A new Discrete Element Method framework to stimulate 3D printing

3D printers are becoming more known. These desktop printers use mostly plastic filaments to print whatever the user draws up. Industry also picked up on this trend and the promises of 3D printing, or Additive Manufacturing (AM). There are many AM methods which are able to print products with many different materials. The technology has made a lot of progress over the recent years, which enabled the printing of structural metal parts. Various industries, such as automotive, aerospace, medical, electrical, energy, high tech and food, already make use of AM in their production chain. AM enables complex shapes, mass personalization, shorter design cycle, improved sustainability and enhanced properties. The framework Bram Dorussen developed in his PhD research can provide insight into laser-powder bed interaction, other laser shapes, powder degradation, and laser polarization. Experiments reveal the relationships between printer settings and product properties.

Despite all these benefits, AM is not yet widely adapted due to the complexity of the process. Products may fail due to large residual stresses and have heterogeneous properties, while it is also difficult to determine the optimal process parameters and repeating the same print might yield different results. Therefore it is essential to better understand the AM processes and the relationship between printer settings and printed products. To this end, numerical tools can be employed to study the fundamental physics of an Additive manufacturing process. Various simulations can be executed in parallel within a short timeframe and at relatively low cost, compared to doing experiments.

From powder to product

A Discrete Element Method (DEM) framework is developed to capture the complete Powder Bed Fusion-Laser Beam printing process, from powder to product. The powder particles are represented by spherical elements which interact with each other. The interaction between two particles is both mechanical and thermodynamic, and can represent a powder or solid volume. Particles bond to represent a solid when they are in contact and the critical bonding temperature is achieved.  Powder can also interact with the boundaries of the printer. The laser-powder bed interaction is modelled in detail with a ray tracer, where numerous rays are traced throughout the powder bed. The printing simulation is controlled by a G-code file, prescribing the laser path, depositing new powder layers and setting the process parameters.

Capabilities of the framework

Single scan line experiments are performed to assess the capabilities of the DEM framework. These are executed for single steel samples, as well as multi-material steel and copper samples. A single scan line with different printer settings is lasered on a cuboid substrate. The dimensions of the single scan lines are determined and compared to numerical predictions. Microscopic images of the cross-sections of the single scan lines are analysed. Next to this, various (numerical) case-studies are performed to demonstrate the DEM framework capabilities and gain new insights: the printing of an H-shape, the effect of laser polarisations, the effect of powder degradation and a multi scan line experiment.

Title of PhD thesis: A Particle Based Numerical Analysis of Metal Laser Powder Bed Additive Manufacturing From Powder to Product. Supervisors: Joris Remmers and Marc Geers.