Surface and Subsurface Cracks in Rolling Contact Fatigue of Hardened Components

The competitive aspect of surface and subsurface fatigue crack propagation in hardened components subjected to rolling contact fatigue is highlighted, the former being greatly affected by the working conditions (in particular the presence of tangential stresses and lubricant), the latter depending mainly on the inclusions content and on the hardness profile. In order to determine which one of these kinds of damage is favoured, initial data consisting of contact load, rolling and sliding speed, theological properties of the lubricant, material hardness and inclusions content are necessary.

The concurrent role of asperities and Hertzian stress field in determining surface crack propagation is explained with the approach of the “quiescent zone.” calculating the stress intensity factors range in a contact cycle and considering the pumping effect of the fluid possibly present on the contact surface.

Inherent defects (especially oxides) are thought to be responsible for subsurface cracks origin and the Murakami formula for short cracks is extrapolated to describe their growth threshold, which also depends on the hardness and therefore on the depth in surface hardened components.

A crack propagation index is then defined as a ratio of applied to threshold stress intensity factor, both for surface and subsurface cracks. Evaluating this index for a general operating condition, it is possible to determine which damage mechanism is favoured, taking into account the decisive effect of the hardness profile.     

Study on the Life Distribution and Reliability of Roller-Based Linear Bearing

A study on the life distribution and reliability for a roller guide with cage was carried out with a total number of 90 test samples in two lots (N _s = 38 and 52), and different fatigue life distribution functions, such as the two and three parameter Weibull distributions, and the log-normal distribution were used for analyzing the test data. The basic dynamic load rating formula standardized by ISO in 2004 was also compared with the life test data in relation to the effect of crowning on both ends of the carriage raceway. This study revealed the following: (1) the best fit for the life test data was achieved by the three-parameter Weibull distribution with the third parameter being the minimum life γ, using a Weibull slope of m = 27/20, and a load life exponent of 10/3. The log-normal distribution came second, and the two-parameter Weibull distribution third. (2) The test data did not fit with the two-parameter Weibull distribution using a Weibull slope of 9/8, which had been used in the ISO linear bearing standard. (3) Fatigue failure on the roller guide was initiated near the starting points of the crowning on both ends of the carriage raceway. Almost 70% of the test specimens were found to have uneven and inclined contact situations with the rollers. (4) The basic dynamic load ratings evaluated theoretically fit well to the life test results, while the λ b _m factors were also observed to fall within a suitable value range.