PSI - Issue 13

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

24

3

2.2. Experimental setup and procedure Quasi-static tensile tests were conducted with the brazed AISI 304L/BAu-4 joints in the as-received and six corrosion conditions. Furthermore, brazed specimens in the as-received condition as well as after 3 and 6 weeks of corrosion were cyclically tested. Two tests were performed for each material condition and mechanical loading. In addition to the extensometers, the DIC was used for strain measurements within tensile and fatigue tests in order to identify the local damage mechanisms in dependence of the corrosion conditions. Using the DIC technique, the deformation state can be visualised with both a temporal and spatial resolution. A frequency-dependent triggered image acquisition was applied to continuously analyse the cyclic deformation behaviour during fatigue testing at a frequency of f = 10 Hz. After the test, engineering tangential strains  xx were computed in tensile direction for line elements (LE) with various lengths in order to analyse the influence of the gauge length on the strain values. Quasi-static tensile tests of the current study were carried out using a universal testing system (F = 100 kN) and an extensometer with a gauge length of 10 mm (GL10). Due to the mounting springs, the maximum length of the DIC line element is limited to 5 mm (LE5). In accordance with ISO 6892, the strain rate was controlled with ̇ = 0.00025 s -1 and with an estimated strain rate over the parallel length ̇ Lc = 0.00025 s -1 , respectively, up to the total strain of approx.  t = 9%. Subsequently, the crosshead separation rate was controlled with v c = 4.8 mm/min until failure. Fatigue tests were performed at a servo-hydraulic testing system (F = ±100 kN) with an extensometer with a gauge length of 12.5 mm (GL12.5), allowing strain measurements of up to ±40%. Cyclic ratcheting strains are well known in the context of asymmetric cyclic loadings due to the directional accumulation of inelastic strains. For evaluation of ratcheting effects, the total maximum strain  max,t , defined in ISO 12106 by  max,t =  m,t +  a,t with the total strain amplitude  a,t and the total mean strain  m,t , was considered. A time-efficient procedure based on stepwise load increase tests (LIT) was applied within this study. Hence, the maximum stress was increased stepwise by  max = 10 MPa each  N = 10 4 cycles with the stress ratio R = 0.1, the frequency f = 10 Hz and sinusoidal load-time functions, starting at  max = 50 MPa. The material response was evaluated with a DIC system and an extensometer as well as with electrical, magnetic and temperature measuring techniques. The experimental setup includes a crack growth monitor for alternating current (AC) potential drop measurements, a feritscope and thermal imaging cameras, Fig. 1d. The change in the deformation-induced electrical voltage  U was recorded in the gauge length, referring to the initial value at the beginning of the test. For a detection of the change in magnetic portion  ζ, which can be correlated with the plasticity-induced transformation from metastable austenite to martensite, the feritscope sensor was positioned at the base material. The thermal imaging camera allows a detection of temperature differences down to 20 mK. After testing, the change in temperature  T = T 1 - 0.5 ∙ (T 2 + T 3 ) was determined based on the temperatures at the centre of the gauge length (T 1 ) and at the shafts of the specimen (T 2 , T 3 ). After testing, the fractured surfaces were investigated using scanning electron microscopy (SEM), with a secondary electron (SE) and a back-scattered electron (BSE) detector . (Schmiedt et al., 2018a) The DIC system was successfully applied for quasi-static tensile tests of the brazed joints (BJ) in the as-received (0w.) and in six corrosion conditions (1w. to 6w.) in order to evaluate the local strain concentrations in the area of the brazing seam. Thus, the engineering tangential strains  xx are computed for line elements with lengths of 0.5, 1, 3 and 5 mm (LE0.5, LE1, LE3, LE5). For as-received brazed specimens, an effect of the gauge length within the elastic deformation range cannot be detected, Fig. 2a. Comparing the 0.5 and 5 mm line elements within the plastic deformation range, the difference in the total strain  t constantly increases with increasing stresses with e.g.  t = 0.2% at 300 MPa up to 5.0% at  UTS and 11.8% close to failure. Strain values of the extensometer with a gauge length of 10 mm (GL10) are well located within the range of the DIC values, Fig. 2a. Considering the LE with a length of 5 mm, the total strain close to failure significantly decreases from 78% down to 22-34% for the corroded specimens. Nevertheless, the influence of the gauge length on the strain measurements is more pronounced for the corrosion conditions, because of notch effects due to the local corrosive attack at the brazing seam. For the LE with a length of 1 and 5 mm, the differences in total strain of  t = 49% at  UTS and of  t = 117% close to failure were computed for the 6 weeks corroded specimen, Fig. 2b. (Schmiedt et al., 2018a) 3. Results and discussion 3.1. Tensile behaviour