Rotordynamic Performance Predictions of Flexure Pivot Tilting Pad Bearings and Comparison to Published Test Data

Abstract

This paper presents performance predictions for a Flexure Pivot® tilting pad bearing (FPTPB) and comparisons to published test data. The FPTPB has pads that tilt about a web pivot. The pivot is designed to have little rotational stiffness, which improves its rotordynamic stability. The Reynolds equation for an isothermal and isoviscous fluid was used to calculate the film pressure using the finite element method. The Newton–Raphson method was used to determine the journal eccentricity, journal attitude angle, and pivot deflections (tilting angle, and radial and circumferential displacements) simultaneously. A small perturbation of the journal center around its equilibrium position was employed to calculate the stiffness and damping coefficients. The whirl frequency ratio (WFR), which is the ratio of the frequency of unstable whirl motion to the rotor threshold speed, was calculated using the predicted dynamic coefficients. The predictive model used was for a four-pad FPTPB with a journal diameter of 116.81 mm and axial length of 76.2 mm. Each pad had a preload of 0.25, pivot offset of 0.5, and web pivot thickness of 2.125 mm. The rotor speed and specific load were varied up to 16 krpm and 345 kPa, respectively. Predictions were performed for the load-on-pad (LOP) and load-between-pad (LBP) configurations. The results show that the predicted journal eccentricity and attitude angle decrease as the rotor speed increases. The direct stiffness coefficients increase as the rotor speed increases, but the cross-coupled stiffness coefficients change little. As the static load increases, the direct stiffness coefficient in the vertical direction increases, but the other stiffness coefficients change little. The damping coefficients are affected little by the rotor speeds and static loads. A comparison of the predictions with published test data shows that the predictive model slightly overestimates the journal eccentricities and underestimates the absolute values of the journal attitude angles. The predicted stiffness coefficients agree well with the test data. However, large discrepancies in the damping coefficients were observed between the predictions and published data. The effect of the pivot thickness on the rotordynamic performance of the FPTPB was also studied. Predictions were performed for changes in the pivot thickness up to ± 20 percent in increments of 10 percent from the original value of 2.125 mm. The results show that the predicted maximum pressure, journal eccentricity, and attitude angle decrease with decreasing pivot thickness, but the minimum film thickness increases. As the pivot thickness decreases, the direct stiffness coefficients, direct damping coefficients, and absolute values of the cross-coupled damping coefficients increase. The whirl frequency ratio (WFR), which was found to be in reasonable agreement with the test data, decreases significantly with decreasing pivot thickness, which suggests improvement in rotordynamic stability.