Predicting and Validating Multiple Defects in Metal Casting Processes Using an Integrated Computational Materials Engineering Approach Dissertation Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By Yan Lu Graduate Program in Materials Science and Engineering The Ohio State University 2019 Dissertation Committee: Professor Alan A. Luo, Advisor Professor Glenn S. Daehn Professor Wei Zhang Copyright by Yan Lu 2019 Abstract Metal casting is a manufacturing process of solidifying molten metal in a mold to make a product with a desired shape. Based on its own unique fabrication benefits, it is one of the most widely used manufacturing processes to economically produce parts with complex geometries in modern industry, especially for transportation and heavy equipment industries where mass production is needed. However, various types of defects typically exist in the as-cast components during the casting processes, which may make it difficult for post-processing and limit the service life and further application of products. It becomes imperative to analyze the processes in actual manufacturing conditions to predict and prevent those casting defects. Since it can be quite time consuming and costly to assess the processes experimentally, a computer-aided approach is highly desirable for product development and process optimization. In recent decades, computer-aided engineering (CAE) techniques have been rapidly developed to simulate different casting processes, which have great benefits to tackle casting defects in a more practical and efficient way. This work focuses on using ProCAST®, a finite element analysis (FEA) software, together with other necessary simulation and modeling techniques, including Computer-Aided Design (CAD), Calculation of Phase Diagrams (CALPHAD) and Cellular Automaton (CA), to study relevant defects in actual metal casting foundries. Specifically, three different cases have been mainly investigated, including (i) veining defect caused by thermal cracking in resin-bonded silica sand molds/inserts for sand casting process; (ii) thermal fatigue cracking in H13 steel dies/inserts for high pressure die ii casting process; and (iii) Hydrogen-induced gas porosity in A356 castings for gravity casting process with permanent molds. For each case, CAD model was designed and FEA model was constructed with validated materials database based on CALPHAD simulation, experiment tests and/or literature references. Coupled calculations of heat transfer, fluid flow for mold filling, and/or stresses and strains were run to obtain thermal and structural data for subsequent defects analyses and predictions. More importantly, key experiments at laboratory scale were designed and performed to reproduce those defects. Test results were employed to correlate and validate the predictions from simulation. The highlight of this dissertation is that an improved model and/or prediction criterion is proposed for each defect case and is dedicated to engineering applications, including (i) a statistics-based cracking criterion of resin-bonded silica sand molds or inserts in casting processes; (ii) a temperature-based fatigue life prediction criterion for thermally-induced cracking in H13 steel dies for die casting; and (iii) a coupled CA-FE model for location-specific prediction of gas porosity in A356 gravity castings with permanent molds. This research is aiming at demonstrating that the integration of different CAE techniques and key experimental validations can help tackle the defects in various casting processes in a time-efficient and cost-effective manner. The results and the approach may be of great benefits to casting engineers for defect assessments and design optimizations in different casting processes. iii Dedicated to my beloved parents, For their unconditional love, unwavering support and endless encouragement. iv Acknowledgements I must first express my deepest appreciation to my advisor, Prof. Alan A. Luo, for his continuous support and overarching guidance during the past five years. My graduate study and research were fully funded by industrial projects owing to his remarkable experience in industry. I would also like to thank him for encouraging me to explore into different research projects and for helping me grow as an independent researcher. I would like to thank Prof. Wei Zhang and Prof. Glenn S. Daehn for serving as my committee members starting from my candidacy exam to my dissertation overview and to my dissertation defense. I also acknowledge Prof. Anthony F. Luscher for being the faculty representative in my dissertation defense. I am truly grateful for their valuable comments and constructive suggestions to make this work more complete. I would also like to