Two Models for Predicting Fatigue Crack Growth


Article Preview

In this investigation two fatigue crack growth models based on the different physical assumptions were systematically analyzed. One model makes use of the size of the stable damage distribution zone near crack as the major contributor to the fatigue crack growth. This model is based on the concept that a material point failures and a new crack will form when the fatigue damage value of the material point reach the critical damage. The other model supposes that fatigue growth can be described as a process with sequentially breaking small volume elements behind the crack tip. The fatigue crack growth can be regarded as successive crack re-initiation over a critical distance. The fatigue crack growth rate can be determined as the ratio of the critical distance to the average life within critical distance. Both models use macro parameter to describe the microscopic mechanism. An elastic-plastic finite element analysis (FEA) was used to obtain the detailed stress-strain history of the notched component with a detailed consideration of the cyclic plasticity of the material using a robust cyclic plasticity model. The fatigue damage distribution and the average damage within the critical distance near the crack tip can be obtained by combining the fatigue damage parameter with the stress strain distribution from the finite element analysis. These two models were evaluated using the experimental results obtained from the crack grow experiments on compact specimens made from 16MnR. The predicting results using these two models correlate well with the experimental data. The results show that two models can well describe the notch effect on the fatigue crack growth.



Advanced Materials Research (Volumes 44-46)

Edited by:

Z.Y. Shen, M.N. James, W.D. Li, and Y.X. Zhao






B. X. Qiu et al., "Two Models for Predicting Fatigue Crack Growth", Advanced Materials Research, Vols. 44-46, pp. 917-924, 2008

Online since:

June 2008




In order to see related information, you need to Login.