UTILIZING DESIGN FOR METAL ADDITIVE MANUFACTURING AND TOPOLOGY OPTIMIZATION TO IMPROVE PRODUCT DESIGNS

摘要

Metal additive manufacturing has transformed the product design process by enabling the fabrication of components with complex geometries that cannot be manufactured using conventional methods. Initial designs can be further enhanced by employing topology optimization software and Design for Metal Additive Manufacturing (DFMAM) guidelines. In this study, a commercially available bicycle spider-crank was optimized for three-dimensional (3D) metal manufacturing. The 3D surface geometry of the original spider-crank was acquired using a white light scanner and used to generate a 3D solid model of the part. Boundary conditions were obtained from cycling loads found in published literature and applied to an ANSYS Finite Element Analysis (FEA) model. The FEA model was analyzed to determine the von Mises stress throughout the part. ANSYS Topology Optimization software was applied to the model. The software uses an iterative process to remove low stress material and recalculate stress within the part until no more material can be removed without exceeding a target maximum stress value. Following topology optimization. DFMAM principles were applied to enable the part to be 3D printed. Results from the FEA showed the DFMAM optimized design to be 41.5% lighter than the original design. The maximum stress increased from 41.2% of the material yield strength to 61.5% in the DFMAM optimized design, which exceeded the target optimization value of 50% yield strength. Analysis results were verified experimentally. The original design and DFMAM optimized design were printed using an EOS M 290 metal additive manufacturing machine. Parts were separated from the support structure and tested on a universal testing machine. A custom testing apparatus was designed and built to conduct the testing. Testing was performed at 15 degrees intervals throughout the range of motion. Strain gages attached to the arm of the crank were used to obtain stress values at specific locations and dial indicators were used to measure the deflection of the crank arm under load. Experimental results closely matched results obtained from the FEA, validating the model. With the model validated at specific locations, it was assumed that the stress calculated by the FEA at the critical points were also accurate. The results showed the topology optimization software to be an effective and useful tool for optimizing the design of 3D metal printed parts. However, topology optimization alone was not enough to finalize a design prior to printing. The application of DFMAM principles were needed to ensure that the overhanging structures would not collapse during printing. Because the determination of what constitutes an overhang is determined by the part orientation when printed, some modification will generally be required prior to printing. In conclusion, using a bicycle spider-crank as an example, this research has shown that the use of topology optimization software and Design for Metal Additive Manufacturing principles is able to reduce the weight of a 3D metal printed part while simultaneously achieving a maximum stress near a target value.