PSI - Issue 5

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

1039

5

The number represents 2.8 % of the weld segments tested, from a total of 15,875 radiographs of which 10,507 refer to tension members. The quality of welded joints is exceedingly low in bridges constructed immediately after World War II. Repeated tests on joints with cracks on many bridges undertaken over 5 to 8-year intervals did not indicate that the cracks were spreading or that new cracks were appearing. The structural-stress analysis given in the paper was limited to the eight oldest welded railway bridges in Poland. The spans of the bridges range from 9.00 to 32.40 m. They were tested from 1 to 4 times over the years 1958 – 1984. A total of 670 photographs were taken of the butt joints. Fig 3. shows a quantitative specification of the splice joints tested in each bridge with a subdivision for acceptable quality levels (upper histogram) and unacceptable quality levels (bottom histogram) in accordance with the requirements of EN ISO 5817:2014. The lower histogram also shows the number of radiographs of welds with cracks for each bridge. The largest number of cracks was in bridge number III, constructed in 1938. The percentage of radiographs of welds with cracks is 47.9 % of the total number of joint segments tested.

Fig. 3. Histogram of quality levels of butt joints and number of radiographs with weld cracks in each bridge.

All the cracks found were technological (hot) cracks which had appeared during welding. Longitudinal (101) linear slag cracks (3011), lack of side-wall and inter-run fusion (4011 and 4012) are included in flat welding imperfections (WI) with an elliptical cross section [Hołowaty and Wichtow ski (2015)]. In the bearing section of a weld weakened by imperfections, stress concentrates and initiates brittle and fatigue fracture. The distribution and magnitude of these stresses may be calculated in accordance to the equations given in Neiber (1947).

4. State of stress in butt welds

To calculate endurance limits for the bridges under discussion, stress analysis in the welded butt splices with cracks was carried out. The results of the analyses are given in Table 1. The characteristic values for stresses in the butt welds (on the axis for the cover plates in flanges and webs – values in numerators) were calculated, and also beyond the cover plates in flange and web plates (values in denominators). The calculations were carried out in accordance with