Cold Cracking in a Thick-Section Pressure-Vessel Weld
A 25 mm thick SA-516 Gr 70 pressure-vessel shell is welded with a cellulosic electrode using 180 A, 22 V, and 3 mm/s travel speed. The plate composition is 0.18% C, 1.25% Mn, 0.35% Si, 0.20% Cr, 0.10% Mo, 0.02% V, 0.10% Ni, and 0.10% Cu. No preheat is used; the shop reports that the plate is only 25 mm thick and that the weather is warm. The joint is radiographed 4 hours after welding and passes. Three days later, a longitudinal crack is found in the weld metal near the fusion line. The HAZ hardness is 360 HV10. The vessel is due to be hydrotested the next day.
1. Calculate the IIW carbon equivalent and assess the preheat requirement for this joint.
Read the worked answer
CE = C + Mn/6 + (Cr + Mo + V)/5 + (Ni + Cu)/15 = 0.18 + 1.25/6 + (0.20 + 0.10 + 0.02)/5 + (0.10 + 0.10)/15 = 0.18 + 0.208 + 0.064 + 0.013 = 0.465.
A CE of about 0.47 in a 25 mm section is not extreme, but it is high enough that hydrogen-assisted cold cracking is a real risk, especially with a cellulosic electrode and no preheat. A typical minimum preheat for this combination is about 100°C, and 150°C is prudent when the heat input is low and the hydrogen level is not controlled. The shop's argument that 25 mm is thin is not sufficient; thickness, CE, hydrogen, and restraint all act together. The 360 HV10 HAZ hardness supports a hard, crack-susceptible microstructure.
2. Calculate the arc heat input and judge whether it is appropriate for this thickness.
Read the worked answer
Heat input HI = eta x V x I / v, with eta about 0.8 for the arc process:
HI = 0.8 x 22 x 180/3 = 1056 J/mm = about 1.06 kJ/mm.
For a 25 mm thick, restrained joint with CE 0.47, 1.06 kJ/mm is on the low side. Low heat input increases the cooling rate, raises HAZ hardness, and promotes martensite. A more typical range for this thickness is about 1.5 to 2.0 kJ/mm, combined with preheat, rather than relying on heat input alone. The calculation also assumes a travel speed of 3 mm/s; if the welder runs faster, the situation is worse.
3. The radiograph passed 4 hours after welding but the crack appeared 3 days later. Explain this and recommend the immediate actions before hydrotest.
Read the worked answer
This is classic delayed hydrogen-assisted cold cracking. Radiography at 4 hours can miss a tight crack that has not yet opened. Hydrogen diffuses into the HAZ and weld metal, and cracking occurs hours to days later when the local hydrogen concentration and restraint stress are high enough. The 360 HV10 hardness shows a susceptible microstructure, and the low heat input and lack of preheat increased the risk.
Immediate actions: do not hydrotest. Hydrotesting a cracked vessel is dangerous and can propagate the crack. Excavate and repair the crack, or remove and reweld the affected area. Before repair, use preheat of at least 150°C, low-hydrogen electrodes that have been baked and stored properly, and heat input of 1.5 to 2.0 kJ/mm. Maintain interpass temperature. After welding, perform a hydrogen bake at about 250°C for 2 hours if the procedure allows, then inspect by UT or MT after at least 48 hours. Qualify the procedure with hardness and crack-susceptibility checks, and radiograph after the delay, not only at 4 hours.