Process and media development of an MR fluid-based finishing for high efficiency material removal

Magnetic field-assisted finishing (MFAF) is a class of non-conventional polishing processes that utilize ferromagnetic smart fluids. As the final stage in a chain of manufacturing processes, these processes are used for surface texture reduction, removal of sub-surface damage layer, or for form corr...

Full description

Saved in:
Bibliographic Details
Main Author: Kum, Chun Wai
Other Authors: Takashi Sato
Format: Theses and Dissertations
Language:English
Published: 2016
Subjects:
Online Access:https://hdl.handle.net/10356/69007
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Magnetic field-assisted finishing (MFAF) is a class of non-conventional polishing processes that utilize ferromagnetic smart fluids. As the final stage in a chain of manufacturing processes, these processes are used for surface texture reduction, removal of sub-surface damage layer, or for form correction. Recently, the aerospace industry has identified MFAF processes as a potential solution for polishing of freeform external surfaces of aerospace components made of titanium alloy (Ti-6Al-4V). Two of the most pressing issues are the process efficiency in terms of material removal rate, and the ability to predict material removal based on process conditions, with the ultimate goal of process automation. The two main focus of this project reflect the two aforementioned issues. For the first focus pertaining to increasing the process efficiency, a new MFAF process is developed. The double-magnet configuration of the polishing unit used in the process, which is a novel concept, is developed to achieve high material removal rate. Two factors contributing to the increased material removal rate are identified and established. The first factor is the high magnetic flux density in the polishing zone, which is verified by magnetostatic analysis and measurement on physical setups with a magnetometer. The second factor is the in situ reformation of finishing media during the process, which allows the finishing media to exert sustainable pressure on the workpiece surface. The mechanism responsible for the in situ reformation of finishing media is also identified, described and established. Following that, the effect of the tool parameters on the process outcome is studied. Specifically, the parameters studied are the magnet-to-magnet gap, magnet-to-workpiece gap, and the thickness of magnet. Using magnetostatic analysis, the relationships between these parameters and the magnetic flux density are established. Additionally, the use of magnetic cap to augment the magnetic flux density is also assessed. Based on the findings from these studies, a set of dimensions is recommended for the polishing unit. The capabilities of the new MFAF process are demonstrated. Surface texture reduction is confirmed for both stainless steel (SUS316) and titanium workpieces, where a final surface texture of 0.016 μm Ra and 0.073 μm Ra are achieved respectively, from an initial surface texture of approximately 1 to 2 μm Ra. The removal rate of the new MFAF process is also compared against similar processes, and its material removal rate of 11.8 μm/min is among the highest. Finally, the feasibility of the process for surface finishing of structured surfaces is also assessed. 2.5-D V-shaped channels with depth of 0.1 mm and width of 0.2 mm are polished. Surface texture of both the peaks and valleys is reduced, but the removal rates are different. As a result, the peak-to-valley height of the channels is reduced from 100 μm to 14 μm. Non-uniformity of removal rate is a weakness of the new MFAF process that needs to be considered in future work. For the second focus pertaining to prediction of material removal rate, a material removal model based on contact mechanics is proposed to better represent the complexity of the finishing media in the MFAF process. The proposed model consists of a base model and two extensions for conditions where assumptions made for the base model are not fulfilled. The complete derivation of the base model and both extensions is presented in this thesis. Then, experiments are conducted to verify the proposed model. The theoretical trends given by the proposed model are found to be in good agreement with the experimental trends.