3D printing and topology optimization

This is worth looking at, if you want a product that is lighter but retains the same strength.

Below, the Danish Technological Institute 's specialist in topology optimization, Malthe Gissel will enlighten us on what computer-generated design is and how, with 3D printing and topology optimization, a product can be made lighter, saving material in the production process.

When optimizing something, it is often made smaller. It is a mathematical task where calculations are made on the component based on specific criteria such as load and material. A limit is set on how much material can be removed while the component still retains the desired strength.

In the initial setup, the objective function will be minimum compliance, which corresponds to maximum stiffness. Without a constraint, the best solution is to keep all the material—which is not very optimal. Therefore, a constraint is added on the volume fraction, for example, 20% of the original volume. This constraint produces the solution that is most rigid, i.e., stiffest, with only 20% of the material, and thus an optimized design. Alternatively, if the maximum allowable deflection is known, this can also be input as a constraint, and then the volume is minimized to get the smallest volume, and thus mass, necessary to meet the deflection requirement.

With topology optimization, one often starts with the original component. The component thus serves as the "design space," and it is the area within which the new design can fit. This ensures that the new design does not unintentionally collide with other components in the machine or application where it is to be mounted.

To perform the calculations, a mathematical formulation is used where the component is divided into small elements. This is called a mesh. It resembles a fine-mesh fishing net that is "pulled" over the component, thus dividing it into tiny parts, often several hundred thousand or millions of elements. Simple equilibrium equations can be set up between all the elements, which the computer can calculate—even if the geometry is complex. In this way, the software performs an iterative process, calculating, removing material, calculating, removing material—and perhaps adding a little if too much has been removed.

Subsequently, an FE analysis (Finite Element Analysis) is conducted to validate that the new geometry meets the requirements for stresses or deflections.

Technology is useful in many contexts. In the food industry, robots are used that move very quickly to efficiently pick and pack goods. This applies, for example, to a robotic arm that needs to pick and pack 200g chicken fillets.

It is crucial that the robotic arm is light so it can move quickly and reach the chicken fillets as they move down the conveyor belt. Here, topology optimization ensures that the robotic arm has the least possible weight while still being able to perform the work it is designed for. At the same time, there may be requirements for the component's natural frequency or, for example, the center of mass, all of which can be included in the optimization setup.

Another example of an industry where the technology is used is in the manufacture of parts for drones and airplanes. Weight significantly affects how much energy is needed to keep the drone or airplane aloft. Therefore, it makes a lot of sense to look at how parts can be made lighter using topology optimization.

At the Danish Technological Institute, there has been a collaboration with the Danish startup Airflight ApS , which produces drones for transporting parts for the wind turbine industry. As part of the design optimization project DfAM (Design for Additive Manufacturing) facilitated by the Danish AM Hub , Airflight has 3D printed a bracket for the drone in titanium, successfully reducing the weight by as much as 67% by design optimizing the brackets. This led to a weight reduction of 11 kg per drone.

Topology optimization software has been crucial in ensuring that the bracket could meet all properties—lightweight, strength, and size. This is possible because the FEM analysis and simulation can modify the design repeatedly until the most optimal design is achieved, which can still lift the same heavy components and tools for wind turbines.

3D printing is a mature production method that many companies have realized over the past 20 years. However, there are still companies that have not yet explored the possibilities within 3D printing. In this context, I believe that topology optimization can help companies see the advantage of lightweighting and redesigning their products. It not only saves material in production but also optimizes the fundamental design of the component. This can be a significant economic advantage, as in large-scale production, a lot of material is saved, resulting in a more optimal design.

Do you have components you can imagine could be 3D printed? And what possibilities do you see with topology optimization?