Critical flaw size in silicon nitride ball bearings

Aircraft engine and bearing manufacturers have been aggressively pursuing advanced materials technology systems solutions to meet main shaft-bearing needs of advanced military aircraft engines. Ceramic silicon nitride hybrid bearings are being developed for such high performance applications. Though silicon nitride exhibits many favorable properties such as high compressive strength, high hardness, a third of the density of steel, low coefficient of thermal expansion, and high corrosion and temperature resistance, they also have low fracture toughness and are susceptible to failure from fatigue spalls emanating from pre-existing surface flaws that can grow under rolling contact fatigue (RCF). Rolling elements and raceways are among the most demanding components in aircraft engines due to a combination of high cyclic contact stresses, long expected component lifetimes, corrosive environment, and the high consequence of fatigue failure. The cost of these rolling elements increases exponentially with the decrease in allowable flaw size for service applications. Hence the range of 3D non-planar surface flaw geometries subject to RCF is simulated to determine the critical flaw size (CFS) or the largest allowable flaw that does not grow under service conditions.;This dissertation is a numerical and experimental investigation of surface flaws in ceramic balls subjected to RCF and has resulted in the following analyses:;Crack Shape Determination: the nucleation of surface flaws from ball impact that occurs during the manufacturing process is simulated. By examining the subsurface Hertzian stresses between contacting spheres, their applicability to predicting and characterizing crack size and shape is established. It is demonstrated that a wide range of cone and partial cone cracks, observed in practice, can be generated using the proposed approaches.;RCF Simulation: the procedure and concerns in modeling nonplanar 3D cracks subject to RCF using FEA for stress intensity factor (SIF) trends observed from parametrically varying different physical effects are plotted and discussed. Included are developments in contact algorithms for 3D nonplanar cracks, meshing of nonplanar cracks for SIFs, parametric studies via MATLAB and other subroutines in python and FORTRAN.;Establishing Fracture Parameters: the fracture toughness, K c, is determined by using numerical techniques on experimental tests namely the Brazilian disc test and a novel compression test on an indented ball. The fatigue threshold for mixed-mode loading, Keff, is determined by using a combination of numerical modeling and results from the V-ring single ball RCF test.;CFS Determination: the range of 3D non-planar surface flaw geometries subject to RCF are simulated to calculate mixed mode SIFs to determine the critical flaw size, or the largest allowable flaw that does not grow under service conditions. The CFS results are presented as a function of Hertzian contact stress, traction magnitude, and crack size.;Empirical Equations: accurate empirical equations (response functions) for the KI, KII, and K III SIFs for semi-elliptical surface cracks subjected to RCF as a function of the contact patch diameter, angle of crack to the surface, max pressure, position along the crack front, and aspect ratio of the crack are developed via parametric 3D FEA.;Statistical Probability of Failure: since the variability in mechanical properties for brittle materials is high a probabilistic investigation of variations in flaw size, flaw orientation, fracture toughness, and Hertzian load on failure probability is conducted to statistically determine the probability of ball failure for an existing flaw subjected to the service conditions. (Full text of this dissertation may be available via the University of Florida Libraries web site. Please check http://www.uflib.ufl.edu/etd.html)...