020-23550
3D MODELING OF ADDITIVE MANUFACTURING PROCESS:
THE CASE OF POLYMER LASER SINTERING
Lan Zhang
1
, M’hamed Boutaous
1
, Shihe Xin
1
, Dennis A. Siginer
2
Univ Lyon, CNRS, INSA-Lyon, Université Claude Bernard Lyon 1, CETHIL
UMR5008, F-69621, Villeurbanne, France
Email: mhamed.boutaous@insa-lyon.fr, Lan.zheng@insa-lyon.fr, shihe.xin@insa-lyon.fr
2
Centro de Investigación en Creatividad y Educación Superior y Departamento de Ingeniería Mecánica
Universidad de Santiago de Chile, Santiago, Chile
Email: dennis.siginer@usach.cl
ABSTRACT
This work focusses on studying multiphysical transient
phenomena in polymer powders occurring during selective laser
sintering in polymers powders. Multiple phenomena stemming
from the interaction of the laser with the polymer powder bed
and the transfer of the laser power to the powder bed including
laser scattering and absorption, polymer heating, melting,
coalescence, densification, and the variation of the material
parameters with the temperature are simulated via the modified
Monte Carlo-ray tracing method coupled with the Mie theory. A
finite volume method is adopted for the heat transfer. The model
couples heat diffusion, melting, coalescence and densification of
the polymer grains, and the crystallization kinetics during the
cooling steps. Laser intensity is concentrated on the surface of
the material contrary to the predictions of the Beer-Lambert law.
Laser acting on thermoplastic material cause the polymer
powder melt, coalescence between melted grains, air diffusion
versus densification, crystallization and volume shrinkage. All
these processes are simulated by a series of multiphysical
models. The reliability of the modeling is tested by comparison
with experiments in the literature, and a parametric analysis is
performed, based on the process characteristics such as laser
sweep speed, its intensity and shape, polymeric grain size among
others. Several recommendations to optimize the process are
proposed.
1. INTRODUCTION
Additive manufacturing of materials is a fast-developing field.
Recent major advances in this area are very noteworthy. But,
several underlying phenomena are still not well understood to
properly model the process and propose process improvements
leading to the manufacturing of better quality parts. The SLS
process fabricates solid objects by means of laser radiation,
which depends on several parameters related to the laser beam:
laser power, scanning velocity and wavelength. The laser energy
input is often assumed as a heat flux with Gaussian density
distribution when computing the temperature field.
Recently, several researchers developed models of heat diffusion
for laser sintering process. However, they underestimate the
effect of the air between grains in the laser sintering process. In
fact, the powder bed is always treated as a homogeneous medium
instead of discrete granular system an assembly of many discrete
solid particles interacting with each other due to dissipative
collisions.
The Selective Laser Sintering (SLS) or selective laser melting
(SLM) is one of the most promising additive manufacturing
techniques for modern material processes. Although, the SLS
(SLM) process for thermoplastics currently finds many industrial
applications, it still leaves quite a lot to be desired concerning the
quality of manufactured parts. Comprehensive modelling of the
complex physics underlying the process is a real challenge that
if met may lead to process optimization and better quality
manufactured parts. Several rather simplified theoretical models
have been proposed in the literature. Most of the contributions
relate to thermal models, simplified thermomechanical modeling
and laser-powder interaction at both powder and part scales in
[1], with empirical Arrhenius evolution for the material
properties in [2]. Berzins et al. [3] investigate the final
microstructure of amorphous polycarbonate by a 2D numerical
approach. The powder densification and shrinkage based on the
empirical Arrhenius evolution proposed in [2] are included in the
analysis. Riedlbauer et al. [4] proposed a modeling of the process
with some experimental comparison and data. Due to the lack of