Statistical Models for Predicting Wear and Friction Coefficient of Valve Tappet Using ANOVA

Abstract

The focus of this study is to propose statistical models for predicting wear and coefficient of friction of the 16MnCr5 steel valve tappet using the analysis of variance (ANOVA). The 16MnCr5 steel valve tappet was fabricated from a steel strip and thermochemically treated by carburizing, quenching, and tempering. Under dry conditions, tribological tests were performed for 16MnCr5 steel valve tappet with a ball-on-plane configuration in a reciprocating slide using an Optimol SRV® v4 device. Subsequently, the results of the coefficient of friction (COF) and the wear rate were then analyzed using the ANOVA. Regression analysis was used to derive the predictive equations for both friction coefficient and wear rate. The applied load was found to be the most significant parameter affecting the COF and wear rate. The proposed statistical models has 88 - 92% percent reliability. These models can be beneficial for predicting the tribological operating conditions for the 16MnCr5 steel valve tappet to avoid premature failure within the tested load and temperature conditions.

Keywords

• anova
• friction
• regression analysis
• valve tappet
• wear