Experimental validation of computer simulations of ceramic core injection processes

In this study, a computer simulation of a three-dimensional transient ceramic injection molding (CIM) process, developed by the Advanced Casting Laboratory at the University of Tennessee for Howmet Research Corporation, was experimentally validated. Experiments were designed and conducted to measure temperature of ceramic core material as a function of time for both filling and packing stages of the injection molding process. In addition to fluid temperature, die wall temperature data were also collected for this analysis. The results were used to validate the CIM model to insure that ProCAST simulation is realistic and accurate.

The experiments were conducted under production conditions at Howmet Casting Support manufacturing facility in Morristown, Tennessee. The temperature data were taken at both 30-milliseconds and 60-milliseconds sampling rates. A total of twenty experimental runs for each sampling rate were conducted. An error band was developed from the results of the experiments. The temperature data are presented here in non- dimensionalized form so that they can be used for validation of computer modeling simulations under various initial and ambient conditions.

The results indicate that the average heat transfer coefficients (h) in the range of 1,800 W/m² K to 2,200 W/m² K best represent experimentally observed thermal characteristics of the filling and solidifying process. The flow patterns observed from short shot pictures are best approximated by flow simulation patterns using isothermal wall boundary conditionAlso, the computer model accurately predicts the high shear heating observed at specific locations in the cavity during the filling process.