No abstract

Seal stability is often evaluated by Whirl frequency ratio (WFR) and Effective damping coefficient (C eff) calculated on the assumption of synchronous whirl. However, the natural frequency of the rotor system must be used for calculation of WFR and C eff when determining self-excited vibration. This paper discusses the evaluation of seal stability using WFR and C eff. First, the stability analysis is performed by FEM for a simply supported Jeffcott rotor model, having a seal near the disk. Next, WFR and C eff are calculated for the following two frequencies; (1) the natural frequency of the rotor system, (2) rotational frequency. As the reference of stability, the logarithmic decrement is used for comparing with WFR and C eff. Effective damping coefficient (C eff) calculated using the natural frequency of the rotor correctly estimates the seal stability. Whirl frequency ratio (WFR) for the natural frequency of the rotor can judge the rotor stability, but cannot evaluate how good the seal is. Both C eff and WFR calculated by using the rotational frequency cannot give correct criterion for the rotor stability.

This study presents an experimental evaluation of the dynamic characteristics of the circular arc spring damper (CASD), which attenuates the vibration of rotating machinery. A major advantage of CASDs over conventional Squeeze Film Dampers (SFDs) with squirrel-cage springs is their compactness and low weight. However, the basic characteristics of this type of damper, including the influence of the geometries, added mass coefficients, and cross-coupled terms of dynamic coefficients have not been investigated. To clarify these characteristics, a series of excitation tests was conducted on three types of CASDs to identify their dynamic coefficients. Tested dampers have the same outer diameter and damper width but different arc patterns and radial clearances. All dampers were tested in both open-end and end-sealed configuration. The influence of the end-seal clearances was also examined. The following results were obtained: (1) 4-arc type CASDs have greater damping than that of 2-arc types; (2) CASDs have a considerable amount of added mass coefficients, especially in the end-sealed condition; (3) Smaller end-seal clearances make the damping and added-mass coefficients significantly larger; (4) A large level of damping can be produced with dashpot configuration (large radial clearances and small end-seal clearances), though the added-mass effect becomes more prominent; (5) Cross-coupled terms are very small in centered / small amplitude motion. These new findings can be utilized for the design and application of CASDs to real rotating machineries.

This study presents an experimental evaluation of the dynamic characteristics of the circular arc spring damper (CASD), which attenuates the vibration of rotating machinery. The CASD is a novel type of fluid film damper that has two or more arc-shaped beams and dual slits (radial clearances) formed by wire electric discharge machining (WEDM). A major advantage of CASDs over conventional Squeeze Film Dampers (SFDs) with squirrel-cage springs is their compactness and low weight. The concept of CASDs was originally proposed by Kanki et al. in 2005. However, the basic characteristics of this type of damper, including the influence of the geometries, added mass coefficients, and cross-coupled terms of dynamic coefficients have not been investigated. To clarify these characteristics, a series of excitation tests was conducted on three types of CASDs to identify their dynamic coefficients. The features of the newly developed test rig and experimental results are presented in this study. Tested dampers have the same outer diameter (160 [mm]) and damper width (44 [mm]) but different arc patterns (2-arc or 4-arc) and radial clearances (about 0.3 or 1.2 [mm]). All dampers were tested in both open-end and end-sealed configuration. The influence of the end-seal clearances (about 0.1 to 0.4 [mm]) was also examined. The following results were obtained: (1) 4-arc type CASDs have greater damping than that of 2-arc types; (2) CASDs have a considerable amount of added mass coefficients, especially in the end-sealed condition; (3) Smaller end-seal clearances make the damping and added-mass coefficients significantly larger; (4) A large level of damping can be produced with dashpot configuration (large radial clearances and small end-seal clearances), though the added-mass effect becomes more prominent; (5) Cross-coupled terms are very small in centered / small amplitude motion. These new findings can be utilized for the design and application of CASDs to real rotating machineries.

A circular arc spring damper (CASD) is a recently proposed fluid-film damper that has two or more arc-shaped centering springs and dual radial clearances formed by wire electric discharge machining. CASD requires less space and weight than a conventional cage-centered squeeze film damper (SFD). It provides linear stiffness and stable damping force in rotor-bearing systems to attenuate vibration due to imbalance or to improve rotordynamic stability. The authors have been investigated the dynamic characteristics of CASD in component-level experiments. However, their performance and applicability to real machines have not been confirmed in system-level experiments. Additionally, a theoretical means of evaluation for CASD should be established to predict its dynamic coefficients, and to understand the mechanism of dynamic force generation. In the first part of this study, a numerical evaluation method using two-way fluid-structure interaction analysis and its theoretical background is presented. Transient structural analysis and fluid-film flow analysis with a simple homogeneous cavitation model are coupled in the commercial multi-physics platform ANSYS. The accuracy of the method was validated by comparing the damping and added-mass coefficients with results from previous experiments. Furthermore, several aspects of the force generation mechanism, and the difference from conventional SFD were studied numerically. The second part of the study addresses the application of CASD in a multi-stage centrifugal compressor. A combined 4-inch diameter, 5-Pad tilting pad journal bearing (TPJB) with 4-arc type CASD was newly designed and manufactured. To prove the applicability of the developed damper bearing, a series of rotating tests was conducted at a high-speed balancing facility with a full-scale dummy rotor with a critical speed ratio of approximately 3.1. The measured unbalance response showed a much lower amplification factor than that of the conventional TPJB without the damper, which infers a significant improvement in the stability. The measured responses agreed with the rotordynamic analysis, which uses the dynamic coefficients of CASD derived from the proposed numerical evaluation method.