Finite element method based sliding wear prediction of steel-on-steel contacts using extrapolation techniques. (October 2019)

• Record Type: Journal Article
• Authors: Bose, Kunal K; Ramkumar, P
• Is Part Of: Proceedings of the Institution of Mechanical Engineers. Volume 233:Number 10(2019:Oct.)
• Page Start: 1446
• Page End: 1463
• Publication Date: 2019-10
• Subjects: Contact mechanics -- extrapolation -- finite element method -- steel -- UMESHMOTION -- wear; Tribology -- Periodicals; 621.89
• DOI: 10.1177/1350650119836813
• Languages: English
• ISSNs: 1350-6501

Abstract

Wear is a complex phenomenon, which depends on various parameters such as load, velocity, material properties, surface, environmental conditions, etc. Hence, wear prediction is a challenging part of engineering. This paper focuses on numerically predicting the wear of 304 stainless steel pin sliding against AISI 52100 bearing steel disc, using pin-on-disc tribometer setup. The experiments are performed for loads of 10 N, 30 N, and 50 N and a sliding speed of 0.4 m/s. The wear coefficient and coefficient of friction obtained from the experiments are given as input to a 2D elastic finite element method model using a commercially available finite element method-based software ABAQUS. The differential form of the Archard's wear law is used to obtain the wear depth at the contact nodes. The UMESHMOTION+ Arbitrary Lagrangian–Eulerian technique is used to update the contact geometry after each wear increment. The major drawback of wear simulation is the large computational time requirement. To address this, three extrapolation techniques are used namely, the constant extrapolation, the linear extrapolation, and the constant pressure extrapolation technique. A new criterion for using extrapolation during sliding wear simulation was proposed. The extrapolation techniques take into consideration the evolution of the contact pressure and contact geometry during sliding wear. The effectiveness of these techniques based on the computational time and accuracy are analysed. Based on the accuracy, the linear extrapolation technique was found to be most effective, while the constant pressure extrapolation technique was most useful in reducing the computational time. The numerical results obtained are validated with the experimental results.