**5.9 Influence of creep**

Because creep effects have to be considered in thermo-mechanical loaded components to take into account stress relaxation phenomena and creep damage, single and multiple step creep tests were carried out. Due to aging effects the decreasing strain rate of the primary creep stage of the single step tests directly merges into the stage of tertiary creep with material softening and therefore increasing creep strain rates. Multiple step tests show, that neither strain nor time are suitable to describe the creep behaviour for that case. Therefore different tests with varying pre-exposure times at test temperature were conducted. The pre-aging time was chosen according to the total time in the multiple step tests. It can be

Comparison of Energy-Based and Damage-Related

Fatigue Life Models for Aluminium Components Under TMF Loading 339

of time-independent plastic deformation and time-dependent creep effects for the deformation behaviour at elevated temperatures is already known from Manson (Manson et al., 1971). At the same time this forms the basis of the *strain range partitioning* concept. A literature review yields a great number of material models which are able to describe the material behaviour for certain kinds of loading. According to (Christ, 1991) they may be classified according to the underlying approach as follows: empirical models, continuum-

The cyclic deformation behaviour is described by the ABAQUS® *Combined Hardening Model* dependent on temperature and ageing. The corresponding variables are the temperaturecorrected time and the current temperature. The necessary parameters can be correlated with the *Rp0,2* yield limit from the Shercliff-Ashby model (Shercliff & Ashby, 1990). For the calculation of different states of ageing a user subroutine has been developed. On the one hand it allows for an accumulation of the temperature-corrected time on the basis of the temperature-time curve for separate calculation steps, and on the other hand it is possible to calculate directly the state of ageing for a certain ageing time according to the local

The calculated hystereses and stress-time curves conform very well to the measured experimental data (fig. 8). Under TMF load both the asymmetry of the stress-strain

**Mechanical strain [%]**

The models available for describing the complex phenomena of thermo-mechanical fatigue range from engineering approaches to physically based models, thereby characterising the

Cycle 2

hystereses and the stress relaxation in the dwell time region are expressed correctly.

mechanical models, physically based models and multi-component models.

**6.2 Using the ABAQUS Combinend Hardening Model** 

maximum temperature.

**Stress**

**7.1 Basics and classification** 

Fig. 8. Comparison of calculated and measured hystereses

**7. Simulation of fatigue life behaviour** 

combined loading in differing complexity.

 **[MPa]**

seen, that the minimum creep strain rate of the single step test at 150 MPa is more than 300 times lower compared to the minimum creep strain rate in multiple step test and single step test with pre-aged specimen at the same stress level. Furthermore the test data of the multiple step test and test with pre-aged specimens show a very similar behaviour. Therefore the time at test temperature determines the minimum strain rate independent of strain (Minichmayr et al., 2005).

Fig. 7. Influence of HCF interaction on the OP-TMF lifetime
