| In this dissertation, the numerical model for the dynamic characteristics of a wavy thrust bearing is presented. Computing software tools with a user-friendly interface are developed to increase the efficiency of parametric analysis and design. The developed model, which is isothermal, incorporates wavy face geometry with variations of pad size, load magnitude, loading position, rotational speed, lubricant viscosity and wave amplitude. A Reynolds based procedure is used to simulate the dynamics of various designs parameters and operational conditions. The equilibrium position of the journal is evaluated based on the equilibrium of forces generated by the fluid film pressures and the external applied loads. A numerical small perturbation technique is applied to calculate the linear stiffness and damping characteristics of the bearing at the equilibrium position. The stability of the wavy thrust bearing is investigated linearly by solving the homogeneous equations of motion, and the results are represented using logarithmic decrement. A transient analysis of the journal motion is performed linearly using the linear bearing coefficients, and nonlinearly by coupling the forces generated by the lubricant film pressure with the initial conditions of the journal motion.; Using the developed procedure, a parametric study is carried out to examine the effects of the loading and operating condition, the bearing geometry, and the bearing clearance on the journal equilibrium position, the bearing stiffness and damping characteristics, and the bearing stability. The results show that the equilibrium film thickness plays the key role in the dynamic characteristics of bearing. In addition, the non-linear behavior of the wavy thrust bearing is analyzed in both the time and the frequency domains. The transient motions of the journal from both nonlinear forces and linear bearing characteristics are compared and examined using frequency response functions to determine the validity and accuracy of the linear stability theory. It is shown that, for the system close to the unstable regime, the linear model fails to predict accurately the dynamic performance and stability of the system while the nonlinear model provides a more realistic result. |
