Gateway to Metal Casting Resources
HomeDie Casting MapDesign    

Foundry Map
Die Casting Map
Research Results to Improve Casting Productivity
Apply for Research Funding
Case Studies and Technology Demonstrations
Reduce Energy Consumption
Training & Workshops
Learn More About the Metal Casting Program




Home | Contact Us

Die Casting Part Distortion: Prediction & Control

ABSTRACT
The long term goal of this research project is to define comprehensive 3-dimensional scaling factors, that is, transformational methods that include all of the distortion phenomena. Although full accomplishment of this goal is not possible within this project, significant progress will be made. The results of this project will provide a description of the role that the load history, the solidification mechanisms and the environmental/process parameters play in the overall dimensions of the product. A critical activity in the first phase was sorting out and evaluating the relative contributions of the various mechanisms to die deflections. This evaluation was accomplished through a series of simple engineering analyses based primarily on the order of magnitude of the influence of each load considered on die deflections. In the current phase, work to date has developed modeling techniques to use a common finite element mesh for modeling casting solidification and cooling with a commercial system (ProCast in particular) and for post ejection distortion and stress modeling using a general purpose code Abaqus. Current work is focusing on performing verification studies of the modeling results; developing techniques to map the distorted shape of the die cavity, developed in the die distortion project, onto the part for post ejection analysis; and preliminary work to develop in cavity stress models to allow researchers to look at ejection force requirements and potential deformation that might occur during ejection. The shape of the cavity within the die defines the shape of a casting. Die cavities are not manufactured to exact dimensions expected from the casting because the die and part expand/contract during the casting cycle. This work will provide the basis for the definition of the guidelines and rules that will enable better dimensional control of the castings. The work includes a determination of nominal die dimensions, residual stress predictions, pressure effect modeling, ejection modeling, and distortion model development. Cases studies and comparison of numerical simulation results with experimental results for a simple test part have been completed. The dimensions checked were the distances between the fins. The data show good correspondence between simulation and experiment except for dimension D1. This difference is due to the effect of a cooling line in the die that was not included in the simulation. The results suggest that thermal distortion alone is a good predictor of final part dimensions for dimensions produced in a single die component. The results also show sensitivity to die temperature and cooling line location as would be expected since a cooling line effects the die and part temperature distribution. Methods were developed and programs completed that allow mapping finite element information from one mesh or simulation to another. This allows use of commercial simulation codes to calculate the solidification pattern and temperature distribution in the cavity and then use of any FEA program to calculate the part distortion. It enables separating die distortion and part distortion calculations in order to keep the computational complexity as low as possible. A related topic is mapping a distorted cavity onto the part to account for the die distortion. Progress has been better than expected in this area and techniques mapping techniques have been completed. Here the distorted mesh of the die cavity is mapped onto a mesh of the part. This provides a better representation of the part as it leaves the cavity than is currently available with commercial codes. While the cavity mapping has gone well, work with the detailed elastic-plastic model has been very slow. Contact conditions between the part and the die during solidification are quite severe requiring very small time increments. In addition, the elastic-plastic simulation is always iterative. The combination of the two is extremely long computation times. The focus at the moment is on improving the models to reduce the computation load.
RELATED REPORTS
PERFORMING ORGANIZATION
Ohio State University
CONTACT INFORMATION
Steve Udvardy
North American Die Casting Association
241 Holbrook Drive
Wheeling, IL  60090
website: www.diecasting.org

phone: (847) 279-0001
fax: (847) 279-0002
email: udvardy@diecasting.org

EERE logoMetal Casting logo