For the investigation of the tribological behavior of lubricated contacts, the choice and the calibration of the adopted asperity contact model is fundamental, in order to properly mimic the mixed lubrication conditions. The Greenwood/Tripp model is extensively adopted by the commercial software commonly employed to simulate lubricated contacts. This model, based on a statistic evaluation of the number of asperities in contact and on the Hertzian contact theory, has the advantage of introducing a simple relationship between oil film thickness and asperity contact pressure, considerably reducing the simulation time. However, in order to calibrate the model, some non-standard roughness parameters are required, that are not available from commercial roughness measuring equipment. Standard values, based on some limited experiences, are typically used, and a limited literature can be found focusing on how to evaluate them, thus reducing the predictivity of the model. In this contribution, the roughness profile of the surfaces of the conrod small-end and the piston pin of a high performance motorbike engine are measured, adopting an optical measurement equipment. An algorithm is developed in order to properly evaluate the Greenwood/Tripp non-standard roughness parameters and set an elastohydrodynamic analysis of the lubricated coupling. A complementarity form of the Reynolds equation capable of handling the cavitation problem is coupled with both Greenwood/Tripp theory and with an alternative complementarity formulation of the asperity contact problem. The results obtained adopting the two different models are critically discussed, comparing them with empirical evidences provided by small-end/piston pin coupling failures, occurred in a severe running condition test at an early stage of the design process of the engine.
