Predicting creep rupture using damage mechanics

Creep rupture strength values are basis for design in many high temperature applications. The application of new heat resistant steel in power plants is a typical example where long rupture time data are required. Different assessment methods are available to get the material creep rupture strength in service condition. Mainly these methods are phenomenological in nature or mere procedures of data fitting. The main challenging issue using these methods concerns their reliability of extrapolation from short term creep data. Numerical methods based on continuum damage mechanics (CDM) are largely used to predict creep strain accumulation and creep rupture. Most of the proposed CDM models are based on the Kachanov's definition of effective stress. Damage accumulates with time (time-fraction) or strain (strain-fraction rule) as a result of void nucleation and growth, and rupture time is predicted to occur when damage reaches a critical value, characteristic for the material. The authors discussed elsewhere the influence of the damage evolution law on the strain accumulation in the tertiary creep stage [1-3]. For a given model formulation, damage model parameters can be identified to reproduce accurately the creep rupture time or the creep rate but usually poor results are obtained in predicting both the strain accumulation in the tertiary stage and the time to rupture. Often a very high value of the critical damage (close to 1) is required to match experimental data, which is not consistent with physical evidences of creep damage. However for time-limited applications a simple linear damage law can be used and a creep rupture formulation can be derived. In this paper the CDM has been used to formulate an engineering approach for creep rupture assessment of metals and alloys.