Additive Manufacturing (AM) promises previously unknown design freedom and much greater flexibility than traditional manufacturing processes. A 3D printer is able to produce a desired component directly, eliminating the need for time-consuming preparatory work such as mold-making.
Like any other manufacturing process, however, additive manufacturing also has its own particularities and fields of application. If you want to succeed with additive manufacturing, the manufacturing requirements have to be taken into account in the design process in order to be able to produce the product successfully.
Let’s have a look at the development process for successfully manufactured products using additive manufacturing:
Virtual first – In a virtual and integrated development process, all relevant design decisions can be investigated beforehand. From design and manufacturing planning to simulation of the print process, the entire process can be examined and validated virtually.
Having a completely virtual product development process adapted to additive manufacturing enables all potentials of the manufacturing method to be exploited: design flexibility, production flexibility, and all this with a minimum number of expensive test prints or misprints.
The successful concept consists of tailor-made applications for additive manufacturing in a collaborative model-based end-to-end process and thus perfectly coordinated product development steps.
But what is model-based?
Until now, separate models have been used with each department and employee working with their own models that are necessary and goal-oriented for their task: geometry model (CAD), the model for manufacturing and assembly, or the function-describing (simulation) models adding system behavior and physics modeling.
A unified data model and a collaborative platform now allow all these models to be linked together – enterprise-wide.
Based on the geometry model (form & fit), the required function models (function) can be added in varying degrees of detail to support all product decisions on the basis of virtual models. In addition, changes to the geometry can be automatically taken into account in the functional models without having to rebuild them from scratch.
Back to the model-based development process for additive manufacturing:
- New Design Practice for AM: ‘Functional Generative Design’
The new AM design freedoms call for new design approaches in order to be able to actually make use of the freedoms gained. This is where the concept of functional generative design comes into play. Algorithms and physics simulation support the designer in finding the ideal component design taking into account the functional requirements and AM specific manufacturing constraints.
For more details please have a look at the short 7 minute webinar: Functional Generative Design
- Manufacturing Planning: ‘Additive Manufacturing Planning’
The component design can be used to start the additive print process preparation. The aim is to supply the selected printing machine with the necessary parameters, settings and geometry data, based on the selected manufacturing method (here: Metal Powder Bed Fabrication). The machine programmer must specify the machine properties, position and orientate the components on the print bed (in the best possible way), define the necessary support structures and their type in order to calculate component slices and the print or laser path. Reusable rules can be defined, which can be used to record best practices and make them available to colleagues. The results can then be exported and transferred to the printing machine.
A major advantage of this approach is the CAD geometry-based generation of support structures. These can be changed and dimensioned manually by the designer and can be directly used in the physics simulation.
- Printing Process Simulation: ‘Additive Manufacturing Simulation’
Instead of printing the component directly, which can be costly and time-consuming, the printing process can be investigated virtually in advance to determine whether production is successful. The machine parameters and print paths that have already been defined can be reused directly – model-based. The component or all components on the print bed are meshed and prepared for finite element analysis. By creating the support structures as geometry, they can be meshed directly. An assistant panel also guides users with less simulation experience through the workflow so that the built-in best practices can be used to simulate the printing process. Using simulation, the resulting deformations, elasto-plastic strains and residual stresses during the production can be investigated. With an extended level of detail in the modeling, microstructures and porosity can also be virtually examined.
This provides the development engineer with a good prediction of the expected printing performance and the physical properties of the final product.
- Iterative Improvement of the Design and Printing Process: ‘As-designed’ vs. ‘As-manufactured’
As with any other manufacturing process, the produced component generally deviates (hopefully only slightly) from the originally constructed design. This can lead to faulty prints or deviating component geometries, especially in the case of 3D printing due to thermal residual stresses or even not optimally positioned or missing support structures. For example, using the selective laser sintering process, component deformations caused by heat input during production can often lead to unexpected permanent deformations.
The presented process allows a detailed comparison of the designed part with the virtually manufactured part. The ‘CATIA Reverse Shape Optimizer’ functionalities can be used to compensate the predicted deviations in the component design: The deviations calculated in the simulation are simply transferred as a vector field with a negative scaling to the original geometry. In a model-based approach, the above analysis steps are now repeated by reusing the models until a sufficient convergence of as-designed and as-manufactured has been achieved.
Other parameters that can be investigated in advance by the virtual process include the influence of different support structure geometries, user-defined or automatically generated variants. The same applies for their positioning, but also the influence of different print or laser paths. A print path adapted to the component can thus cause considerably lower residual stresses and thermal deformations…
What to learn more about Dassault Systèmes’ simulation solutions for additive manufacturing? Visit: go.3ds.com/Print2Perform
This post was originally posted on our German language blog: Puls der Innovation.
