As part of the December 2020 update of Fusion 360, we delivered a hidden gift just in time for the holidays: a new method for generating additive manufacturing outcomes. Internally, we call this project Additive 2.0, and it is part of our Experimental Generative Solvers and Features preview. This new method should go a long way to improve the shape quality and printability of the outcomes compared to what you previously saw in generative design. This was a present I know many of us have been eagerly waiting for, and my team and I would like to “unwrap” everything we delivered.
Using the New Additive 2.0 Algorithm
First off, you must turn on our Experimental Generative Solvers and Features preview from your User Preferences.
With the preview enabled, you will need to turn on Alternative Outcomes in the Study Settings. From there, just set up your study as usual. Enable the additive manufacturing constraint in the Manufacturing Constraints dialog, choose the desired print orientations, and let our solvers take care of the rest.
Once in Explore, for each set of additive outcomes (where a set equals given material and direction), Additive 2.0 will generate the second outcome in each set. Quick tip: the easiest way to see this is to switch to the properties view, isolate down to the additive manufacturing method, and set the Sort By dropdown to Material.
How is Additive 2.0 Better?
Our initial approach to generating additive outcomes left a lot to be desired in shape quality in many scenarios (many of you told us that, and we listened!). It was common to see outcomes with pronounced stair-stepping and other artifacts as we tried to meet overhang requirements and be fully self-supporting. With our new approach, our solvers do a much better job of meeting minimum thickness requirements, balancing design mass, and minimizing support material. This new approach will not sacrifice shape quality to be fully self-supporting.
Let’s dig into a couple of examples that demonstrate how these improvements impact the outcomes we create.
Our first example is the simple generative setup shown below (loads and constraints are hidden). We intend to manufacture this part in the orientation shown in Figure 1 (below), replacing the red obstacle with the build plate.
Figure 1 – Design space for v-shaped tension bar example
This setup produces an unconstrained outcome, as seen in Figure 2 (below), which is difficult to manufacture for various reasons. First, a large area under the part geometry requires support structures (red highlighted regions in the second half of the figure). Also, the thin features near the center of the part are fragile and may break during support removal. While this design can be additively manufactured, it is far from optimized.
Figure 2 – Unrestricted outcome and required support area
When we start to look at Additive 2.0 side by side with our existing additive constraints (Additive 1.0), the differences become obvious, as shown in Figure 3 (below). In both cases, we configured the additive constraint with an overhang angle of 45 degrees and a minimum thickness of 3mm.
Figure 3 – Production Additive Result (1.0) vs. Experimental Preview Additive Result (2.0)
In both cases, the amount of support material required is reduced but not eliminated. However, the Additive 1.0 result, which builds material up from the build plate to support the entire structure, is 50% heavier than the unconstrained outcome. It exhibits stair-stepping artifacts that were a common annoyance with the original additive manufacturing solution. Additive 2.0 also significantly reducesContinue reading