Virtual and Physical Prototyping (Jul 2019)

Numerical and experimental assessment of the mechanical properties of 3D printed 18-Ni300 steel trabecular structures produced by Selective Laser Melting – a lean design approach

  • Franco Concli,
  • Andrea Gilioli

DOI
https://doi.org/10.1080/17452759.2019.1565596
Journal volume & issue
Vol. 14, no. 3
pp. 267 – 276

Abstract

Read online

Lattice and trabecular structures are excellent candidates for energy absorbing applications such as personal protective equipment or any sort of bumper. Additive manufacturing technologies allow more freedom in the design of new topologies such as trabecular and lattice structures overcoming the limitation of the traditional manufacturing processes. In the present research, an investigation on the ductile behaviour of additive structures is presented. In a first phase, a series of 18-Ni300 steel specimens with different geometries has been printed using the SLM (Selective Laser Melting) technology. Experimental quasi-static tests on those samples were numerically reproduced, in order to retrieve the actual stress state, quantify the plastic strain at failure and calibrate a ductile damage model. In a second phase, trabecular structures, made of the same material and processed with the same technology as the samples, were produced and experimentally tested in a compression test. Simulations including the calibrated model were used to reproduce the response of the elementary trabecular cell subjected to different loading conditions (micro-scale simulations). This kind of simulations is very time-consuming and not suitable for the design/optimisation of large structures made by thousands of elementary cells. To overcome this limitation, in a third phase of the project, an effective and efficient design methodology has been implemented. Each elementary cell is modelled as an equivalent non-linear three-dimensional spring. The force–displacement (torque–rotation) relations in different directions were obtained with the previously described micro-scale FE simulations. In this manner, the computational costs can be reduced by many orders of magnitude allowing the study of complex real systems.

Keywords