EFFECT OF PROCESS VARIABLES ON THE QUALITY OF AI-4.5%Cu ALLOY CASTINGS PRODUCED BY V-PROCESS

Abstract

The V—Process has been investigated to study the effect of process variables on the quality of casting produced by V-Process. The following process variables were selected based on their importanCe: i) Sand grain size and its distribution characterised by AFS grain fineness number ii) Percentage graphite mixed with silica sand of AFS gfn 50 iii) Vibrating frequency iv) Vibrating time v) Degree of vacuum imposed, and vi) Pouring temperature of metal. The casting quality characteristics such .as dimensional accuracy, surface finish, casting soundness, and mechanical properties were analysed. A mathematical model was developed to simulate the deviation of casting dimensions from pattern dimensions. It has been found that the coefficient. of relative compressibility of sand and shrinkage of metal are mainly responsible for this deviation. The effect of process variables on the coefficient of relative compressibility of sand has been studied using a specially designed sand testing apparatus. iii Another model has been developed to simulate the Ra- thearithmetic average deviation of microirregulari-ties occurring on the casting surface produced by V-Process. The casting soundness and as-cast microstructure have been quantified by measuring volume porosity and secondary dendritic arm spacing. Also, mechanical properties viz, tensile strength, impact strength, and hardness have been measured. The effect of process variables on casting quality characteristics have been evaluated using the response surface methodology approach which correlates the independent selected process variables to the des'ired quality characteristics by the regression equations. The adequacy of these second order regression models have been ascertained by statistical analysis. An empirical relation estimating the tensile strength of Al-4.5% Cu allOy castings with the help of combined effect of porosity and dendritic arm spacing has been developed. Lastly; the optimal values of process variables have been found using multiple objective non-linear goal programming technique: