Effects of dressing parameters on grinding wheel surface topography

Abstract

Grinding is a critical manufacturing process and is often the only alternative when producing precision components or when machining brittle materials such as ceramics. Characterizing and modeling the surface finish in the grinding process is a difficult task due to the stochastic nature of the size, shape and spatial distribution of abrasive grains that make up the surface of grinding wheels. Since the surface finish obtained in grinding is a direct function of the wheel surface topography, which is conditioned by a single point dressing process, understanding the effects of dressing parameters on the wheel topography is essential. Therefore, the main objectives of this thesis are: 1) to experimentally characterize the three-dimensional surface topography of a conventional grinding wheel including attributes such as the abrasive grain height distribution, grain geometry and spacing parameters and their respective statistical distributions, 2) to determine the effects of single point dressing conditions on the three-dimensional wheel surface topography parameters and their distributions, 3) to model and simulate the three-dimensional wheel surface topography, and 4) to experimentally validate the wheel topography model. In this research, new and existing characterization methods are used to characterize the wheel surface and the individual abrasive grains. The new techniques include the use of X-ray micro-tomography (μCT) to obtain a better understanding of the grinding wheel's internal micro-structure, and a focus variation based optical measurement method and scanning electron microscopy to characterize previously ignored attributes such as the number of sides and aspect ratio of individual grains. A seeded gel (SG) vitrified bond conventional grinding wheel is used in the study. A full factorial design of single point wheel dressing experiments is performed to investigate the effects infeed and lead dressing parameters on the grinding wheel surface topography. A custom wheel indexing apparatus is built to facilitate precision relocation of the grinding wheel surface to enable optical comparison of the pre- and post-dressing wheel surface topography to observe wheel surface generation mechanisms such as macro-fracture and grain dislodgement. Quantitative descriptions of how each dressing parameter affects the wheel surface characteristics are given in terms of the wheel surface roughness amplitude parameters (Sp, Ssk, Sku) and areal and volume parameters (Spk, Sk, Vmp, Vmp, Vvc, Smr1) derived from the bearing area curve. A three-dimensional wheel topography simulation model that takes as input the abrasive grain height distribution and the statistical distributions for the various abrasive grain geometry parameters is developed and experimentally validated. The results of wheel characterization studies show that the actual abrasive grain height distribution in the SG wheel follows a beta distribution. The μCT work shows that the abrasives are polyhedral in shape, as opposed to the spherical or conical shapes commonly assumed in grinding literature. Grain spacing is found to follow a beta distribution while the number of sides of the grain and the grain aspect ratio are found to follow the gamma and the Weibull distribution, respectively. The results of the dressing study show that the lead dressing parameter has the strongest effect on wheel topography. Using statistical distributions for the key parameters (e.g. grain height, number of sides, grain spacing), a stochastic three-dimensional model is developed to simulate the wheel surface topography under different dressing conditions. The resulting model is shown to yield realistic results compared to existing models mainly due the fact that additional abrasive grain geometry parameters and more realistic assumptions of the different grain attributes are used in the model. It is shown that the model follows the overall wheel surface topography trends during dressing but has difficulty in accurately simulating some of the wheel characteristics under specific dressing conditions. The thesis then concludes with a summary of the main findings and possible future research avenues including extending the model to rotary dressing and simulation of wheel-workpiece interaction.