Static Performance Characteristics and Rotordynamic Coefficients for a Four-Pad Ball-in-Socket Tilting Pad Journal Bearing

Static performance characteristics and rotordynamic coefficients were experimentally determined for a four-pad, spherical-seat, tilting-pad journal bearing in load-between-pad configuration. Measured static characteristics include journal static equilibrium position, estimated power loss, and trailing-edge pad temperatures. Rotordynamic coefficients were determined from curve fits of measured complex dynamic-stiffness coefficients as a functions of the excitation frequency. A frequency-independent [M]-[C]-[K] model did a good job of fitting the measurements. Test conditions included speeds from 4 to 12 krpm and unit loads from 0 to 1896 kPa (0 to 275 psi). The bearing uses cool inlet oil to decrease the pad operating temperatures and increase the bearing’s load and speed capacity. The bearing has a nominal diameter of 101.78 mm (4.0070 in). Measurements indicated significant bearing crush with a radial bearing clearance of 99.63 μm (3.92 mils) in the axis 45° counterclockwise from the loaded axis and 54.60 μm (2.15 mils) in the axis 45° clockwise from the loaded axis. The pad length is 101.60 mm (4.00 in), giving L/D = 1.00. The pad arc angle is 73°, and the pivot offset ratio is 65%. Testing was performed using a test rig described by Kaul [1], and rotordynamic coefficients were extracted using a procedure adapted from Childs and Hale [2]. A bulk-flow Navier-Stokes model was used for predictions, using adiabatic conditions for the fluid in the bearings. However, the model assumes constant nominal clearances at all pads, and an average clearance was used based on measured clearances. Measured static eccentricities and attitude angles were significantly lower than predicted. Attitude angles varied from 6° to 39° and decreased with load. Power loss was well-predicted, with a maximum value of 25 kW (34 hp). The maximum detected pad temperature was 71°C (160°C) while the temperature rise from inlet to exit was over-predicted by 8°C (14°F). Direct stiffness and damping coefficients were significantly over-predicted, but the addition of a simple pivot-stiffness in series with the measured stiffness and damping values vastly improved the agreement between theory and experiment. Direct added masses were negative to a higher degree for M_yy (y load direction) at low speeds and increased with speed. With the exception of M_yy at zero load, they became positive before reaching 8,000 rpm. Although significant cross-coupled stiffness terms were present, they always had the same sign, producing a whirl frequency ratio of zero and netting unconditional stability over all test conditions.