Modelling of the stability of variable helix end mills

Abstract Chatter in milling is still the main obstacle in achieving high-performance machining operations in industry. In this paper, an analytical model is presented to predict the chatter stability of the variable helix end mills. Owing to the lack of accurate and rapid modelling, variable helix tools are used simply on a trial and error basis, in the hope that they will improve the stability of the process. This work provides a comparative study of the performance of variable helix and variable pitch end mills. Time domain chatter recognition techniques and analytical models are explored and tested against experimental results. For the experimental validation, aluminium test pieces were used to enable a broad range of spindle speeds to be covered. The linearity of the machine tool dynamics is explored through validation of standard stability lobes. A comparison of the predicted and observed performance of variable helix against constant helix, variable pitch end mills are presented. The stability lobes are validated for each of the variable helix, variable pitch and standard tools. The analytical model assumes the variable helix tools to behave as variable pitch tools, calculating the average pitch angle for each flute. This approximation is proven to be accurate for some of the cases. For certain combinations of pitch and helix angle greatly enhanced stability is demonstrated empirically with up to a 20-fold increase in depth of cut for the variable helix over the equivalent variable pitch tool. This enhanced stability is neither predicted by the analytical nor time domain solutions. The time domain chatter recognition criterion is investigated and found to have little influence on the predicted stability plots. It is concluded that the enhanced stability is a result of some mechanism not represented in the well-established time domain model. It is possible that this is a result of the disturbance of regeneration in the manner of an alternating spindle speed or due to a non-linear process damping effect.