Influence of microphone tilt angle on instability identification in micromilling of thin-walled Ti6Al4V

Abstract

High speed micromilling is the most preferred manufacturing process to fabricate complex 3D featured thin-walled structures like stents, turbine blades, micro-fins, cabin parts etc., that are extensively used in bio-medical, energy, automobile and aerospace sectors. Fabricating thin-walled structures with high dimensional and geometrical accuracy requires chatter free machining parameters. Different sensors like accelerometer, current sensors, laser based sensors, vibration pick-ups etc. are available for capturing the vibration while machining thin-walled structures. Each sensor has its own advantages and disadvantages, like laser based sensors adds noise to response as chips fly past in the laser path and accelerometer captures deflection of both micro-cutting tool and thin-walled workpiece. Hence, In-process chatter identification is a challenging task in thin-wall micromilling. Also, the Filtering of signals to remove the noise from signals captured via contact type and laser based sensors results in loss of low-amplitude chatter data in high-speed micromilling. Consequently, there is a need of sensor, which functions similar to contact type with less or no loss in low-amplitude chatter data. The proposed work compares vibration signal captured using accelerometer and microphone at different tilt angles for radial milling of thin-walled Ti6Al4V at different radial immersion to analyse the effect of tilt-angle on chatter identification. The recorded signals from both accelerometer and microphone are analysed in time domain by estimation and comparison of statistical parameters, in frequency domain by analysing the distributed frequency and in time–frequency domain by analysing the energy of the signals. Complete characterization of the signals in time, frequency and time–frequency domain/methods indicated that microphone at 60° tilt angle can be used as an alternative sensor for chatter onset identification without loss in chatter data irrespective of cutting condition. Machined surface analysis of thin-walled structures complements the microphone as an effective chatter identification sensor.