PSI - Issue 38

Tuomas Skriko et al. / Procedia Structural Integrity 38 (2022) 393–400 Skriko et al. / Structural Integrity Procedia 00 (2021) 000 – 000

397

5

2.3. Measurements and fatigue tests Prior to fatigue tests, different measurements, such as local joint geometry with 3D laser scanning device and residual stresses with X-ray diffraction method, were performed to the specimens in order to have essential information for the numerical analyses and modellings of the laser-dressed non-load-carrying fillet weld joints. The fatigue tests of specimens were conducted in room temperature using constant amplitude cyclic loading. The applied stress range (Δ σ ) varied between different test specimens and the tests were executed at two different stress ratio levels, i.e. R = 0.1 and R = 0.5. For each laser-dressed specimen, the load and displacement values were monitored during the test from the servo-hydraulic test rig and strain gauges were used to measure and define the structural stresses and stress concentrations, respectively (Fig. 3a). In each fatigue test, the failure criterion was the total rupture of the test specimen (Fig. 3b). In general, tensile residual stresses were observed in the vicinity of weld toe areas (i.e. laser-dressed zones) and the local geometries of the weld toes were very smooth as a result of laser dressing treatment that can also be seen from the Fig. 2b. On an average, the toe radii were 3 - 4 mm in laser-dressed condition, whereas in normal as-welded condition, the toe radii are usually less than 1 mm and very often the interface between the weld metal and base material possess a sharp connection. More information and comprehensive data from the geometry and residual stress measurements, along with the experimental fatigue tests, of laser-dressed non-load-carrying cruciform joints are presented in the publication made by Skriko (2018).

(a)

(b)

Fig. 3. Experimental fatigue tests: (a) test set-up and (b) a fractured test specimen before and after total rupture.

3. Statistical and numerical analyses Statistical calculation methods presented by Hobbacher (2017) and Niemi et al. (2018) were applied for analyzing the experimental fatigue test results of laser-dressed non-load-carrying fillet weld joints. In addition, finite element (FE) models and analyses were made with appropriate meshing and elements according to the guidelines and recommendations given by Fricke (2012) and Baumgartner and Bruder (2013). The laser-dressed cruciform joints were examined and analyzed by means of four different method: nominal stress, structural hot-spot stress and effective notch stress (ENS) approaches as well as the 4R method, of which detailed information can be found in publications made by Nykänen and Björk (2016) and Ahola et al. (2020).