Stability modeling with dynamic run-out in high speed micromilling of Ti6Al4V

Abstract

The low flexural rigidity of the micro-tool makes it susceptible to failure under high cutting forces and/or unstable process conditions. One way to counter the effect of low flexural rigidity is to use high rotational speed which reduces the chip load and, therefore, the cutting forces acting at the tip of the micro-end mill. However, high rotational speeds can amplify the effects of run-out and misalignment. This speed-dependent tool eccentricity, known as dynamic run-out, can be much larger than the quasi-static (speed independent) run-out. The dynamic run-out is more pronounced at high rotational speeds (∼100,000 RPM) due to the increased centrifugal force. The chip thickness is modified due to the dynamic run-out which affects the cutting forces and the stability. It is necessary to incorporate the dynamic run-out in the chip thickness model for an accurate prediction of the forces and the stability limits in high speed micromilling. Consequently, this paper is focused on characterizing the dynamic run-out and incorporating its effect on the undeformed chip thickness, cutting forces and stability limits. The predicted cutting forces with the modified chip thickness due to dynamic run-out shows a better agreement with the experimentally measured cutting forces as compared to the quasi-static run-out and ideal cycloidal chip thickness based force models. The predicted cutting force with dynamic run-out in the feed-direction gives an error of 11.54% as opposed to a relatively large error of 73.1% with a quasi-static run-out based model at 100,000 rpm. The Nyquist stability criterion has been used to predict the stability for 2-DOF chatter model of micromilling process. At a chip load of 3 µm/flute, the predicted stable depth of cut is higher if the dynamic run-out is included at high rotational speeds (> 51,500 rpm). This increase is attributed to a high dynamic run-out which results in cutting via only one flute. However, if the dynamic run-out is relatively low (between 29,500 rpm and 51,500 rpm), both the flutes are engaged in machining and the predicted limit is lower than that of the quasi-static run-out. The predicted values are in reasonably good agreement with the experimentally determined chatter onset points for most cases.