OPTIMIZATION OF BUSH BEARING LUBRICATION FOR REDUCING MAINTENANCE TIME AND IMPROVING SERVICE LIFE

Abstract

Bush bearings are critical tribological components widely used in heavy-duty rotating machinery to support high radial loads. To mitigate boundary friction, suppress adhesive wear, and prevent seizure, high-viscosity lubricating grease is extensively utilized. However, because grease exhibits a non-Newtonian yield stress, effective delivery across the sliding journal interface requires internal lubrication grooves. Machining these internal channels introduces a fundamental mechanical trade-off: while grooves establish necessary lubricant reservoirs, they simultaneously reduce the effective contact land area and act as localized stress concentration raisers. This research presents an integrated analytical, numerical, and experimental investigation into the predictive maintenance and structural mechanics of grease-lubricated bush bearings. Eight distinct bush bearing models are systematically evaluated across two groove pattern geometries—Straight/Straight grooves and Zigzag/Chevron grooves each iterated across four fine groove depths (0.25 mm, 0.50 mm, 0.75 mm, and 1.00 mm). Precision 3D parametric models are developed in SolidWorks Premium 2025, incorporating specialized transitional fillets to alleviate numerical stress singularities. Analytical design calculations establish baseline sizing (Vsolid = 24,836.38 mm³) and extract exact grease consumption capacities ranging from 2.09 mm³ to 338.61 mm³, demonstrating that the zigzag pattern achieves a 255% higher lubricant storage capacity than the straight groove at 1.00 mm depth, accompanied by an increase in total internal surface area from 5,027.15 mm² to 6,340.23 mm². A finite element simulation setup is established in ANSYS Mechanical, utilizing fixed outer housing press-fit boundary conditions and operational radial loads to evaluate equivalent von Mises stress, total structural deformation, and stress concentration factors (Kt). Furthermore, a dedicated experimental test-rig architecture is engineered, integrating triaxial piezoelectric accelerometers and a high-speed data acquisition system. A predictive maintenance framework is formulated using time-domain statistical metrics (RMS, Peak, Crest Factor, Kurtosis) and Fast Fourier Transform (FFT) spectral decomposition to track operational health and detect lubricant starvation in accordance with ISO 10816 standards. Structured placeholders are provided for ongoing ANSYS solver executions and experimental acquisitions, providing an end-to-end foundation for bearing health monitoring.