Experimental analysis of deep slot milling in EN AW 2024-T3 alloy by stretched trochoidal toolpath and variable helix angle tool

Aluminium alloys are widely used in manufacturing of parts by subtraction of a high percentage of the initial volume. CAD-CAM offers dynamic roughing strategies as trochoidal toolpath, based in the principle of low values of the cutting forces and heat dissipation due to inherent intermittent cut. In this work, the deep slot milling process of EN AW 2024-T3 aluminium alloy by a stretched trochoidal toolpath, using a tool with variable helix angle and chip splitter, has been evaluated through a factorial design of experiments with the aim of stablish the optimisation rules for this solution. The material removal rate (MRR), surface roughness (Ra) and electric energy consumption (EEC) were analysed within the recommended range of the machining conditions. The results revealed the cutting speed (v(c)) and feed/ tooth (fz) as the most influential factors in MRR, while the surface quality was mainly affected by the radial depth of cut (a(e)) and feed/tooth (f(z)). The electric energy consumption depends mainly on vc and fz. Linear regression models were generated from the results which have been shown good reliability to predict and optimize MRR, EEC and Ra-wall and Ra-bottom. Three additional experiments validated the models showing minimum errors lower than 5 % for MRR and Ra-wall and around 10 % for EEC and Rabottom between experimental and predicted results. (C) 2021 CIRP.

Automated summary

The study evaluates the deep slot milling process of EN AW 2024-T3 aluminium alloy using a stretched trochoidal toolpath, with a focus on optimizing rules for material removal rate, surface roughness, and electric energy consumption under different machining conditions. Cutting speed and feed/tooth have significant effects on material removal rate, while surface quality is mainly influenced by radial depth of cut and feed/tooth. Linear regression models generated from the results show good reliability in predicting and optimizing MRR, EEC, and Ra-wall and Ra-bottom.