Implicit Slicing for Functionally Tailored Additive Manufacturing

Abstract

One crucial component of the additive manufacturing software toolchain is a class of geometric algorithms known as “slicers.” The purpose of the slicer is to compute a parametric toolpath defined at the mesoscale and associated g-code commands, which direct an additive manufacturing system to produce a physical realization of a three-dimensional input model. Existing slicing algorithms operate by application of geometric transformations upon the input geometry in order to produce the toolpath. In this paper we introduce an implicit slicing algorithm that computes mesoscale toolpaths from the level sets of heuristics-based or physics-based fields defined over the input geometry. This enables computationally efficient slicing of arbitrarily complex geometries in a straight forward fashion. The calculation of component “infill” is explored, as a process control parameter, due to its strong influence on the produced component’s functional performance. Several examples of the application of the proposed implicit slicer are presented. It is demonstrated — via proper experimentation — that the implicit slicer can produce a mesoscale structure leading to objects of superior functional performance such as greatly increased stiffness and ultimate strength without an increase of mass. We conclude with remarks regarding the strengths of the implicit approach relative to existing explicit approaches, and discuss future work required in order to extend the methodology.

Homogenization of material properties in additively manufactured structures

“…Infill A Infill B Linear Measured modulus [10] 1 Even though our predicted modulus for infill A is higher than the measured result, it agrees with the stress -strain curve shown in Figure 8, in which the tensile moduli for Infill A and B are much closer.…”

“…Direct simulation of the design model is likely to yield poor results due to the various inconsistencies between the design model and the printed part. The proposed method is verified with the physical testing result in [10] on the right branch in Section 5.…”

“…We also compare the predicted homogenization results with the physically measured effective tensile modulus reported in [10]. The samples for the physical testing are shown in Figure 6.…”

“…The proposed printed model does not preserve the volume of the filament and underestimate the volume of the material in printed part with overfills. Since the G-code generator used in [10] is custom-built to study the effects of different infill patterns, the overfill is less likely to happen in most of the established process planning procedures. We also observed that the measured modulus of Infill A is around the midpoint of Infill B and linear infill.…”

“…Table 1: The measured and homogenized tensile modulus by the proposed method. The first row shows the measured modulus (GPa) in [10]. The rest of the rows show the simulated modulus with Ex = 2.25 GPa and different stiffness in the transverse direction.…”