PSI - Issue 42

Kai Donnerbauer et al. / Procedia Structural Integrity 42 (2022) 738–744 Kai Donnerbauer / Structural Integrity Procedia 00 (2019) 000 – 000

743

6

The procedure is making use of relationships by Morrow (1964), Manson (1953), Coffin (1954) and Basquin (1910). It is described in more detail in Bill et al. (2021) and Acosta et al. (2021). Total strain S-N curves are calculated by equation (1).

, M e n

−

1

−

, M e

5

1

n

5

1



+

n



+

(2 )

(2 )

B N

C N

a 

= 

+ 

(1)

, M p

,

t

f

f

Slopes B and C characterizing the elastic and plastic parts of the material’s strain-life behavior under constant amplitude loading are evaluated from the data of two CATs using the Basquin (elastic strain) and Manson-Coffin (plastic strain) relationships respectively. Exponents n M,e and n M,p are derived from a relationship proposed by Morrow (1964) and most commonly used to describe the stress amplitude as a function of plastic strain amplitude. With StrainLife though it is possible to calculate the exponents from NDT data and therefore introduce information about microstructural changes. Results based on stress amplitude, E OCP,a and Hall voltage are shown in Fig. 5.

Fig. 5. StrainLife results from stress amplitude, open circuit potential and Hall voltage.

The total strain S-N curves determined fit very well to the results obtained from CATs. This holds true for tests in distilled water and in air, indicating that the strain correlation works well and that there is no influence onto fatigue life of AISI 347 from testing in distilled water. Beyond 2·10 3 cycles to failure all three curves fit very well to each other. CAT results at high strain amplitudes match best with the evaluations based on open circuit potential data. Ferromagnetic phase volume and mechanical material properties like yield strength of specimens change a lot during testing with higher strain amplitudes due to increasing martensitic transformation with increased loading. This could be a reason for obtaining a better result through the results from an NDT method, being less dependent on martensitic transformation and strain hardening. To confirm this, further CATs should be performed and additional insights into E OCP and the respective NDT methods should be generated. 4. Conclusions and outlook Nondestructive testing (NDT) methods based on magnetic as well as electrochemical measurements have been successfully applied to characterize the cyclic behavior of AISI 347 steel in total strain-controlled constant amplitude and strain-increase tests in distilled water. With open circuit potential amplitude a parameter has been suggested, which is sensitive to surface related microstructure changes and loading state. The conclusion is, that data obtained through NDT-based parameters can become a strong alternative to the classical mechanical parameters traditionally used for fatigue life evaluations. This holds true when the NDT parameter correlates better with the various microstructural changes being relevant to the fatigue process and should therefore be chosen with the respective