24th International Colloquium Tribology - January 2024 81 Numerical and Experimental Analysis of the Tribological Performance of a DLC-Coated Piston Ring-Cylinder Liner Contact Thomas Lubrecht 1,2 , Nans Biboulet 1 , Antonius A. Lubrecht 1 , Johnny Dufils 2* 1 Univ Lyon, INSA-Lyon, CNRS UMR5259, LaMCoS, Villeurbanne, France. 2 Institut de Recherche En Ingénierie des Surfaces (IREIS) ICE-T platform, HEF GROUPE, Andrézieux-Bouthéon, France * Corresponding author: jdufils@hef.group 1. Introduction According to the “Fit for 55” European package, greenhouse gas (GHG) emissions should be reduced by 55 % in 2030 compared to 1990. In particular, CO 2 emissions related to transportation should be drastically reduced. In this respect, electric battery vehicles seem to emerge as a reference solution for passenger cars however it is poorly adapted for heavy duty transportation. Internal combustion engines, either hybridized or H 2 fueled, will seemingly play a role in the future of transportation. Therefore, the quest for reducing friction within internal combustion engines is still meaningful for efficient transportation. Surface coatings, such as Diamond-Like Carbon (DLC), may help improving the efficiency, reliability and sustainability of future internal combustion engines since they present excellent tribological properties. The application of DLC coatings to all parts of the crucial Piston-Ring/ Cylinder-Liner (PRCL) contact has not been widely studied yet. In this work, both experimental and numerical methods were developed in order to analyze the tribological performance of the DLC coated PRCL contact. 2. Materials and methods 2.1 Semi-analytical model The PRCL contact operates mainly in mixed and hydrodynamic lubrication regimes. A semi-analytical, time-dependent, line-contact, mixed lubrication solver was developed. The fluid and asperity load carrying capacities are computed separately (Figure 1). The full film lubrication model is based on a fast and simple line-contact, time-dependent, semi-analytical solver developed by Biboulet et al. [1]. The asperity contact part of the model is based on a “load-distance” curves computed from measured surface topographies. The load-distance curves are obtained from a numerical tool developed by Sainsot [2]. Contrary to the regular stochastic theories, this method relies on deterministic contact mechanics results using measured topographies. Figure 1: Illustration of the asperity and hydrodynamic load carrying capacities in the mixed lubrication modelling of the piston ring/ cylinder liner contact The solver allows for a rapid prediction of the contact friction forces accounting for oil starvation (geometrical or by lack of lubricant) as well as oil transport. The solver was validated using fully numerical solutions and measurements obtained on a line contact reciprocating tribometer. To overcome the model limitations related to the sample macro-geometry defects, analytical correction coefficients evaluated using the real geometry were introduced. 2.2 Experimental methods In order to validate the semi-analytical model, friction measurements were performed on a linear reciprocating tribometer in a cylinder on plane configuration. Both the cylinder and the plane were DLC coated (a-C: H). The upper specimen is a 20-mm long roller with a diameter of 14 mm which is fixed. The cylinders underwent an additional machining to mimic the limited geometry of a piston ring. A 24N normal load was applied by deadweights leading a maximum Hertzian pressure of 82 MPa. The lower specimen is mounted into a lubricant reservoir, which is set in reciprocating motion by means of a crank/ conrod mechanism driven by an electric motor. The mean sliding speed was varied between 5 mm/ s and 100 mm/ s. In order to be close to the PRCL operating conditions and because the sliding speed was low, a 10W-60 grade oil with a high viscosity at room temperature was selected. 82 24th International Colloquium Tribology - January 2024 Numerical and Experimental Analysis of the Tribological Performance of a DLC-Coated Piston Ring-Cylinder Liner Contact Additional experiments were performed on a PRCL testbench using real engine parts. Piston rings mounted onto a piston are set in reciprocating motion inside a full cylinder liner thanks to a crank/ conrod system powered by an electric motor. The piston underwent an additional machining in order to remove the skirt and only measure the piston rings/ cylinder liner friction. The liner is assembled into a liner holder itself mounted onto a piezoelectric force sensor centered onto the piston axis in order to measure the piston rings/ cylinder liner friction force. The PRCL contact is lubricated on the inner surface of the liner by means of a nozzle located at the bottom end of the liner. The experiments were carried out at low rotational speed (from 100 to 400 RPM) with a 10W-60 grade oil at 50-°C in order to mimic the piston ring lubrication at higher speeds and higher oil temperatures. The experiments were performed with only one piston ring (a compression ring) and with and without DLC coatings on the ring and the liner. 3. Results and discussion 3.1 Validation of the semi-analytical model The semi-analytical model was able to efficiently predict the friction force both in mixed and hydrodynamic lubrication regimes. Figure 2(b) shows a comparison of the friction force predicted by the model and the friction force experimetal measurement. At high speeds, the transient term induces a damping effect of the lubricant which generates load carrying capacity and thus a non zero film thickness (squeeze effect). Friction and film thickness predictions with and without the correction coefficients are distinctly different. Systematically, a lower film thickness (Figure 2(a)) is predicted for the real geometry, which is consistent with the coefficient theory. 3.2 Effect of DLC on the PRCL contact and comparison with the model In the experiments performed on the PRCL testrig, a 10 to 60% reduction in friction at the top and bottom dead centers was obtained compared to the uncoated case as well as an improved wear resistance. The experimental results were compared to results from the semi-analytical model. There is good fit in the hydrodynamic regime but not in the mixed lubrication regime (at the top and bottom dead centers) probably due to the waviness defects on the rings and the liner which were not considered in the PRCL modelling so far. Figure 2: (a): film thickness (black dotted line: Moes and Venner solution, blue: 1d transient model, red: upgraded model with correction coefficients); (b): friction force (plus signs: measurement, black dotted line: Moes and Venner solution, blue: 1d transient model, red: upgraded model with correction coefficients) 4. Conclusions A solver allowing for a rapid prediction of the PRCL contact friction forces accounting for oil starvation and oil transport in mixed lubrication was developed and validated using a linear reciprocating tribometer in a cylinder on plane configuration in which the cylinder was machined to mimic a piston ring shape. In the experiments on the PRCL testrig, an excellent tribological performance of the DLC/ DLC compression ring/ liner contact was observed compared to the uncoated contact. However, the model prediction did not match well with the experiments in the mixed lubrication regime probably because of the waviness defects of the parts not being considered in the model. References [1] N. Biboulet and A. A. Lubrecht, Tribology Letters, vol. 70, 2022. [2] P. Sainsot and A. A. Lubrecht, Proceedings of the Institution of Mechanical Engineers, Part J: Journal of Engineering Tribology, vol. 225, no. 6, pp. 441-448, 2011.N.