DS0303 -

Development of an eXtended Friction Energy –third body Wear approach to predict the fretting wear rate of metallic interfaces. – X-FEW

XFEW: eXtended Friction Energy Wear model

This project aims at developing a new wear model that unifies the energy approaches (wear volume as a function of friction work dissipated in the interface), the third body theory (ejection flow of the debris bed from the interface) and the contact oxygenation concept (oxygen present in the debris bed modifies the wear process).

Introduction

Fretting resulting from micro-displacement oscillatory motion is considered a serious problem for many industrial applications (bridge cables, turbojet engines, nuclear power plants, etc.). It leads to substantial expenses for the maintenance and replacement of equipment. On the other hand, before the XFEW project, there was no wear model sufficiently reliable to predict the life of an assembly for a very wide range of stresses or contact sizes. The prediction of fretting wear was until now addressed by two approaches:<br />- An energy approach which consists in establishing the evolution of the wear volume as a function of the frictional energy dissipated in the interface. This approach is simple and can be easily integrated into a calculation code. However, it does not take into account the process of ejection of debris from the interface nor the influence of the contact oxygenation which modifies the mechanisms and therefore the kinetics of wear. <br />- A third body approach which consists in explaining the evolution of the wear kinetics according to the process of the ejection of the debris from the interface. This approach is more physical than the energy approach but remains very difficult to implement in a numerical code.<br />- Finally, neither the energy approach nor the third-body approach were able to take into account the effect of oxygen diffusion in the rubbed interface which, by modifying the wear mechanisms (abrasive wear versus adhesive wear), modifies the wear kinetics. <br />Thus a major scientific challenge, which motivated the XFEW project, was to develop a unified approach to these different concepts in order to establish a reliable wear law that can be easily integrated into a numerical code.

The fretting-wear kinetics of a homogeneous 34NiCrMo16 steel crossed flat-on-flat interface is studied applying a very large spectrum of loading parameters (pressure, sliding amplitude, frequency, mono-contact, textured surfaces, contact size, etc.). These tests were carried out on a fretting wear bench specifically set up for this project. In a first step, the evolution of the partition between the adhesive wear zones in the center of the contact and the abrasive wear zones observed at the contact edges were analyzed using post-mortem analyses (SEM, EDS, Raman,...). This partition was formalized through the contact oxygenation concept (COC): if the concentration of oxygen present in the debris bed is lower than a threshold value, the metal surfaces can no longer oxidize and the interface evolves towards an adhesive wear. This concept was modeled using an ADR (Advection, Diffusion, Reaction) approach. It has been validated for a coupled experimental/simulation analysis. In parallel, we studied the debris ejection process by adjusting the contact size in order to formalize the kinetics of the third body ejection (TBT: Third Body Theory). Indeed, the longer the length of the contact, the longer the debris residence time in the interface, the thicker the debris bed, the more the contact surfaces will be protected (screened by the debris bed) and the lower the wear kinetics will be. Finally, we formalized the kinetics of the wear volume increase (V) as a function of the friction energy dissipated in the interface (SEd). Thus, by establishing characteristic lengths that characterize the contact oxygenation L_COC and the ejection of the debris bed L_TBT, but also by integrating the effect of the frequency (which acts on the kinetics of oxidation) we proposed an extended energy wear law (XFEW) for which the energy wear coefficient is no longer constant but expressed as a function of weight of the various parameters of loading (pressure, sliding amplitude, frequency) but also of the quantities L_COC and L_TBT. Thus, by integrating the physical mechanisms of the wear processes, this very simple formulation can be easily implemented in a numerical code. The comparison with a wide range of experimental conditions confirmed the stability of the model. In parallel to this work, a more fundamental research was conducted to study the influence of the cohesion between the debris on the friction using a meshless multi-body DEM model. To confirm the results of the DEM model, the adhesion between debris was experimentally modified by controlling the gaseous environment, ambient air or pure argon, in which fretting tests in sphere-on-flat configuration were performed.

