Sammendrag
The possible sources causing weld metal hydrogen and oxygen pick-up during offshore hyperbaric tie-in TIG welding have been identified and simulated at a hyperbaric chamber pressure of 12 bar, i.e. 110 meter sea depth. The base material was supermartensific 13 % Cr stainless steel. Matching filler wire was used. The weld metal hydrogen and oxygen pick-Up from water vapor in the shielding and chamber gases has been investigated by girth welding ofpipes. Moist chamber gas seems to have insignificant effect on hydrogen and oxygen pick-up. The largest contribution is from moist shielding gas. Most of the hydrogen content in the supermartensitic welds is diffusible. By applying post-heat, it is possible to reduce the hydrogen content in the weld metal. Microcracks in the cap were observed for welds deposited with 14 mbar water vapor in the shielding gas (117 ppm) and above. Microcracks in the root welds were not observed, which was also confirmed for self restrained cracking tests. During constant load test at ISO MPa simulating hydrostatic testing ofpipelines, the microcracks grew, and for the weld deposited with fully moistened shielding gas (175 ppm), a macrocrack appeared in the centre of the capping pass longitudinally to the welding direction. Crack Tip Opening Displacement (CTOD) values based on Single Edge Notched Bend (SENE) testing were all low, and the high constraint of the SENE specimen did not show any effects of hydrogen on fracture toughness. CTOD values based on Single Edge Notched Tension (SENT) testing decreased with increasing shielding gas moisture contents. For specimens without microcracks it was possible to increase the CTOD (SENT) values by post-heat, close to the toughness observed for specimens deposited with dry shielding gas.