Effects of partition wall type, partition wall number and cavity depth on the leakage and rotordynamic characteristics of the pocket damper seal (PDS) were numerically investigated using a presented 3D transient CFD method based on the multi-frequency elliptical whirling orbit model. The accuracy and availability of this transient CFD method and the multi-frequency elliptical whirling orbit model were demonstrated with the experimental data of the experimental eight-bladed fully-partitioned pocket damper seal (FPDS). The leakage flow rates and frequency-dependent rotordynamic coefficients of PDS were computed for two types of partition wall (namely conventional PDS and fully-partitioned PDS), four partition wall numbers including the labyrinth seal (no partition wall) and six cavity depths including the plain smooth seal (zero cavity depth) at operational conditions with or without inlet preswirl and 15000rpm rotational speed. The numerical results show that the FPDS has the similar leakage performance and more superior stability capacity than the conventional PDS. The FPDS possesses slightly larger leakage flow rate (∼2.6–4.0% larger) compared to the labyrinth seal. Eight is a preferable value for the partition wall number to gain the best leakage performance of the FPDS with the least manufacturing cost. The FPDS possesses significantly larger stiffness and damping than the labyrinth seal. Increasing partition wall number results in a significant increase in the direct stiffness but limited desirable effect on the effective damping. The FPDS possesses the lowest leakage flow rate when the cavity depth is about 2.0mm. Compared to the plain smooth seal, the FPDS possesses larger positive direct stiffness and significantly less direct damping and effective damping. Increasing cavity depth results in a significant decrease in the stabilizing direct damping and the magnitude of the destabilizing cross-coupling stiffness. H = 3.175mm is a preferable value of the cavity depth for which the effective damping of the FPDS is largest, especially for the concerned frequencies (80–120Hz) where most multistage high-pressure centrifugal compressors have stability problem.Copyright © 2014 by ASME
[1]
John M. Vance,et al.
Fluid Seals and Their Effect on Rotordynamics
,
2010
.
[2]
Giuseppe Vannini,et al.
Rotordynamic Force Coefficients for Three Types of Annular Gas Seals With Inlet Preswirl and High Differential Pressure Ratio
,
2012
.
[3]
Dara W. Childs,et al.
Turbomachinery Rotordynamics: Phenomena, Modeling, and Analysis
,
1993
.
[4]
John M. Vance,et al.
Using A Damper Seal To Eliminate Subsynchronous Vibrations In Three Back-To-Back Compressors.
,
1995
.
[5]
Robert X. Perez,et al.
The Use Of Honeycomb Seals In Stabilizing Two Centrifugal Compressors.
,
1993
.
[6]
Jun Li,et al.
Investigations on the Rotordynamic Coefficients of Pocket Damper Seals Using the Multifrequency, One-Dimensional, Whirling Orbit Model and RANS Solutions
,
2012
.
[7]
Bruce M. Steinetz,et al.
Sealing in Turbomachinery
,
2006
.
[8]
John M. Vance,et al.
A Bulk-Flow Analysis of Multiple-Pocket Gas Damper Seals
,
1999
.
[9]
B. Lakshminarayana.
Fluid dynamics and heat transfer of turbomachinery
,
1995
.
[10]
John M. Vance,et al.
Rotordynamic Force Coefficients for a New Damper Seal Design
,
2007
.
[11]
John M. Vance,et al.
Rotordynamic Force Coefficients of Pocket Damper Seals
,
2006
.
[12]
Jun Li,et al.
Multiple Frequencies Elliptical Whirling Orbit Model and Transient RANS Solution Approach to Rotordynamic Coefficients of Annual Gas Seals Prediction
,
2013
.
[13]
John M. Vance,et al.
Machinery Vibration and Rotordynamics
,
2010
.