Sensor-Assisted Monitoring and Optimization of Process Parameters in Micro-end Milling of Ti-6Al- 4V Titanium Alloy ICOMM/4M 2010 No. 40 T. Thepsonthi 1 and T. Özel 2 1 T. Thepsonthi; MARL, Industrial & Systems Engr, Rutgers University, USA; thanongs@eden.rutgers.edu 2 T. Özel, MARL, Industrial & Systems Engr, Rutgers University, USA; ozel@rci.rutgers.edu ABSTRACT Micro-end milling is one of the promising methods for rapid fabrication of medical devices and implants with 3D complex shapes. However, controlling the micro-end milling process to obtaining the desired results is much challenging compared to that of macro-end milling due to the size effect and some uncontrollable factors. The problem is much pronounced when workpiece material is a difficult- to-process material such as Titanium alloys which are widely used as material of choice for small medical devices and implants. Therefore, in this study the feasibility of using acoustic emission (AE) signal to monitor and optimize surface generation in micro-end milling of Ti-6Al-4V was investigated. The results revealed that the mean, deviation and density of AE signal sensitively change in respond to a change in cutting parameters and generation of machined surface. Therefore, monitoring or predicting surface generation and burr formation in micro-end milling process is feasible by using of acoustic emission sensor assistance. In addition, experiments using Taguchi method revealed effects of spindle speed, feed rate and axial depth of cut on surface quality and burr formation. Feed rate seems to be a significant factor affecting the surface roughness in micro- end milling of Ti-6Al-4V. It is observed that mostly shearing dominated cutting occurs at high feed rate and results in better surface roughness. It is also found that the best process performance in terms of surface roughness and burr formation are not necessarily obtained at the same cutting condition. 1. INTRODUCTION Micro-end milling possesses several advantages such as ease of use, process flexibility, low set-up cost, unlimited part materials and high material removal rate. Therefore, it is one of the most promising methods for rapid fabrication of medical devices and implants with 3D complex shapes. However, scaling the conventional milling process down to a micro-scale results in encountering several problems. Many factors that can be ignored in macro-scale suddenly become significant in micro-scale; for instance, vibration, deflection, temperature, micro structure, etc. As a result, obtaining the desired performance in micro-end milling is much challenging than that of macro-end milling [1]. The problem is much more pronounced when workpiece material is a difficult-to-process material such as titanium alloys. Titanium alloys are widely used as material of choice for small medical devices and implants, because of their strength, high corrosion resistance, and most importantly, their biocompatibility with the host tissue. Nevertheless, it is generally known that titanium alloys are difficult-to-process materials due to their high yield strength, low elasticity modulus and low thermal conductivity resulting in undesired surface and burr formation and rapid tool wear. Fast development of tool wear poses many problems in terms of surface quality. Even though, tool wear in micro-end milling can be monitored using AE signal, it is not guaranteed that the desired machined surface is still attainable. Therefore, in order to obtain high quality surfaces in micro-end milling, it is necessary to develop a capability of direct monitoring the surface generation so that an adaptive control can be applied. On the other hand, surface formation (surface roughness, waviness and texture) in micro-end milling is highly affected by tool condition and cutting parameters (feed rate, axial and radial depth of cuts and spindle speed). In addition, at the edges of the surfaces there are often various shapes of burrs formed during micro-end milling of most metals. Although, various sensors such as acoustic emission (AE), accelerometer and force sensors are widely utilized to monitor tool wear and tool breakage in micro-end milling [2, 3] use of AE sensors for monitoring surface generation is not fully explored. It is also well known that force sensors with limited frequency bandwidth (<1 kHz) are not suitable to monitor micro-end milling process where tooth passing frequencies are in the order of 1-5 kHz for two-flute end mills rotating between 50000-150000 rpm [3]. Acoustic emission is a class of phenomena whereby transient elastic waves are generated by rapid release of energy from localized sources within a solid material and has been used in monitoring machining processes for more than two decades [4] . An advantage of using AE for process monitoring over the other approaches is that the frequency bandwidth of AE is much higher (1-10 4 kHz) than that of machine vibrations and ambient acoustic noise [5]. 227