Prediction model of surface integrity characteristics in ball end milling TC17 titanium alloy

Surface integrity is important to improve the fatigue property of components. Proper selection of the cutting parameters is extremely important in ensuring high surface integrity. In this paper, ball end milling of TC17 alloy has been carried out utilizing response surface methodology. The effects of cutting speed, feed per tooth, cutting depth, and cutting width on the surface integrity characteristics, including surface roughness (R_a), surface topography, residual stress, and microstructure were examined. Moreover, predictive metamodels for surface roughness, residual stress, and microhardness as a function of milling parameters were proposed. According to the experimental results obtained, the surface roughness increases with the increase of milling parameters, the (R_a) values vary from 0.4 μm to 1.2 μm along the feed direction, which are much lower compared to that along the pick feed direction. The surface compressive residual stress increases with the increase of feed per tooth, cutting depth, and cutting width, while that decreases at high cutting speed. The depth of the compressive residual stress layer is mostly in the range of 25–40 μm. The milled surface microhardness represents 6.4% compared with the initial state; the work-hardened layer depth is approximately 20 μm. Moreover, plastic deformation and strain streamlines are observed within 3 μm depth beneath the surface. The empirical model of surface integrity characteristics is developed using the results of ten experiments and validated by two extra experiments. The prediction errors of the three surface integrity characteristics are within 27%; the empirical model of microhardness has the lowest prediction errors.