People and Projects

Geometry-Driven Manufacturability Analysis of
Cast Parts
Durgesh Joshi

Background: An industrial survey of 155 Indian foundries revealed that the average lead-time (9 weeks) and the average rejection levels (7.5%) are still very high. One of the main reasons is that cast parts are not designed considering process capabilities and constraints, because product designers lack the necessary domain knowledge. There is a need for a suitable methodology and tools for design for manufacturability (DFM), in which potential manufacturing problems are predicted early in product life cycle, and prevented by suitable modifications to part design.

Methods : A scientific approach for early manufacturability evaluation of cast parts has been evolved, involving quantitative assessment of the influence of part and tooling design on casting quality and cost. The effect of part geometry, particularly junctions, on casting solidification leading to porosity defect has been investigated using simulation and experimental validation. The knowledge gained is used for developing design guidelines for 3D junctions. The effect of part geometry on tooling cost is explored in terms of volume ratio, area ratio, depth ratio, thickness ratio, number of cores, and undercut volume. A relation for tooling cost based on the above shape complexity parameters is established by performing regression analysis on data compiled from 40 industrial castings.

Results : A case study of benchmark part and its revised version illustrate how the guidelines can assist in eliminating shrinkage porosity defect. The tooling cost equation has been successfully validated on castings not covered by the analysis. This is useful for comparing alternative product process designs in terms of tooling cost.

Conclusion : Castability can be assessed in terms of mouldability (shape fidelity, core aspect ratio, parting flatness), fillability (minimum thickness, gate velocity and air accumulation) and feedability (thickness gradient, feeder fettling and yield). The methodology can be applied to industrial castings for early assessment and improvement of castability, and is expected to significantly improve the quality and economy of castings.