The main results of this project are as follows:
1 - Implementation of a high load (hydraulic) wear fretting bench allowing studying very large flat-on-flat contacts.
2 - Formalization of the composite structure (abrasive wear zone / adhesive wear zone) of the wear scars generated by fretting based on the Contact Oxygenation Concept (COC): if the oxygen concentration present in the debris bed is lower than a threshold value, the metal surfaces can no longer oxidize and the interface evolves towards an adhesive wear. If this concentration is higher than this threshold value, the oxidation of the metallic surfaces is possible and the interface evolves locally towards an abrasive wear. This explains why adhesive wear is observed at the center of the contacts and abrasive zones are located at the edge where the oxygen concentration is higher.
3 - Modeling of the COC (Contact Oxygenation Concept) from an ADR (Advection, Diffusion, Reaction) approach taking into account the porosity of the debris bed. This model allows predicting the position of the border between adhesive and abrasive zones. 3D numerical code was developed (ADR_COC) which was then simplified to give explicit mathematical expressions for axisymmetric and 2D contacts.
4 - Coupling the ADR-COC code with a local wear model (FEM Wear Box) taking into account the presence of the debris bed. This model has made it possible to simulate for the first time the «W« wear scars generated in gross slip condition which are usually observed experimentally but which until now could never be simulated with conventional wear models.
5 - Analysis of the wear kinetics for a very wide range of loads (pressure, sliding amplitude, frequency, contact size etc...). Introduction of two characteristic lengths L_COC and L_TBT allowing the formalization of the effect of the contact size on the wear kinetics.
6 - Introduction of an extended wear energy coefficient (XFEW) taking into account the loading conditions, the size of the contact but also the activation of the COC and TBT processes through a formulation of weight functions. Established with a limited number of tests, this formulation allowed predicting the wear kinetics of test conditions very different from those used to establish the model. Based on an energy concept, this approach can easily be implemented in numerical codes (e.g. FEM).
7 - Development of a DEM code for the local dynamic analysis of the third body in order to link the effect of the rheology of the third body to the evolution of the friction coefficient.

We would like to deepen our research on the following axes:
1- Coupling the COC-ADR model and the DEM approach (third body rheology) in order to enrich the XFEW mesoscopic model with a more accurate description of the third body rheology.
2 - Development of a 3D model (ADR-WEAR-BOX) in order to propose an «engineering« code to better simulate the risk of fretting wear in interfaces subject to fretting wear.
3 - Studying the effect of the environment (neutral atmosphere, vacuum) in order to control the oxygenation process of the interface.
4 - Studying different materials presenting a different oxidation behavior but also giving rise to wear debris with various cohesive properties.
5 - Coupling this wear model with the study of cracking, in particular for the so-called mixed regime conditions where the activation of adhesive zones by generating over-stresses can favor the initiation of cracks.
We hope to be able to benefit from a support for these research projects either within the framework of an ANR project or via other funding.

This ANR project has been fruitful in terms of scientific publications. In addition to a doctoral thesis, eight papers (7 published + 1 submitted) are from LTDS and LAMCOS laboratories. The work carried out in these publications was awarded two prizes (HIRN prize and Ken Ludema Best paper award) during the JIFT2021 and WOM2021 conferences respectively, reflecting the strong scientific impact and the originality of this ANR project at the national and international levels.
It will be reminded that this work has been disruptive in the field of tribology and more particularly fretting wear. This approach (XFEW model) is now considered by industrial companies like SAFRAN to better formalize their wear kinetics on large contacts. Finally, it is widely quoted in the international literature and «taken up« in different forms by our British colleagues.
Scientific publications:
1) Y. Zhang, G. Mollon, S. Descartes, Significance of third body rheology in friction at a dry sliding interface observed by a multibody meshfree model: Influence of cohesion between particles, Tribol. Int. 145 (2020) 106188. doi:10.1016/j.triboint.2020.106188.
2) S. Baydoun, S. Fouvry, S. Descartes, P. Arnaud, Fretting wear rate evolution of a flat-on-flat low alloyed steel contact: A weighted friction energy formulation, Wear. 426–427 (2019) 676–693. doi:10.1016/j.wear.2018.12.022.
3) S. Baydoun, S. Fouvry, An experimental investigation of adhesive wear extension in fretting interface: Application of the contact oxygenation concept, Tribol. Int. 147 (2020) 106266. doi:10.1016/j.triboint.2020.106266.
4) S. Baydoun, P. Arnaud, S. Fouvry, Modelling adhesive wear extension in fretting interfaces: An advection-dispersion-reaction contact oxygenation approach, Tribol. Int. 151 (2020) 106490. doi:10.1016/j.triboint.2020.106490.
5) S. Baydoun, P. Arnaud, S. Fouvry, Modeling contact oxygenation and adhesive wear extension in axisymmetric flat circular fretting interfaces, Wear. (2021). doi:10.1016/j.wear.2021.203822.
6) P. Arnaud, S. Baydoun, S. Fouvry, Modeling adhesive and abrasive wear phenomena in fretting interfaces: A multiphysics approach coupling friction energy, third body and contact oxygenation concepts, Tribol. Int. 161 (2021) 107077. doi:10.1016/j.triboint.2021.107077.
7) S. Baydoun, Investigation of fretting wear of a flat-on-flat 34NiCrMo16 interface: Application and modelling of the contact oxygenation concept, PhD Thesis - Ecole Centrale de Lyon, 2020.
8) S. Baydoun, S. Fouvry, S. Descartes, Modeling contact size effect on fretting wear: a combined contact oxygenation - third body approach. Wear (submitted).
9) S. Baydoun, P. Arnaud, S. Fouvry, Explicit formulations of adhesive wear extension in fretting interfaces applying the contact oxygenation concept. Wear (accepted).


