Parametric Optimization of Centrifugal Force-Assisted Abrasive Flow Machining (CFAAFM) by the Taguchi Method

Abstract

Extrusion honing, known as abrasive flow machining (AFM), deburrs, polishes, and radiuses surfaces and edges by flowing an abrasive-laden media over these areas. The process is particularly used on internal shapes that are difficult to process by other nonconventional machining processes. Because abrasive action occurs only in areas where the flow is restricted, tooling is used to direct the media to the appropriate areas. Like other nonconventional machining processes, AFM has the limitation of lower material removal rates. The application of centrifugal force (by using rotating rectangular rod inside the hollow workpiece) has been explored for the productivity enhancement of the process. This article reports that centrifugal force enhances the material removal rate (MRR) and improves the scatter of surface roughness (SSR) value in AFM. It outlines the development of a system that determines sets of viable process parameters for a new process called centrifugal force-assisted abrasive flow machining (CFAAFM). Cylindrical workpieces of brass are used for the experiment. During the experiments, parameters, such as rotational speed of rectangular rod, extrusion pressure, and grit size, were varied to explore their effect on material removal and scatter of surface roughness. Taguchi's parameter design strategy has been applied to investigate the effect of process parameters on the MRR and SSR values.

Keywords:

• AFM
• CFAAFM
• Centrifugal force
• CFG rod
• Optimization
• Taguchi method

Notes

Polymer to gel ratio: 4:3; abrasive concentration: 1:2; workpiece material: brass; number of cycles: 3; abrasive type: Al₂O₃; temperature: 32 ± 2°C.

R₁, R₂, R₃ represent responses values for three repetitions of each trial. The 1's, 2's, and 3's represent levels 1, 2, and 3 of the parameters, which appear at the top of the column. (–) represents no assignment in the column. Y_ij are the measured values of the quality characteristic (response).

Ll, L2, and L3 represent average values raw data at levels 1, 2, 3, respectively, of parameters. L2–L1 and L3–L2 are the average main effect when the corresponding parameter changes from level 1 to level 2 and level 2 to level 3, respectively.

∗Significant at 95% confidence level.

SS = sum of squares; DOF = degree of freedom; V = variance; SS′ = pure sum of squares.

∗Significant at 95% confidence level.