DETERMINATION OF TOLERABLE FLAW SIZES IN FULL SIZE WELDED BRIDGE DETAILS

Twenty-four full size beams with welded details and three welded gusset details were fabricated from A36, A588 and A514 steels which met the 1975 AASHTO toughness specifications. The types of beam details tested were cover-plated beams, lateral attachments, transverse stiffeners, and flange transitions. These beam and gusset details were cyclically loaded at room temperature for 2 million cycles and then at temperatures minus 40 degrees F (minus 40 degrees C) and lower until rapid fracture occurred. The fracture resistance of each beam and gusset was estimated using Linear Elastic Fracture Mechanics and compared to the material toughness test results. Current material toughness and fatigue specifications were also checked for applicability to full scale beams. Results of the beam and gusset fracture resistance estimations were in direct correlation with the slow bend (one second loading), 3 point bend, material tests. The residual stresses had a significant contribution to the fracture resistance estimation. Category E of the current AASHTO fatigue specifications was found to be applicable to the 12 in. (304.8 mm) flange attachment. However, this category was observed to overestimate the fatigue strength of the full size cover-plated beams. Categories B and C were found to be applicable to the flange transition and transverse stiffener details, respectively. At the time of fracture most of the fatigue life of the welded girders was exhausted. All beams except those with cover-plated details equaled to exceeded their design fatigue life before brittle fracture occurred. These tests confirm the necessity of an adequate fatigue design in any fracture control plan for bridges.