On the Effect of Supplied Flow Rate to the Performance of a Tilting-Pad Journal Bearing: Static Load and Dynamic Force Measurements

Rotating machinery relies on engineered tilting-pad journal bearings (TPJB) to provide static load support with minimal drag power losses, safe pad temperatures, and ensuring a rotor dynamic stable rotor operation. End users focus on reducing the supplied oil flow rate into a bearing to both lower operational costs and to increase drive power efficiency. However, a too low oil flow might significantly rise pads’ (Babbitt) temperatures whilst increasing rotor vibrations, even producing sub synchronous whirl frequency motions. This paper presents measurements of the steady-state and dynamic forced performance of a TPJB whilst focusing on the influence of supplied oil flow rate, below and above a nominal condition (50% and 150%). The test bearing has five pads, a slenderness ratio L/D = 0.4, spherical pivots with pad offset = 50% and pad preload ~ 0.40, and a clearance to radius ratio (C_r/R) ≈ 0.001 at room temperature. The bearing is installed under a load-between-pads (LBP) orientation and has a flooded housing with end seals. The test conditions include operation at various shaft surface speeds (15-85 m/s) and specific static loads (0.17 to 2.1 MPa). A typical turbine oil lubricates the bearing with a speed-dependent flow rate delivered at a constant supply temperature. Measurements obtained at a steady thermal equilibrium include the journal static eccentricity and attitude angle, the oil exit temperature rise, and the pads’ subsurface temperature rise at various locations, circumferential and axial. The experimental set up also includes a direct measurement of the torque using a strain gage in-line meter, and the bearing drag power loss follows from the product of angular speed x torque. Orthogonally mounted hydraulic shaker heads excite the test bearing with dynamic loads spanning frequencies to 300 Hz. Measurements of the ensuing bearing accelerations and displacements relative to the shaft produce the test system complex dynamic stiffness coefficients. Dynamic force coefficients; namely stiffness, damping, and virtual-mass, follow from curve-fits to the real and imaginary parts of the said complex stiffnesses. As expected, the measured drag power and the lubricant exit temperature rise depend mainly on shaft speed rather than on applied static load. A reduction in oil flow rate to 50% of its nominal magnitude causes a modest increase in journal eccentricity, a 15% reduction in drag power loss, a moderate raise (6°C) in pads’ subsurface temperatures, a slight increase (up to 6%) in the direct stiffnesses, and a decrease (up to 7%) in direct damping coefficients. Conversely, 1.5 times increase in oil flow rate causes a slight increase (up to 9 %) in drag power loss, a moderate reduction of pads’ temperatures (up to 3°C), a maximum 5% reduction in direct stiffnesses, and a significant 10% increase in direct damping. The paper also presents comparisons of the experimental results against predictions derived from a thermoelastohydrodynamic lubrication model. In conclusion, a 50% reduced oil flow rate only causes a slight degradation in the test bearing static and dynamic force performance and does not make the bearing operation unsafe for tests with surface speed up to 74 m/s. As an important corollary, the measured bearing drag power differs from the conventional estimate derived from the product of the supplied flow rate, the lubricant specific heat and the oil exit temperature rise.