Performance enhancement of material removal using a surface-refinement model based on spatial frequency–response characteristics in magnetorheological finishing

The efficiency of processes plays a crucial role in the manufacturing industry. While numerous studies have been conducted to improve process accuracy and time, investigations into the fundamental causes of process efficiency degradation remain insufficient. Additionally, many existing studies require complex methodologies and significant investment costs, making them less applicable in real-world manufacturing settings. To address these vulnerabilities, this research explores the control limitations of the magnetorheological finishing (MRF) process and proposes a surface-refinement model (SRM) aimed at enhancing process efficiency. The control limits of the MRF are examined by designing various spatial frequencies based on the size of the tool interface function (TIF) and analyzing the corresponding response characteristics. The proposed SRM improves process accuracy by filtering residual surface errors in accordance with the control limits of the MRF to achieve refined surfaces. To evaluate the practical effects of the proposed technology, comparative experiments were conducted to compare the performance of processes with and without the SRM. The results indicate that the application of the SRM yields approximately 25% greater accuracy. Furthermore, to validate long-term stability and applicability, procedures resembling actual manufacturing scenarios were designed, resulting in a surface error of 9.1 nm RMS for a Zerodur aspherical mirror. Notably, the experimental accuracy across all correction processes consistently exceeded 80%. These findings demonstrate the potential applicability of the SRM in various optical fields. Moreover, the straightforward technical approach of surface filtering not only reduces the complexity of technological implementation in the manufacturing industry but also yields superior outcomes, suggesting immediate applicability in industrial settings.