Contents lists available at ScienceDirect Tribology International journal homepage: www.elsevier.com/locate/triboint Finite element analysis of fretting wear under variable coefficient of friction and different contact regimes Tongyan Yue a , Magd Abdel Wahab b,c,d, ⁎ a Department of Electrical Energy, Systems and Automation, Faculty of Engineering and Architecture, Ghent University, Belgium b Division of Computational Mechanics, Ton Duc Thang University, Ho Chi Minh City, Vietnam c Faculty of Civil Engineering, Ton Duc Thang University, Ho Chi Minh City, Vietnam d Soete Laboratory, Faculty of Engineering and Architecture, Ghent University, Technologiepark Zwijnaarde 903, Zwijnaarde B-9052, Belgium ARTICLE INFO Keywords: Fretting wear Coefficient of friction Finite element method (FEM) ABSTRACT Fretting wear is a material damage in contact surfaces due to micro relative displacement between two bodies. It causes some unexpected results, such as loosening of fasteners or sticking in components supposed to move relative to each other. Since this micro motion of fretting wear is difficult to measure in experiments, finite element method (FEM) is widely used for investigating the evolution of contact variables and wear scars during fretting wear process. In most FEM simulations of fretting wear, coefficient of friction (CoF) is assumed to be constant in order to simplify the models. As measured in experiments, however, the evolution of CoF has a relation with the wear number of cycles, especially during the running-in stage. In this research, the effects of variable CoF are considered in both gross sliding and partial slip conditions of fretting wear. The wear scar and wear volume predicted by FEM models for constant and variable CoF cases are calculated. Results indicate that, in gross sliding condition, whether or not using a variable CoF has little effect on wear volume at the end of the steady state stage of fretting wear cycles. However, when considering partial slip or running-in stage of gross sliding conditions, FE models with variable CoF achieve predictions that are closer to experimental results. 1. Introduction Fretting is a small movement between contact surfaces. Depending on different loading conditions, namely combinations of the normal load imposed in the contact bodies and the tangential displacement between them, it could result in two fretting conditions: partial slip and gross sliding. If the normal load is sufficiently high or the oscillatory displacement is small enough, points of the contact centre are in stick regime, while the remaining of the contact is in slip regime. This condition is called partial slip regime, since both sticking and slipping exist at the contact surface [1–10]. Decreasing the normal load or increasing the applied displacement amplitude, the sticking area reduces until it vanishes. In this case, the whole contact surfaces will slide with each other, which is known as gross sliding regime [11–14]. Fretting wear, defined as wear due to fretting in ASTM [15], occurs at both partial slip and gross sliding regimes. As material damage, fretting wear is still a challenge to engineers for design of engineering components that undergo vibration or oscillation, such as stem/cement of hip joint [16], blade/disk of dovetail joint in turbine [17], etc. This is because of the continuous change of the contact surfaces in component, which is difficult to measure during fretting wear experiments. Therefore, finite element method (FEM) is intensively used for predicting the process of fretting wear since it is suitable to solve problems like non-linearity of boundary conditions, changes in geo- metry and time integration effect, which all happen in fretting wear simulations. Finite element modelling of fretting problems has been reported in the literature by several authors [8,9,11,18–27]. Generally, two wear models are widely employed for FE wear simulations. The classical Archard model [28], which calculates wear volume based on the sliding distance, the normal load and the wear coefficient, is firstly integrated into FE model of fretting wear by McColl et al. [18] in 2004. Energy model is another wear model simulating fretting wear damage. This model is based on the experimental finding that wear volume is linearly related to the accumulated dissipated energy converting from frictional work, proposed by Fouvry in 1996 [7]. Later on, this energy model is implemented to simulate fretting wear surface evolution of Ti–6A1–4 V contact, which is the first time that energy model is combined with FE fretting wear simulation [19]. Using any of these two wear models, different aspects of fretting wear were modelled and studied including the plasticity behaviour of material [29], debris effects [13,20,30], the impacts of fretting wear on fretting fatigue [31–33] and coating performance under fretting http://dx.doi.org/10.1016/j.triboint.2016.11.044 Received 1 November 2016; Received in revised form 28 November 2016; Accepted 30 November 2016 ⁎ Corresponding author at: Soete Laboratory, Faculty of Engineering and Architecture, Ghent University, Technologiepark Zwijnaarde 903, Zwijnaarde B-9052, Belgium. E-mail addresses: magd.abdelwahab@tdt.edu.vn, magd.abdelwahab@ugent.be (M. Abdel Wahab). Tribology International 107 (2017) 274–282 Available online 02 December 2016 0301-679X/ © 2016 Elsevier Ltd. All rights reserved. MARK