Film-Stiffness Characterization for Supercritical CO2 Film-Riding Seals

Low-leakage film-riding seals are a key enabling technology for utility-scale supercritical carbon dioxide (sCO2) power cycles. Fluid film-riding rotor-stator seals (operating with sCO2 as the working fluid) are designed to track rotor movements and provide effective sealing by maintaining a tight operating clearance from the spinning rotor. The operating equilibrium clearance of the seal is determined by the balance of opening and closing forces, while the rotor tracking ability of the seal at this equilibrium gap depends on the stiffness of fluid film and its insensitivity to expected distortions of the seal and/or rotor faces. Consequently, for designing a reliable film-riding seal, it is important to characterize the fluid film stiffness and its sensitivity to equilibrium gap, pressure ratio, and seal/rotor geometrical parameters. In this paper, we describe a non-rotating experimental test rig designed for measuring the fluid film stiffness of representative seal/rotor geometries along with the instrumentation and actuation mechanisms incorporated in the test rig. The rig consists of a pressure vessel, where the top cover forms the stator portion of the seal. Inside the vessel, a non-rotating dummy rotor floats on piezo-electric actuators that precisely locate the rotor surface relative to the seal bearing surface. Several rotor and stator configurations have been fabricated and tested. The rig has three independently controlled pressure cavities (supply pressure, high pressure and low pressures) and is designed to be run with medium-pressure air and with high-pressure supercritical CO2. We present typical non-dimensionalized test data from this rig with air as the working fluid. Furthermore, we present a 3D computational fluid dynamics (CFD) model with air for predicting the film stiffness, and compare the predictions of this model with the test data acquired from the rig. The CFD model predictions for film stiffness are in excellent agreement with the test data for the two tested configurations, with the CFD-based bearing pressures overpredicting the measured bearing pressures by about 10%. Unavoidable friction in the moving rig interfaces is one of the main reasons for this mismatch. Testing on this rig with sCO2 as the working fluid and comparison with sCO2-based CFD remain as future work.