On Establishing the Limits of a Virtual Test Rig

Abstract

The use of a virtual test rig to numerically test turbomachinery hardware can be extremely cost effective if the results obtained are physically correct and relatively accurate. The literature clearly shows that a lot of emphasis has been placed on single component validation optimized for a single operation point. There are few studies, however, that have clearly documented the numerical issues surrounding the modeling of a complete stage of a centrifugal compressor across its operating range. This effort uses a generic low flow stage design to demonstrate the accuracy to expect from the current state-of-the-art technology found in both commercial and research computational fluid dynamics (CFD) software. Even effects stemming from secondary flow paths are considered in this study. For design and off-design operation toward surge, 360-degree transient calculations are compared to those obtained from using the steady state fixed-rotor approximation. Finally, all work is ultimately compared to detailed test data obtained from single stage testing.