Return to search

Multiphysics modeling and statistical process optimization of the scanning laser epitaxy process applied to additive manufacturing of turbine engine hot-section superalloy components

Scanning Laser Epitaxy (SLE) is a new laser-based layer-by-layer generative manufacturing technology being developed in the Direct Digital Manufacturing Laboratory at Georgia Tech. SLE allows creation of geometrically complex three-dimensional components with as-desired microstructure through controlled melting and solidification of stationary metal-alloy powder placed on top of like-chemistry substrates. The proposed research seeks to garner knowledge about the fundamental physics of SLE through simulation-based studies and apply this knowledge for hot section turbine component repair and ultimately extend the process capability to enable one-step manufacture of complex gas turbine components. Prior methods of repair specifically for hot-section Ni-base superalloys have shown limited success, failed to consistently maintain epitaxy in the repaired part and suffered from several mechanical and metallurgical defects. The use of a fine focused laser beam, close thermal control and overlapping raster scan pattern allows SLE to perform significantly better on a range of so-called “non-weldable” Ni-base superalloys. The process capability is expanded further through closed-loop feedback control of melt pool temperature using an infra-red thermal camera. The process produces dense, crack-free and epitaxial deposit for single-crystal (SX) (CMSX4), equiaxed (René-80, IN 100) and directionally solidified (DS) (René-142) Ni-based superalloys.
However, to enable consistent and repeatable production of defect-free parts and future commercial implementation of the technology several concerns related to process capabilities and fundamental physics need to be addressed. To explore the process capability, the fabricated components are characterized in terms of several geometrical, mechanical and metallurgical parameters. An active-contour based image analysis technique has been developed to obtain several microstructural responses from the optical metallography of sample cross-sections and the process goes through continuous improvement through optimization of the process parameters through subsequent design of experiments. The simulation-based study is aimed at developing a multiphysics model that captures the fundamental physics of the fabrication process and allows the generation of constitutive equations for microstructural transitions and properties. For this purpose, a computational fluid dynamics (CFD) finite-volume solver is used to model the melting and solidification process. The development work also focuses on studying process response to different superalloy materials and implementing a multivariate statistical process control that allows efficient management and optimization of the design parameter space. In contrast to the prior work on single-bead laser scan, the model incorporates the raster scan pattern in SLE and the temperature dependent local property variations. The model is validated through thermal imaging data. The flow-thermal model is further tied to an empirical microstructural model through the active-contour based optical image analysis technique, which enables the identification of several microstructural transitions for laser beam describing a raster scan pattern.
The CFD model can effectively be coupled with finite element solver to assess the stress and deformation and can be coupled with meso-scale models (Cellular Automata) to predict different microstructural evolutions. The research thus allows extending the SLE process to different superalloy materials, performs statistical monitoring of the process, and studies the fundamental physics of the process to enable formulation of constitutive relations for use in closed-loop feedback control; thus imparting ground breaking capability to SLE to fabricate superalloy components with as-desired microstructures.

Identiferoai:union.ndltd.org:GATECH/oai:smartech.gatech.edu:1853/54294
Date07 January 2016
CreatorsAcharya, Ranadip
ContributorsDas, Suman
PublisherGeorgia Institute of Technology
Source SetsGeorgia Tech Electronic Thesis and Dissertation Archive
Languageen_US
Detected LanguageEnglish
TypeDissertation
Formatapplication/pdf

Page generated in 0.0026 seconds