PSI - Issue 13

Anke Schmiedt et al. / Procedia Structural Integrity 13 (2018) 22–27 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

26

5

Fig. 3. a) LIT and b) excerpt of the test with a DIC analysis of a brazed specimen in the as-received condition. (Schmiedt et al., 2018a)

The pre-corrosion leads to a significant influence on the fatigue behaviour with a reduction of the failure maximum stress  max,f (LIT) down to 240 MPa (46%) after 3 weeks and down to 220 MPa (42%) after 6 weeks, Fig. 4a. The forces are related to the gross cross-section of the brazed joints. Due to reduced fatigue stresses, ratcheting fatigue effects are less pronounced for the corrosion conditions with a decrease of  max,t at the failure down to approx. 1.5%, Fig. 4a. Within LIT, the change in AC voltage  U corresponds well with  max,t and a damage-induced progressive increase can be announced at approx. 85% of the respective failure maximum stress. As a result of the reduced deformation behaviour of the pre-corroded brazed joints, the maximum value of  U at failure is decreased from 330 to 40 mV. Thus, electrical measuring techniques are suitable to describe the ratcheting fatigue behaviour of the brazed joints in various corrosion conditions.

Fig. 4. a), b) LIT of brazed joints prior to and after pre-corrosions (Schmiedt et al., 2018a); c) DIC analysis for a pre-corroded brazed joint. The DIC system allows to visualise local strain concentrations in the area of the brazing seam for as-received and pre-corroded specimens, Fig. 4c. Locally increased strain values are more pronounced for higher fatigue stresses. For quantification, the total maximum strain  max,t within the LIT was determined for 0.5 to 12.5 mm line elements (LE0.5 to LE12.5), using the triggered DIC system, Fig. 3b and Fig. 4b. DIC strain values for 12 and 12.5 mm line elements are in very good accordance with the results of the extensometer (GL12.5). For the as-received brazed specimen, the comparison of the LE of 0.5 and 5 mm close to failure shows a maximum difference in strain values of  max,t ≈ 3%, Fig. 3b. Deviations  max,t < 0.5% are determined for  max < 450 MPa. For the 6 weeks pre-corroded specimen, a decrease of the LE length from 5 to 0.5 mm leads to  max,t ≈ 3% at 17·10 4 cycles. In accordance with the tensile tests, the influence of the gauge length on the strain values is more pronounced for notch-containing surfaces, when taking comparable strain levels into consideration. (Schmiedt et al., 2018a) 3.3. Fractographic analysis For an evaluation of the fatigue and corrosion fatigue damage processes, fracture surfaces and polished sections of fractured specimens were investigated using SEM. By applying a BSE detector, the stainless steel appears dark grey in contrast to the light grey gold-base filler metal. For an as-received brazed specimen, the fatigue crack initiated at an imperfection of the brazing seam close to the surface, at the top of Fig. 5a. After initiation, the crack propagated within the interfaces and the base material, respectively. In contrast, the final fracture occurred in the centre of the brazing seam with pronounced deformation characteristics, Fig. 5a. (Schmiedt et al., 2018a)