Probabilistic modeling of solidification grain structure in investment castings

Conference · Sun Dec 31 04:00:00 EST 1995

OSTI ID:227777

Upadhya, G K; Yu, K O; Layton, M A; Paul, A J ^[1]

---

1. Concurrent Technologies Corp., Johnstown, PA (United States)

A probabilistic approach for modeling the evolution of grain structure in investment castings has been developed. The approach differs from the classical Monte Carlo simulations of microstructural evolution in that it uses the results from a heat transfer simulation of the investment casting process for determining the probabilities of nucleation and growth. The model was used to predict the solidification grain structure in castings. The model is quasi-3D, since it uses the information from a 3D simulation of heat transfer to predict the grain structure developed in any 2D-section of the casting. Structural transitions such as columnar/equiaxed transition can also be predicted, using suitable transition criteria. Results from the model have been validated by comparison with actual micrographs from experimental investment castings. In the first case, simulations were performed for a simple plate shaped casting of superalloy Rene 77. The effects of mold insulation as well as metal pour and mold preheat temperatures on the grain size of the casting were studied. In the second example, the casting of a complex-shaped jet engine component made of superalloy IN718 was simulated. Simulation results were seen to match very well with experiments.

• OSTI ID: 227777
• Report Number(s): CONF-9509118--; ISBN 0-87339-297-3
• Country of Publication: United States
• Language: English

Related Subjects

36 MATERIALS SCIENCE
99 GENERAL AND MISCELLANEOUS
COMPUTERIZED SIMULATION
DENDRITES
FINITE DIFFERENCE METHOD
GRAIN GROWTH
HEAT RESISTING ALLOYS
HEAT TRANSFER
INCONEL 718
MATHEMATICAL MODELS
METALS
MICROSTRUCTURE
NUCLEATION
ORIENTATION
SOLIDIFICATION
TEMPERATURE GRADIENTS
VALIDATION