SBIR-STTR Award

Additive manufacturing (AM) technology offers the potential to fabricate complex geometries that cannot be realized using conventional subtractive methods. In current industrial AM processes, support structure is typically needed for complex geometries that contain overhangs. Determination of the type and distribution of support structure is often based solely on technician experience. This process is far from optimal and does not take into account thermally-induced residual stress buildup and subsequent part distortion that can occur when the support material is removed. Hence significant non-recurring engineering costs are incurred to develop a repeatable AM fabrication process for production. Currently, a computational tool that is capable of optimizing the support structure design to minimize thermally-induced residual stress or distortion does not exist. The key challenge to the development of an efficient topology optimization tool for support structures is the lack of an ultra-fast method for predicting residual stress and distortion in an AM part. Materials Sciences Corporation with team with the University of Pittsburgh to develop a computational tool cable of simulating the residual stress and distortion of a realistic AM part in minutes for use in developing optimal support structure by employing a numerically efficient finite element (FE) based process simulation model.

Additive manufacturing (AM) technology offers the potential to fabricate complex geometries that cannot be realized using conventional subtractive methods. In current industrial AM processes, support structure is typically needed for complex geometries that contain overhangs. Determination of the type and distribution of support structure is often based solely on technician experience with a focus on maintaining structural stability throughout the build process. This process is far from optimal and does not take into account thermally-induced residual stress buildup and subsequent part distortion that can occur when the support material is removed. Directed Energy Deposition (DED) additive manufacturing (AM) utilizes a high energy beam source such as a laser to create a melt pool on the part while feedstock material in powder or wire form is fed into the melt pool. The material deposition rate for a DED process is 10-100 times faster than for a Laser Powder Bed Fusion (LPBF) process, e.g., Selective Laser Melting (SLM). Therefore, heat and residual thermal stress accumulation in the build during DED processing is a much greater issue than during LPBF processing. For example, the temperature rise in each layer would cause the build to be inconsistent in dimension and mechanical properties. Excessive residual stress induced by the process would cause the build plate to deform severely after removal from the build platform, which would make post-machining very difficult or sometimes even impossible for large parts. Computational modeling can be utilized to overcome these issues by drastically reducing the number of experiments needed to optimize the process. Materials Sciences LLC and the University of Pittsburgh will build on their recent advances in fast process simulation based on the modified inherent strain method and topology optimization enabled prediction of residual distortion at part-scale for DED to accelerate insertion of this AM technology in missile and aviation product development efforts.