Fretting wear damages (micro oscillating slidings) are observed in numerous interfaces subjected to vibrations. These wear damages can be very detrimental in many industrial applications (turbine engine contacts, oscillatory bearings, etc). To optimize the design of new assemblies, fretting wear need to be simulated using numerical methods like finite element (FEM) or finite discrete element methods. Unfortunately no fretting wear model is currently able to provide reliable predictions. The actual strategy consists in applying the Archard’s wear equation which relates the wear volume extension to the product of the normal force multiplied by the sliding distance. This simple approach does not consider the effect of debris layer (third body) entrapped between the interface which consumes a significant part of the imputed friction energy. Consequently it can significantly modify the fretting wear rate inducing large dispersions between experiments and predictions. Through an experimental and simulations strategy, the objective of this project is to palliate this limitation by developing an eXtended Friction Energy Wear approach; the wear volume extension is predicted by multiplying the friction energy dissipated in the interface by an energy wear coefficient expressed as a function of the third body thickness and contact configuration.
In the aim to generate abrasive wear process Steel/Steel and Inconel/Stellite material contacts have been selected as model tribo-systems to establish this fretting wear approach. Additionally, this choice has also been done in concertation with associated industrials SKF and AIRBUS (Involved in a so called industrial committee of X-FEW project) to correspond to industrial applications (ball bearings, joints assemblies).
This research will implies various experimental fretting wear experiments and interface analyses developed at LTDS and LAMCOS laboratories. One main experimental issue is to target and control the third body thickness and related properties thus to establish the relationship between energy wear rate and third body properties for a given running in fretting wear contact configuration. This will achieve by investigating different debris trapping and untrapping contact geometries and by texturing the surface to control the debris flow evacuation and debris layer thickness.
Other aspects like contact size, contact pressure, displacement amplitude and sliding velocity will be examined. In situ observations (sapphire counterface) and dedicated postmortem analysis will be undertaken to estimate the third body thickness. Finally by comparing the obtained energy wear rates with the measured third body debris layer and eXtended Friction Energy Wear approach will be established linking the friction energy wear efficiency as a function of the contact configuration and related debris trapping properties of the fretting interface. This new wear formulation will be implemented in LTDS and LAMCOS numerical code and case studies provided by SKF and AIRBUS will be simulated to establish the stability of this new approach.








Project coordination

Siegfried Fouvry (LABORATOIRE DE TRIBOLOGIE ET DYNAMIQUE DES SYSTEMES)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partner

INSA Lyon - LAMCOS Institut National des Sciences Appliquées de Lyon - Laboratoire de Mécanique des Contacts et des Structures
LTDS LABORATOIRE DE TRIBOLOGIE ET DYNAMIQUE DES SYSTEMES

Help of the ANR 326,371 euros
Beginning and duration of the scientific project: - 48 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter