Development of an Experimental Test Facility to Measure Leakage Through Labyrinth Seals

by Mark A. Rodarte, Mark H. Anderson, Gregory F. Nellis, & Sanford A. Klein
University of Wisconsin - Madison, Solar Energy Laboratory

Click Here To Access The Complete Paper
Click Here To Access The Presentation

Supercritical carbon-dioxide Brayton cycles have potential advantages related to high efficiency and low cost. However, one potential disadvantage of these cycles is that the very high density fluid that is in contact with the high speed surfaces in the turbomachinery will lead to large windage losses. In order to control these losses it is necessary to reduce the pressure within the rotor cavity. The required pumping power is directly related to the effectiveness of the shaft seals that isolate the rotor cavity from the compression and expansion spaces. The primary objective of this research is the measurement of the flow of carbon dioxide through a labyrinth shaft seal. The behavior of supercritical carbon dioxide expanding through a shaft seal under two-phase, choked conditions cannot be predicted accurately using any currently available theoretical model. This investigation will enable a more rigorous analysis of the trade-off between windage and pumping power in order to determine the optimal rotor caviypressure.

This paper describes a test facility that has been designed and fabricated in order to accurately characterize the flow resistance associated with an arbitrary, static (i.e., non-rotating) seal geometry over a broad range of operating conditions. The static test section will be used to investigate the leakage rate into the rotor-cavity region for the large pressure drops that are expected in the S-CO2 Brayton cycle. The test section has been carefully designed in order to precisely control the flow geometry and therefore minimize the uncertainty associated with the annular area between the shaft and seal. Even minor deviations in this geometry will lead to large experimental uncertainty.

The test facility was designed to support upstream pressures in the range of 7-14 MPa with a minimum downstream pressure of 2 MPa. The test facility supports upstream temperatures in the 280-320K range and flow rates as large as 108 kg/hr. The upstream and downstream pressures and the upstream temperature are actively controlled using valves and heaters. The test facility has a two-phase downstream recovery system following the test section, which allows the working fluid to be cycled back through the compressor. This paper presents initial data obtained using high pressure air in order to demonstrate the capability of the facility. Initial data obtained with two-phase carbon dioxide are also presented.