Experimental Investigation of Induced Tool Vibration in Turning of Hardened AISI52100 Steel

Purpose The demand for hardened steel for different applications particularly in the field of aviation and the automotive industry is increasing day by day. Different machining procedures like grinding, electric discharge were utilized for the machining steel. These machining measures are tedious because of the lesser material evacuation rate and an impediment to cutting complex profiles. The new emerging advanced process, hard turning, has the potential to produce parts from the hardened steel using single-point advanced cutting tools like ceramics, cubic boron nitride, and carbide. The cutting tool is exposed to large mechanical and high-temperature loads as a result, vibrations are generated throughout the machining process consequently the event of rapid wear and catastrophic failure of the cutting edge. As a result, the ability to estimate tool wear for hard turning is critical. Objective This paper aims to evaluate tool vibration acceleration with varying machining parameters such as cutting speed, feed, and depth of cut to create a predictive mathematical model and tool condition monitoring system (TCMS). Methods The Central Composite Rotatable Design approach is used to conduct the trials. The experimental results are examined, and regression analysis is utilized to create a mathematical model. The model is being used to create a tool monitoring system based on a microcontroller. The prediction model is compared to the proposed system, and the results are presented. Results After analyzing the data, it was revealed that the cutting conditions and vibration signals have a significant effect on tool wear. The DataFit Statistical commercial tool is used to do statistical analysis on the gathered results. DataFit Statistical commercial tool. The tool wear model's R-2 value is 0.93, indicating that the factors cutting speed, feed, and depth of cut, as well as accelerations V-x, V-y, and V-z, have a considerable impact on tool wear and offer reliable flank wear estimations. The residuals are close to a straight line with a maximum error of 9%, indicating that error is normally distributed and the model does not suggest any insufficiency and gives a trustworthy prediction, according to the normal probability plot. It is seen that tool wear obtained from developed TCMS has the maximum percentage error of 12.24% with experimental value and minimum error is 7.62%, whereas the maximum and minimum error of RA model with TWMS results is 12.37% and 1.04%, respectively.