Computational modelling of dynamically loaded journal bearings

Abstract

Most of the mechanical systems are expected to work in different regimes. In what concerns to journal bearings, this means that variations of rotating speed and of magnitude and direction of the applied load must be accounted for in the design stage. Examples include reciprocating machinery such as internal combustion engines, compressors and other industrial processing machinery. Lubricated joints are designed so that, even in the worst conditions, journal and bearing are not expected to come into contact. Two very good reasons for designing journal bearings in this way are to reduce friction and to minimize wear and rupture risks. Furthermore, the hydrodynamic fluid film developed plays a crucial role in the stability of the mechanical systems, due to its damping characteristics, which cannot be disregarded. Lubrication theory for dynamically loaded journal bearings is mathematically complex and, over the last few decades, several analytical approaches have been proposed. However, these models are often quite complex to be used on design stage or, else, based on simplifications that constrain their application to the most common situations. Thus, the main purpose of this work is to present a general computational methodology for modelling of journal bearings subjected to dynamic loads. In the process, the fundamental aspects related to the Reynolds’ equation are reviewed and the main assumptions and simplifications discussed. The proposed methodology mathematically combines the well known infinitely short and the long journal-bearings theories.