Dual-frequency prediction using the crack closure concept

by | Oct 24, 2012

Researchers report on the prediction of low cycle, high cycle and combined cycle fatigue in turbine blades using the crack closure code FASTRAN.

Turbine blades in power generation or aircraft engines have to withstand a dual-frequency loading. A combination of centrifugal and thermal stresses produces a low-frequency loading, which is large enough to result in a low cycle fatigue (LCF) failure. On the other hand, vibrations due to imbalances, unsteady flows, and self-excitations cause a high-frequency loading, which is superimposed on the low-frequency loading. These vibrations usually induce only small stress amplitudes, and so typically result in high cycle fatigue (HCF) failures. The combined effect of both loadings is called combined cycle fatigue (CCF).

For the most common titanium alloy Ti-6Al-4V the linear summation model has in the past been successfully used for  predicting crack growth rates due to CCF. It is the simplest model and assumes that there are no load interaction effects. However, it has not been successful when load interactions or short crack effects occurred.

In new work Domnin Gelmedin and Dr. Karl-Heinz Lang from the Karlsruhe Institute of Technology, Germany, and Prof. James C. Newman Jr. from the Mississippi State University, USA, report on the prediction of fatigue lives with the aid of the crack closure concept. They used the fatigue crack growth structural analysis program FASTRAN, which was developed to predict fatigue crack growth under aircraft spectrum loading accounts for load interactions and short crack effects. Gelmedin, Lang, and Newman compared the predicted lifetimes with experimental fatigue lives over a wide range in fatigue lives.

The researchers carried out LCF, HCF, and CCF tests on a nickel-based superalloy at 650°C in air. Under combined LCF and HCF loading, they used block striations form on the fracture surface to complete an effective crack growth curve by applying the linear summation model. Crack growth lives starting from equivalent initial flaw sizes are calculated by the crack closure code and compared with experimental fatigue lives.

Even when the results show a good agreement between experimental and predicted life, the crack initiation and propagation in the superalloy under HCF loading differ from the way FASTRAN calculates. The crack closure concept cannot describe these interactions which lead to changes in the fracture surface morphology. Lower lifetimes are overestimated, indicating that the linear summation model is not valid for this material in this loading range.

Image: David Castillo Dominici / FreeDigitalPhotos.net