|About this Abstract
||Liquid Metal Processing & Casting Conference (LMPC 2019)
||Liquid Metal Processing & Casting
||Uncertainty in the Modeling of Nitinol Solidification in VAR
||Kyle Fezi, Matthew J.M. Krane
|On-Site Speaker (Planned)
Nitinol’s unique shape memory and super-elastic properties make it desirable for many biomedical, automotive, and aerospace applications, but are highly dependent on the Ni/Ti ratio and other outcomes of solidification behavior in VAR. A cost-efficient method to explore the processing parameter space is numerical modeling, but simulation results reliability is dependent on uncertainty in the model and the input data. The VAR process is modeled using the commercially available MeltFlow-VAR package. The sensitivity of the sump depth prediction to several input parameters, primarily material properties and boundary conditions, were analyzed. Level one sensitivity analysis determined which inputs had the most impact and level two analysis quantified sump depth uncertainty due to those parameters. Uncertainty in the thermal boundary condition on the ingot’s sidewall was found to have a large influence on the sump depth, providing justification for an experimental program to more accurately measure heat loss during the process.
||Definite: At-meeting proceedings