PSI - Issue 5

Hołowaty J. et al. / Procedia Structural Integrity 5 (2017) 1035 – 1042 Wichtowski and Hołowaty / Structural Integrity Procedia 00 ( 2017) 000 – 000

1041

7

5. Fatigue tests on butt welds with cracks

For a general endurance assessment of welded butt splices in old bridge structures, laboratory fatigue tests on specimens with butt welds showing internal cracks were performed [Wichtowski (2002), Wichtows ki and Hołowaty (2016)]. The regression line and the results of fatigue tests on eighteen specimens with welded butt splices showing internal cracks are given in Fig. 4. The average infinite life fatigue strength value was calculated from the regression line equation in a logarithmic scale for x = lgN i for 2·10 6 cycles and is equal to Z rj = 89.8 MPa. This value is 28 % of the yield strength for the steel tested f y = 320 MPa and each time it is larger than the service stress, whose maximum value is  ser,max = 74.4 MPa (see Table 1). An approximation of fatigue classes to Eurocode 3 is given in Wichtowski and Hołowaty (2016) for different specimen conditions and joints . The welds were executed manually in a down-hand position using EA1.46 acid electrodes with a diameter of 4.0 mm. The fatigue tests were performed in a pulsar machine at a frequency of 300 cycles per minute using a one sided elongation cycle with six levels of nominal stresses: 140, 130, 120, 100, 80 and 50 MPa with stress ratio R = 0.1. Radiographic tests (RT) on the butt welds from 18 specimens showed that, in addition to cracks of length from 14 to 145 mm there were also gas cavities and clustered slag inclusions with, in two cases, linear slag inclusions and a segmental lack of fusion. The imperfections in the specimens are considered to be inherent in early welded railway bridges which have a rather low weld quality.

Fig. 4. Linear regression and fatigue test results on 18 specimens with weld imperfections.

6. Summary

In-situ investigative analysis of a bridge structure allows a real service load model to be determined and degradation changes in the material mechanical properties due to ageing to be calculated. In radiographic testing of the 8 oldest railway bridges in Poland, the welded butt splices were found to be of very low quality, with 69.3 % of the butt weld segments tested featuring welded imperfections of quality level D and higher. On 124 radiographs, cracks in the welds were found. Multiple testing did not uncover any further increase in the existing cracks. A welded imperfection in a joint is permitted if, during operation, the defect will not enlarge and will not limit the resistance and rigidity of the structure.