Assessing Reynolds and Jakobsson–Floberg–Olsson models for infinitely long water-lubricated partial journal bearings: Static and dynamic performance predictions compared to computational fluid dynamics analysis

This study evaluates the validity of two numerical models in predicting the performance of infinitely long water-lubricated partial journal bearings (IWLPJB): the non-mass-conserving Reynolds model (RBC) and the mass-conserving Jakobsson–Floberg–Olsson model (JFO). Building upon previous research on inadequately lubricated partial, long water-lubricated bearings, this work employs an efficient modeling approach (EMA) to implement these models, with results compared against computational fluid dynamics (CFD) analysis. A comprehensive evaluation of numerical models under realistic operating conditions is provided, specifically addressing the negative pressure cavitation phenomena of water-lubricated systems. The analysis focuses on conformal surfaces of IWLPJB, examining both static and dynamic performance predictions at the high eccentricity ratios typically encountered in industrial applications. The study examines both untextured and textured cases, assessing bearing performance parameters including maximum hydrodynamic pressure, attitude angle, load-carrying capacity, coefficient of friction (COF), and eight dynamic coefficients. The results indicate that for untextured and partially textured cases, both models provide reliable predictions of the performance parameter values at the studied high eccentricity ratios (0.9–0.98), despite RBC's inaccurate prediction of the cavitation region. Additionally, the differences between the two models are not sensitive to the studied texture depths (7.8–52 μm). For the fully textured case, although prediction errors increased, both models remain applicable for a wide range of performance parameters except for the COF. Notably, this study reveals that the predictive capability of the RBC model for both static and dynamic performance, particularly in textured cases, has been somewhat underestimated in previous research.