Computational fluid dynamics (CFD) studies for annular seals continues to gain popularity as a powerful tool for seal rotordynamic analysis, but the wide variety in setup and modeling choices without sufficient justification prevents more widespread use of these methods. This study applies the quasi-steady (QS) method for rotordynamic coefficient prediction to an incompressible grooved seal model in order to rigorously quantify the influence of various prominent modeling choices on prediction results. Variation of the upstream region configuration confirms the dependence of the cross-coupled stiffness on the development of circumferential velocity, which is a strong function of upstream geometry. An axial inlet upstream region is shown to be a sufficient approximation to a radial inlet with axial symmetry, representative of a back-to-back seal configuration. Significant stiffness variation is observed for particular downstream region configurations, though the additional pressure perturbation is mostly confined to the downstream region itself. When no downstream region is included, sensitivity to the amount of pressure profile blending enforced at the seal outlet plane is demonstrated, further underscoring the need for an experimentally accurate downstream geometry. This is the first paper dedicated to a quantitative sensitivity investigation of this nature specific to incompressible grooved seals that examines upstream and downstream region configuration, rotor centrifugal growth effects, and modeling whirl amplitude. Use of these results will foster more accurate application of the QS method for rotordynamic predictions of incompressible grooved seals, ultimately enabling more widespread use of CFD methods for general seals predictions.

This content is only available via PDF.
You do not currently have access to this content.