ORIGINAL ARTICLE Mechanical characterization and quality of iron castings using optimized mold design: simulations and experimental validation Muhammad Azhar Ali Khan 1 & Anwar Khalil Sheikh 1 Received: 6 February 2018 /Accepted: 5 June 2018 # Springer-Verlag London Ltd., part of Springer Nature 2018 Abstract This paper presents a new approach to analyze the quality of ductile iron castings through simulations and experiments. Standard tensile test specimens are considered as simple cast products for which a multi-cavity mold is designed, simulated, and optimized to minimize porosity using MAGMASoft. X-ray imaging, hardness measurement, and tensile testing are done for selected specimens produced using optimized mold design. Next, finite element simulation of tensile testing until fracture is done in ABAQUS using elastic-plastic material model and porous metal plasticity model. Simulation results for sound specimen are found to be in good agreement with the experimental results. Since mold design optimization is solely based on porosity minimization, no porosity is observed in the final mold design. However, if multi-criteria optimization of mold is done, the specimens may show some porosity which can be integrated in the developed finite element model of tensile testing. It is concluded that simulation-based mold design optimization can produce nearly defect-free castings and at the same time exhibit the similar mechanical properties as their sound counterparts produced with other manufacturing processes. Keywords Metal casting . Mold design . Simulation . Optimization . Quality 1 Introduction Ductile iron has always remained a material of interest in metal casting. It is extensively used in engineering applica- tions owing to its exceptional castability and excellent me- chanical properties. The mechanical properties of ductile iron castings are a function of the microstructure developed and the defects produced during casting process. Major defects which adversely affect the mechanical properties are often related to inclusions, shrinkage porosities, and the shape and dimen- sions of graphite particles in thermal center of castings where the solidification is delayed. Such defects can be minimized through controlled chemical composition, a robust mold de- sign, and carefully selected process parameters. In pursuit of high-quality ductile iron castings, a physical trial-and-error method adds unnecessary cost and time to product develop- ment. Moreover, the rigs for experimental testing of the cast prototypes, which reflects their real-time conditions, are diffi- cult to design and sometimes are not economically viable. On the other hand, simulation-based casting allows method engi- neers to evaluate and thus develop a robust casting design through a proof-of-concept approach. Next, the quality and performance of these castings (with defects predicted in cast- ing simulations) can be analyzed in a virtual domain using commercially available engineering software before they ac- tually put into service. Previous studies on ductile iron castings suggest that most of work is done to improve the casting design using simula- tions [1–3]. Sun et al. [2] worked on the improvement in casting design of a truck rear axle made from nodular cast iron. Mold filling and solidification simulations were done using a 3D FDM (finite difference method) solidification model and the Z-CAST simulation software. Defects in cast- ing were analyzed based on temperature field and flow front progression within the mold. A new pouring system was de- veloped where the height and length of the ingate were in- creased and decreased respectively, the runner was elongated in horizontal direction, the diameter at sprue bottom was in- creased, and the position of the feeder was altered. With the * Muhammad Azhar Ali Khan azharali@kfupm.edu.sa Anwar Khalil Sheikh anwarks@kfupm.edu.sa 1 Department of Mechanical Engineering, King Fahd University of Petroleum and Minerals (KFUPM), Dhahran, Saudi Arabia The International Journal of Advanced Manufacturing Technology https://doi.org/10.1007/s00170-018-2325-y