Lanterna Rally

Author:

Publisher:

Published: 2014

Total Pages: 241

ISBN-13:

This report is divided into four parts. First part of the report describes the methods used to measure and model the flow of supercritical carbon dioxide (S-CO2) through annuli and straight- through labyrinth seals. The effects of shaft eccentricity in small diameter annuli were observed for length-to-hydraulic diameter (L/D) ratios of 6, 12, 143, and 235. Flow rates through tooth- cavity labyrinth seals were measured for inlet pressures of 7.7, 10, and 11 MPa with corresponding inlet densities of 325, 475, and 630 kg/m3. Various leakage models were compared to this result to describe their applicability in supercritical carbon dioxide applications. Flow rate measurements were made varying tooth number for labyrinth seals of same total length. Flow rate measurements were also made for a stepped labyrinth seal similar to shaft seal found in Sandia National Laboratories research facility. The effect of eccentricity on flow through small diameter annuli was found to be minimal for the lengths typically found in labyrinth type shaft seals. There is an expected flow increase when moving a shaft from a concentric position to an eccentric position which is driven by a change in the fluid friction. This flow increase was found to be small for short annular orifices. An observed increase in flow rate of 3% was observed for short length annular orifices and was increased to 8.5% when the orifice length was increased beyond a distance equal to the developing entrance length and frictional effects were manifested. Flow rate measurements for a straight through labyrinth seal with three teeth were made at inlet pressures of 7.7, 10, and 11 MPa with corresponding inlet densities of 325, 475, and 630 kg/m3. Various labyrinth seal leakage models were applied to the data calculated to compare applicability. Applying the Homogeneous Equilibrium Model (HEM) with an experimentally determined discharge coefficient to predict the mass flow rate gave results with less than 5% error for the higher pressure cases of 10 and 11 MPa and less that 14% error for the lower pressure case of 7.7 MPa. The HEM model works well when the inlet condition chokes prior to entering the two phase region and begins to deviate when two-phase effects become more prevalent. Other models were unable to predict property changes along with poor response to changes in geometry due to their lack of complexity. A Stepped labyrinth seal was designed to mimic the geometry used in the supercritical flow research loop at Sandia National Laboratories. This provided a more complex geometry to further test the capabilities of the facility and validate models. The results showed that the data could be used to scale to larger diameters and apply to more practical geometries. Three-tooth and four-tooth cases were tested an inlet pressure of 10 MPa with a corresponding inlet density of 325 kg/m3. It was found that increasing the tooth number decreased the flow by 5% from the three-tooth case to the four-tooth case. Second part of the report describes the computational study performed to understand the leakage through the labyrinth seals using Open source CFD package OpenFOAM. Fluid Property Interpolation Tables (FIT) program was implemented in OpenFOAM to accurately model the properties of CO2 required to solve the governing equations. To predict the flow behavior in the two phase dome Homogeneous Equilibrium Model (HEM) is assumed to be valid. Experimental results for plain orifice (L/D ~ 5) were used to show the capabilities of the FIT model implemented in OpenFOAM. Error analysis indicated that OpenFOAM is capable of predicting experimental data within "10% error with the majority of data close to "5% error. Following the validation of computational model, effects of geometrical parameters and operating conditions are isolated from each other and a parametric study was performed in two parts to understand their effects on leakage flow.