Ultrasonic cavitation abrasive finishing

• Main Author: Tan, Kai Liang
• Other Authors: Yeo Swee Hock
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2019
• Subjects: DRNTU::Engineering::Mechanical engineering
• Online Access: https://hdl.handle.net/10356/106122; http://hdl.handle.net/10220/47908
• Institution: Nanyang Technological University

The development of Powder Bed Fusion (PBF) additive manufacturing process has built a new manufacturing paradigm for metal end-use components. However, the as-manufactured components often have poor surface quality and require post-process finishing. The unique morphology of PBF surfaces poses a huge challenge to most existing processes. To resolve this, a targeted finishing strategy is desired. The present thesis aims to develop a novel surface finishing process for PBF components based on intense ultrasonic cavitation in a solid-liquid medium. Two main research questions are addressed in this explorative study. Firstly, is the developed process, often associated with surface deterioration, capable of improving the surface quality of a PBF component without compromising its surface integrity? Secondly, in view of the well-known cavitation shielding effects, what are the capabilities and limitations of the developed process in internal channel finishing? An experimental rig, termed as the Ultrasonic Cavitation Abrasive Finishing (UCAF) system, was designed and assembled. A 20 kHz horn-type ultrasonic system was used to induce cavitation effects in a mixture of deionised water and micro-abrasives. To answer the first research question, the surface quality of PBF surfaces post-UCAF was characterised. Microscopic observations were also made extensively to gain insights into the UCAF mechanisms. This was followed by a parametric study to enhance process understanding. The second research question is answered through a comprehensive high-speed visualisation and acoustic spectrum analysis of cavitation phenomenon within millimetre-scaled channels. Actual UCAF on PBF channels was then conducted to verify the results. The surface morphology, topography, dimensions and microhardness of post-UCAF specimens were characterised. UCAF removes most surface irregularities and alters the surface morphology of PBF surfaces visibly. The final Ra values after 45 min processing duration is between 2.7 µm and 3.8 µm for surfaces built at 3 different orientations: 0° (top), 45° (sloping) and 90° (side). On sloping and side surfaces with rougher initial morphology, %Ra and %Rz improvements are more than 30 %; whereas the top surface registers negligible changes. The post-process dimensions of all 3 types of surfaces are within 10 µm of the intended design thickness. This illustrates the feasibility of UCAF on PBF components with varied types of surfaces. The largest microhardness enhancement recorded is approximately 15 %, with no significant phase changes or grain refinement observed. Two main modes of material removal have been identified through extensive microscopic observations. The first is the removals of surface irregularities through direct collapses of heterogeneously nucleated bubbles. The second is the gradual erosion of fully-melted surface structures such as stair-steps through micro-abrasive impacts. Mass losses and surface topography improvement can be mostly attributed to the direct bubble collapse mechanism. Nonetheless, micro-abrasives are crucial in curbing further erosion progression and aiding in surface homogenisation. In the parametric study, the effects of abrasive size, concentration, ultrasonic amplitude and working gap were investigated. Increasing abrasive size or concentration inhibits the direct bubble collapse mechanism and enhances micro-abrasive impacts. A moderate abrasive size of 12.5 µm and concentration of 5 wt% results in the best %Ra improvement at 41 %. Increasing the working gap from 0.8 mm to 2 mm leads to severe attenuation, as evident from the drop in %Ra improvement from 41 % to 22 % and the reduction of finishing area by half. Cavitation phenomenon within channel confinements was then characterised. The ratio of channel diameter over the horn diameter, d/ϕ, and ultrasonic amplitude were varied. At d/ϕ < 1, the conical bubble structure typically observed under an ultrasonic horn is disrupted. Bubble density increases and bubble propagation through the channel improves. However, at d/ϕ < 0.50, excessive bubble crowding at the channel entrance reduces cavitation intensity at the deeper regions. Across all conditions, a unique phenomenon of bubble cluster formation near the channel exit region (~ 20 mm away from the ultrasonic horn) is observed. These large clusters have maximum diameters of up to 6 mm and collapse transiently at the maximum ultrasonic amplitude. They grow and collapse repetitively up to hundreds of cycles at a constant position. The findings suggest that the transient cavitation zone under an ultrasonic horn is lengthened with the introduction of channel confinements. Pure cavitation, without micro-abrasives, was first attempted on channels of various d. The surface irregularity removal efficiency shows a decreasing trend along the channel length from the entrance section. The severity of attenuation increases as d decreases. The finishing effects are only observable within the first 8 mm from the channel entrance at the smallest d of 3 mm, a drop from 20 mm at d = 9 mm. Upon micro-abrasive addition, the surface irregularity removal efficiency near the exit section more than doubles. This illustrates that micro-abrasives are crucial in enhancing the finishing range. With the abrasive size at 50 µm and maximum ultrasonic amplitude, %Ra and %Rz improvements are over 20 % along the entire channel length of 20 mm. The findings support the hypothesis that intense cavitation in a solid-liquid medium is a feasible surface finishing principle on PBF surfaces. This thesis also prompts a rethinking of surface finishing approach to the challenges posed by PBF components. Investigations into the research questions have led to new insights on cavitation silt-erosion and cavitation phenomenon within channel confinements. On this basis, it is recommended that further research be conducted to develop process models, extend UCAF to components with higher complexity and explore other cavitation generation setups.