Chatter is detrimental to turning operations and leads to inferior surface topography, reduced productivity, dimensional accuracy, and shortened tool life. Avoidance of chatter has mostly been through reliance on heuristics such as: limiting material removal rates or selecting low spindle speeds and shallow depth of cuts. But, modern industries demand increased output and not steady operational limits. Various research efforts have therefore focused on developing mathematical models for chatter formation. However, as yet there is no existent model that meets all experimental verification. This research employed a novel technique based on the synergy of statistical modeling and experimental investigations in order to develop an effective empirical mathematical model for chatter amplitude and to subsequently find optimal machining conditions. Ti-6Al-4V, Titanium alloy, was used as the work-piece due to its increased popularity in applications related to aerospace, automotive, nuclear, medical, marine etc. A sequence of 15 experimental runs was conducted based on a small Central Composite Design (CCD) model in Response Surface Methodology (RSM). The primary (independent) parameters were: cutting speed, feed, and depth of cut. The tool overhang was kept constant at 70 mm. An engine lathe (Harrison M390) was employed for turning purposes. The data acquisition system comprised a vibration sensor (accelerometer) and a signal conditioning unit. The resultant vibrations were analyzed using the DASYLab 5.6 software. The best model was found to be quadratic which had a confidence level of 95% (ANOVA) and insignificant Lack of Fit (LOF) in Fit and Summary analyses. Desirability Function (DF) approach predicted minimum vibration amplitude of 0.0276 Volts and overlay plots identified two preferred machining regimes for optimal vibration amplitude.
Estimation of Surface Roughness in Face Milling Using Response Surface Methodology and Desirability Function Analysis