acknowledge Mr. Keith Ripplinger from Honda of America Manufacturing, Inc. and Mr. Duane Detwiler from Honda R&D Americas, Inc. for their kind help and collaboration. Their friendly inputs and discussions in the regular project update meetings help improve this work. I must also extend my sincere thanks to each and every colleague in the group for their great amount of support. I had the great pleasure of working with and learning from them, including Dr. Weihua Sun, Dr. Renhai Shi, Dr. Jiashi Miao, Andrew Klarner, Scott Sutton, Zhi Liang, Emre Cinkilic, Xuejun Huang, Colin Ridgeway, Janet Meier and Thomas Avey. Specifically, thanks to Dr. Huimin Wang for helping get started on my research here. Many thanks to Mr. Geoffrey Taber, a truly skilled artisan, for helping me build a key test unit in my research. Also many thanks to Dr. Cheng Gu for sharing his expertise in Cellular Automaton to help with my research. v I would also like to acknowledge SIMCenter at OSU for providing supercomputer resources to run the simulation and for offering access to quite a few softwares for my research purpose. I also acknowledge the technical assistance from Ross Baldwin, Steve Bright, John (Pete) Gosser and Kenneth Kushner for their expertise and help in some parts of my experiments. Last but not least, I am deeply indebted to my parents for their unconditional love. They have always supported me and encouraged me to be brave and confident, no matter the path in life I choose. I also feel so lucky to meet my beloved girlfriend here at OSU, who has shared my joys and pains throughout my PhD life. For her constant support and dedication as always by my side, I thank her. vi Vita 2010…………………………………………...Shanghai High School 2014…………………………………………...B.E. Materials Science and Engineering, Shanghai Jiao Tong University 2014 to present..................................................Graduate Research Associate, Department of Materials Science and Engineering, The Ohio State University Publications 1. Y. Lu, K. Ripplinger, X. Huang, Y. Mao, D. Detwiler, A.A. Luo, “A New Fatigue Life Model for Thermally-Induced Cracking in H13 Steel Dies for Die Casting”, Journal of Materials Processing Technology, September 2019, vol. 271, 444-454. 2. G. Cheng, Y. Lu, E. Cinkilic, J. Miao, A.D. Klarner, X. Yan, A.A. Luo, “Predicting Grain Structure in High Pressure Die Casting of Aluminum Alloys: A Coupled Cellular Automaton and Process Model”, Computational Materials Science, April 2019, vol. 161, 64-75. 3. E. Cinkilic, A.D. Klarner, Y. Lu, J. Brevick, A.A. Luo, M. Zolnowski, X. Yan, K. Sadayappan, G. Birsan, “High Integrity Structural Aluminum Castings Produced with Vacuum High Pressure Die Casting”, Die Casting Engineer, November 2018, 24-28. 4. Y. Lu, A.A. Luo, K. Ripplinger, D. Detwiler, “Simulation and Experimental Evaluation of H13 Steel Thermal Fatigue Life in Die Casting”, North American Die Casting Association Transactions, October 2018, T18-011. vii 5. A.D. Klarner, E. Cinkilic, Y. Lu, J. Brevick, A.A. Luo, J. Shah, M. Zolnowski, X. Yan, “A New Fluidity Die for Castability Evaluation of High Pressure Die Cast Alloys”, North American Die Casting Association Transactions, September 2017, T17-101. 6. Y. Lu, H. Wang, K. Ripplinger, A.A. Luo, “Process Simulation and Experimental Validation of Resin-Bonded Silica Sand Mold Casting”, American Foundry Society Transactions, 2017, vol. 125, 215-220. 7. H. Wang, Y. Lu, K. Ripplinger, D. Detwiler, A.A. Luo, “A Statistics-Based Cracking Criterion of Resin-Bonded Silica Sand for Casting Process Simulation”, Metallurgical and Materials Transactions B, February 2017, 48(1), 260-267. Fields of Study Major Field: Materials Science and Engineering viii Table of Contents Abstract .......................................................................................................................... ii Acknowledgements .........................................................................................................v Vita .............................................................................................................................. vii List of Tables............................................................................................................... xiii List of Figures ..............................................................................................................xiv Chapter 1: Introduction ....................................................................................................1 1.1 Metal Casting .........................................................................................................1 1.2 Why Modeling? ......................................................................................................2