Owusu Y A, Dargan T, Piwonka T S

Abstract: A statistical method of evaluating geometric variations of casting products using point cloud data measured on a coordinate measuring machine (CMM) is presented. The paper deals with a statistical analysis approach to evaluating the differences in dimensional accuracy of test castings as influenced by gating system, molding process, casting section thickness, and casting orientation. Feature parameters of flatness, parallelism, tapered angle, overall length, and height of test casting were obtained from CMM point cloud data; and these were used to identify the relative importance of the factors that contribute to the dimensional inaccuracies. The scope of this study included three casting thicknesses: ¼ in., ½ in., and ¾ in. The three sand molding processes used for the research included resin-bonded sand, lost foam (silica sand), and lost foam (low expansion sand). CMM point cloud data were statistically analyzed successfully to evaluate casting distortion. The research established that casting distortion (or casting quality) is influenced by the following parameters: casting thickness, orientation, gating system, and type of molding process at different degrees. The characteristic parameters monitored for the distortion includes casting length, height, thickness, parallelism, flatness, and casting angularity (angle). From the analyses, it was found that resin bonded sand process, top gating, and orientation down contributed most to the casting distortion.