Innovative bionic mirror design for highly dynamic applications
Topology optimization is a powerful tool for the development of new mechanical components. New manufacturing methods and materials offer a great deal of scope for optimizing scanner mirrors. These optical elements for highly dynamic beam deflection applications can be significantly improved through simulations.
LIGHTWEIGHT MIRROR SUBSTRATE MADE FROM ALUMINUM
The use of aluminum-silicon alloys (AlSi) as a mirror substrate has already been investigated in a research project sponsored by the German Federal Ministry of Education and Research. Current solutions for highly dynamic applications are based on beryllium or ceramics. These materials are associated with very high costs, however, on account of the toxic properties of beryllium oxide, and the high hardness and processing limits of ceramics.
Aluminum-silicon substrates can be manufactured using conventional methods, and enable mirrors of lightweight construction to be produced. A weight-reduced mirror pair is shown in Figure 1.
MAXIMUM PERFORMANCE THANKS TO SIMULATION
The R&D department at ARGES GmbH has been analyzing mechanical components using the finite element method (FEM) for a number of years now. A broad range of simulation techniques are employed for this. On the one hand, individual components are analyzed based on specific loads and boundary conditions. The main focus in this case is on linear, dynamic, and static simulations, while non-linear, dynamic simulations, deliver insights into the extremely fast movements of the mirror systems. On the other hand, processes such as laser hardening of steels are simulated in order to determine the optimum process parameters.
When developing new opto-mechanical components, it is important to come up with a robust and efficient design. The finite element method (FEM) in conjunction with topology optimization provides an automated, simulation-supported means of improving components. After specifying the boundary conditions, loads, and manufacturing constraints, the software then performs the iterative optimization process. Material is successively removed from those areas that are not subjected to internal forces. At the end of the optimization process, a smooth surface is generated that represents the ideal distribution of material.
NOT A ONE-SIZE-FITS-ALL SIMULATION
Topology optimization relies on the correct application of boundary conditions, loads and design limits. To verify the optimization algorithm, simulations are performed on minimal examples.
As a first step, the mirror model is approximated as a flat, twodimensional and symmetrical plate. The corresponding threequarter section view is shown in Figure 2. The plate is restrained so that it corresponds to the plane of symmetry of the mirror. The reflective surface (gray) is excluded from the design space (blue).
The two-dimensional optimization is performed for various load cases. These include a force F (at the external edge of the mirror), a pressure p (acting on the entire reflective surface), an acceleration g (perpendicular to the reflective surface), and an angular acceleration φ about the pivot point of the mirror. These models are also illustrated in Table 3.
It is apparent that different optimal support structures are obtained depending on the prescribed objectives and constraints, such as mass or inertia. To be able to also correctly optimize three-dimensional structures, interpretation of the results and verification of plausibility are essential.
3D TOPOLOGY OPTIMIZATION AND ADDITIVE MANUFACTURING
To obtain a good solution for three-dimensional cases as well, the design space (cf. Figure 2, transparent volume) must be modeled in very fine detail. This means a large number of discrete calculation points or element nodes in the simulation.
A typical simulation may have up to 1.5 million degrees of freedom (DOFs). Convergence is achieved with the sensitivity-based optimization algorithm after approx. 80 calculation iterations. These simulations are clearly very resource intensive.
Figure 4 shows an initial design study that has already been produced with additive manufacturing method.
In the further course of the project, the single optimization goal – maximum stiffness – used thus far will be extended by further goals. This includes the various loads that arise during the manufacturing process, for example the shear stresses resulting from the application of the highly reflective coating system onto the reflective surface. Furthermore, the thermal deformations due to absorption of the laser beam during a highly dynamic application will be investigated. The objective in this case is to already reduce the optical surface deformation during the design phase.
The optimized design for the additively manufactured mirror will also be compared and contrasted with currently available substrates.
Additive manufacturing opens up a large spectrum of possibilities. Since the dynamic requirements on the scan system also define the loads acting on the mirror substrate, a high level of individuality can be achieved using this novel manufacturing technology. This makes it possible to implement very specific customer requirements.
This research and development project is being sponsored by the German Federal Ministry of Education and Research (BMBF) under the thematic area “Additive manufacturing – individualized products, complex mass products, innovative materials (ProMat_3D)”. The BMBF has commissioned the project manager Karlsruhe Institute of Technology (KIT) to coordinate this funding programme.