by Malin Storkamp and Therese Sandberg, students at Linköping University

The rise of Additive Manufacturing (AM) in industrial production brings along a need for new ways to design and adapt products to fully take advantage of the unique capabilities of AM; among them the fewer limitations imposed by the manufacturing process. Together with AMEXCI, who helps pave the way for industrial adoption of AM technologies, we wanted to explore tools for utilizing this greater design freedom and to find a procedure that effectively integrates the tools in the design process.

Topology Optimization (TO) is one such tool. TO is a mathematical method for optimizing material distribution within a set design space with the aim to generate lightweight and high-performance structures. The results often take shapes that would be hard (if not impossible) to produce using traditional manufacturing methods. The unification of AM and TO has the potential to make the most of their respective strengths.

The goal of this thesis has therefore been to investigate and develop a methodology for redesigning components for AM using TO, and to demonstrate this on an existing component. The work consists of an initial literature study, the development of the redesign process (see below) and the case study involving the implementation of the process. The software used in this thesis for TO is Siemens NX Topology Optimization for Designers.

The redesign process was drawn up based on the insights from the literature study regarding i.a. Design for Additive Manufacturing (DfAM), the workflow for TO and validation using Finite Element Analysis (FEA) and tools for analysing DfAM aspects. Some of the main steps include analysing the current design and context to determine the suitability for AM and using FEA on the initial design for later comparison. With the insights from the initial evaluation, the TO is set up. For the software used in this project this includes drawing and designating design volumes in CAD; the design space that the optimization will work inside and the frozen features that should remain the same. On these loads and constraints are imposed, to e.g. create self-supporting structures for AM. The optimization is then run with different parameters to generate optimized shapes for the case component. These shapes are then evaluated and adjusted for further improvement or aesthetic appeal. The promising designs are verified using structural analysis and DfAM analysis tools.

The process was implemented on an Autonomous Underwater Vehicle (AUV) hull component, courtesy of Saab, with the goal to reduce weight and adjust the design for manufacturing with SLM (some pictures from the process are included below) without sacrificing structural functionality. For this specific part, many features were important and could not be changed. Even then, we were able to reduce the weight 25% with the proposed method, and give some suggestions for further weight reduction that could not be structurally verified within the project due to lacking computer capability (the project was completed working remotely) and time limitations.

1. High stress areas identified in the initial design
2. New inner channels to avoid trapped supports
3. An iteration of the design space
4. Some generated TO results using different setups

In conclusion, the case study results with the improved design verified that TO is a suitable tool for redesign for AM and that the process used in this study is effective.