Influência da geometria do inserto de fresamento sobre a integridade superficial de uma liga de titânio

Abstract

Titanium alloys have got features that classify them as excellent candidates for medical applications. However, they have porr machinability mainly due to their low thermal conductivity, high chemical affinity to almost every tool material and their small modulus of elasticity. Regarding commercial milling tools, inserts usually differ one to another in terms of grade (such as substrate and coating composition and microstructure) and geometric characteristics (nose radius, rake angle and chipbreaker shape), so that it is difficult to find two products differentiated by one single variable. Due to these factors, the general goal of this research is to investigate the correlation between tool nose radius and the roughness after milling of Ti-6Al-4V ELI titanium alloy. For that, it was necessary to grind carbide inserts to prepare the three different nose radii (re = 0.8 mm; 1.2 mm or 1.6 mm) as well as their characterization, besides designing, manufacturing and evaluating a customized tool holder with a single insert seat. Tools once prepared were tested through face milling experiments, keeping fixed cutting speed (vc) at 60 m/min, axial depth of cut (ap) at 1 mm, in addition to radial depth of cut (ae) at 16.8 mm and flood coolant. Five diferente levels were used for feed per tooth ( fz): 0.10 mm; 0.15 mm; 0.20 mm; 0.25 mm or 0.30 mm. After the experiments, the specimens were evaluated in terms of different roughness parameters, and all data were statistically analyzed by means of analysis of variance (ANOVA). Regarding the characterization of the ground inserts, there was no significant variation in the irregularity of the edges as a function of the different nose radii. The design of the tool holder proved out to be adequate to the operation, providing a high design safety factor. Regarding the surface integrity of the machined surfaces, results show that, in exception of one tested condition, all other values of Ra satisfy N6 roughness grade. The pass associated to re = 0.8 mm and fz = 0.15 mm probably provided higher vibrations. ANOVA indicates that, in statistical terms, only feed ( f ) and the interaction between feed and nose radius ( f x re) affect the roughness parameters Ra, Rq, Rt, Rz, and Rmax, although these effects are not obvious. There was no statistically significant difference for skewness (Rsk), which tends to zero and, therefore, expresses an approximately symmetrical surface in terms of peaks and valleys. The same for kurtosis (Rku), which remained close to three, which means a distribution close to normal.