Until recently, designing lattice structures was a long, difficult process. Conventional design tools don’t allow enough flexibility, and conventional analysis & optimization methods have struggled to integrate well into engineering workflow. But with a new collaboration between Abaqus and nTopology, lattice design, analysis, and optimization is a seamless and repeatable process.
TOPOLOGY
Lattice design begins with topology—the locations and connectivity of nodes and beams in the structure. Topology design determines the load paths and structural rigidity of the design. Lattice topologies can be generated by a number of methods—some periodic/repeating and some aperiodic/stochastic. Regardless of the topology generation method, we begin by understanding our design space. By using Abaqus to analyze the solid region that we’ll be generating a lattice in, we can create a topology that is effective and efficient (see figure 1).
Using this result, we can use nTopology Element to design a topology. Stochastic topologies will vary beam density based on Abaqus field outputs, whereas periodic topologies will use different lattice cells in different regions of the part. Here, we’ll use variable periodic topologies (all based on a hex prism cell) to create regions with different properties in our part (see figure 2).
Once we have a topology, we can reanalyze the part to understand how that topology will perform. Using nTopology Element to export an Abaqus input file, apply a default beam thickness based on the 3D printing process. Finally, run a simple beam analysis on the part (see figure 3). If necessary, we can use these analysis results to modify the topology design. Otherwise, we can use Tosca to optimize beam sizes. We limit the beam thickness to the printable range of our printing process, staying thick enough to print successfully and thin enough to not require support structures. Even with tens of thousands of beams, these optimizations run very quickly. We can be confident that the results are efficient and reliable because we use Tosca’s time-proven optimization methods (see figure 4). Once the optimization has run and we’re happy with the results, we export the design back out as an LTCX file and bring it into nTopology Element. At this stage, we can make any necessary design edits before converting the part into a printable mesh (3MF or STL are preferred) (see figure 5).
Through this process we can create parts that have extremely high strength to weight ratios, efficient mass distributions, and variable mechanical properties (stiffness/compliance, etc). Whether you’re designing for aerospace, medical implants, or consumer tech, this workflow can be tuned to meet the key metrics that your customers require. Regardless of your specific goals, the workflow is smooth and flexible. Learn more about nTopology Element , Abaqus and Tosca.
ABOUT THE AUTHOR
Spencer Wright manages integrations & R&D at nTopology in New York City. He writes about additive manufacturing at pencerw.com and about the engineering industry at theprepared.org.
For More Information, visit: www.ntopology.com
This article was originally published in the September 2017 issue of SIMULIA Community News magazine.
