Contents lists available at ScienceDirect Tribology International journal homepage: www.elsevier.com/locate/triboint Semi analytical fretting wear simulation including wear debris Vamshidhar Done a,c , D. Kesavan a , Murali Krishna R b , Thibaut Chaise c , Daniel Nelias c, ⁎ a GE GRC, Bangalore, India b GE Gas Power Systems, Bangalore, India c Univ Lyon INSA-Lyon, LaMCoS UMR CNRS 5259, F69621 Villeurbanne, France ARTICLE INFO Keywords: Fretting Wear debris Numerical wear modeling Semi-analytical method ABSTRACT Many numerical models are proposed in the literature using finite element and finite discrete element methods to study fretting wear, barely including the effect of wear debris. These models being computationally expensive, simulating large number of fretting wear cycles is not practically feasible. A new methodology is proposed which needs only bulk material properties like friction/wear coefficients and uses semi-analytical methods to simulate fretting wear with entrapped debris. In this approach, debris are assumed to be attached to one of the surfaces during the fretting process. The results obtained from this approach were compared with fretting experiments. The proposed method permits to capture the wear depth and scar width, and results are very close to that observed in the experiments. 1. Introduction Fretting wear is the material removal process occurring when two contacting bodies are in micro level relative motion and subjected to contact load. The wear rate during fretting depends on many para- meters, from experiments conducted with cylinder and flats speci- ments, Warmuth et al. [16] concluded that the diameter at the interface and slip play a prominent role in fretting wear. Large diameter and low slip causes least wear, whereas a large slip causes significant material removal irrespective of the diameter. Li and Lu [17] studied the effect of displacement amplitude on fretting wear of Inconel alloy. They observed micro-cracks at the junction of adhesion and sliding zone. For large amplitudes, 'plow' effect was found at the edge of the wear scar. Apart from contact load and displacement, the nature of debris formed during the fretting wear process determines the wear modes. When debris particles formed during the fretting wear are entrapped at the contact zone, there will be significant effect in altering the wear mechanism and the resultant wear scar. Specially, contact profile and contact pressure changes based on the amount of debris present at the contact region. Everitt et al. [18] concluded from their work that formation of debris within the fretting interface generally indicates that fretting wear has reached the steady state. Also, that debris layer moves into the substrate by an oxidation process. Gas turbine combustor components such as hula seal, liner stops, crossfire tubes, collars, swirlers etc. are subjected to dry fretting wear at the contact interfaces and most of the time the contact is not opened frequently to release the debris. A numerical model which can include the effect of debris during fretting wear will be very useful to predict the wear profiles. Whereas faster ejection of debris occurs for fretting wear of valve-valve seat interface of internal combustor engines because valve opens to let out the combustion gases and allow fresh air into the cylinder. But, performing fretting wear experiments to replicate opening and closing of valves is time consuming and generally experiments are performed without opening the contact which leads to different wear profiles compared to the wear profiles observed in the field. To bridge this gap of wear profiles obtained from experiments to the real wear profiles observed in the field, numerical models capable of considering different level of debris ejection rates will be very useful. Different types of fretting wear prediction methodologies without considering debris are found in the literature. Rodriguez-Tembleque et al. [19] applied 3D boundary elements to simulate 3D fretting wear problems. The methodology uses Lagrangian formulation to solve the contact problem and Archard wear law to compute wear. Fouvry et al. [20] performed 2D FEM wear simulation and showed that the wear profile evolves from an initial Hertzian, then elliptical and finally flat distribution. Tang et al. [21] proposed a multilayer node update method in FEM to simulate fretting wear involving large depth. The nodes interpolation method was provided for both 2D and 3D contact problems. Arnab el. al. [22] presented a damage mechanics stress based wear law to model fretting wear of Hertzian contact. They used FEM to obtain contact stresses and the wear coefficients obtained from their model was comparable to the values reported in the earlier literature. Lee et al. [23] formulated 2D contact model in terms of Cauchy integral equation to perform fretting wear analysis. They had http://dx.doi.org/10.1016/j.triboint.2016.12.012 Received 5 August 2016; Received in revised form 24 November 2016; Accepted 9 December 2016 ⁎ Corresponding author. E-mail address: daniel.nelias@insa-lyon.fr (D. Nelias). Tribology International 109 (2017) 1–9 Available online 10 December 2016 0301-679X/ © 2016 Elsevier Ltd. All rights reserved. MARK