A new theoretical model for surface roughness prediction in rotational abrasive finishing process

Azami, Aref and Khoshanjam, Ali and Jerez-Mesa, Ramon and Lluma-Fuentes, Jordi and Travieso-Rodriguez, Jose Antonio (2023) A new theoretical model for surface roughness prediction in rotational abrasive finishing process. Wear, 524-525. 204772. ISSN 0043-1648 (https://doi.org/10.1016/j.wear.2023.204772)

[thumbnail of Azami-etal-Wear-2023-A-new-theoretical-model-for-surface-roughness-prediction-in-rotational-abrasive-finishing-process]
Preview
 Text. Filename: Azami-etal-Wear-2023-A-new-theoretical-model-for-surface-roughness-prediction-in-rotational-abrasive-finishing-process.pdf
Accepted Author Manuscript
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo
Download (2MB)| Preview

Abstract

Rotational abrasive finishing (RAF) is a new nano-finishing technique in which the finishing forces are applied to the workpiece by the opposite rotations of a stirring-blades and the workpiece. The RAF process allows for finishing the inner and outer surfaces of workpieces, particularly complex ones with axial symmetry. The present study aims to propose a new theoretical model to obtain deeper insights into the material removal mechanism, surface roughness prediction, and forces in RAF process. Since the abrasive-workpiece interaction is random and complex in nature in the RAF, some initial assumptions were considered. To validate the repeatability of the experimental results, the experiments were designed based on the Response surface method (RSM). To validate the new proposed theoretical model, a number of influential parameters were investigated. It was found that the stirring-blade speed (S), working gap (W), and abrasive grain size (A) had significant effects on Ra. The minimum surface roughness (Ra) was obtained to be 46.87 nm at a rotational speed of 600 rpm, a working gap of 1 mm, and grain size of 18 μm. The experimental results were relatively in good agreement with the theoretical results so that the maximum error was about 24%. This can be assumed that the most important explanation for the difference between the theoretical and experimental results can be attributed to the initial theoretical assumptions.

CORE (COnnecting REpositories)