Crack propagation at root toe using GTAW and SMAW Process

[jkkt87]

Dear All Master welding,

I have a project of welding DN300 thk 40mm A420 WPL6 (LTCS) 90 Degree elbow to a DN300 thk 40mm A333 Gr 6 straight pipe. The process of welding is GTAW(root using Hyundai ST-50G 2.4mm) and SMAW(hot pass and capping using welding rod oerlikon supercito 3.2mm). We have cut out and re-weld the joint as much as 3 times due to crack found at root toes of elbow side. The crack propagate from time to time. Even our welder have control the welding in such a way like pre-heat 95 degree before welding, dye pent/MPI after root or layer of hot pass to check for surface crack (found no crack) but after capping. PAUT scan after cooling, it found the crack at the root toes. What is the possible causing of HAZ crack? Is it material properties, stress or welding process problem?

[ttoks]

Dear All Master welding,

I have a project of welding DN300 thk 40mm A420 WPL6 (LTCS) 90 Degree elbow to a DN300 thk 40mm A333 Gr 6 straight pipe. The process of welding is GTAW(root using Hyundai ST-50G 2.4mm) and SMAW(hot pass and capping using welding rod oerlikon supercito 3.2mm). We have cut out and re-weld the joint as much as 3 times due to crack found at root toes of elbow side. The crack propagate from time to time. Even our welder have control the welding in such a way like pre-heat 95 degree before welding, dye pent/MPI after root or layer of hot pass to check for surface crack (found no crack) but after capping. PAUT scan after cooling, it found the crack at the root toes. What is the possible causing of HAZ crack? Is it material properties, stress or welding process problem?

I'm gonna be straight up at say it's pretty damn obvious you are an engineer and not a welder, because the jargon is all engineer speak, this is also one of those situations where you should know better because this is a material science thing, and I should really charge you a consultant fee of this.

This is a pre heat and heat input issue, A420 and A333 are both low temperature grades of carbon steel, are more prone to cracking, you have a few options.

Most piping codes, including ASME B31.3 and B31.1 allow for TIG welding out to 11.4mm of wall thickness, TIG welding when done within parameters is less likely to cause cracking of the root than running straight over the root with stick, at least a hot pass and one extra fill pass with TIG before switching to stick, also more pre heat OUTSIDE of the weld zone by using electric heating blankets with thermocouples to maintain a higher preheat (around 180 C) will help as this will drastically reduce the cooling rate of the weld as its welded.

The other and honestly more fool proof method of welding lower temp carbon steel is to use the Lincoln STT process which has very low heat input, and thus almost no chance of cracking in low temp steels, but this does require a great deal of operator skill as it is still a hard wire process and so is prone to lack of fusion if not done properly.

all of these require highly skilled personnel which will adhere to the WPS, but being that the job is with A420 and A333 in a fairly heavy wall that is what it will take.

[milomilo]

Did you do this weld to a qualified welding procedure?

[jkkt87]

Hi ttoks,

Thanks for your professional opinion and yes you are correct, i am a production engineer. You are really experience who will know whether someone is welder or not through the language they speak. Salute. i should buy you a drink. haha.

[jkkt87]

Yes. Our welder weld as per approved WPS.

[ttoks]

[QUOTE=jkkt87;8834946]Yes. Our welder weld as per approved WPS.[/QUOTE

I would guess the reason an approved WPS isn't working the real world is that the wps will be qualified to a range of thicknesses and pipe diameters, and generally the qualification weld is done on a smaller piece of pipe (thinner wall, smaller diameter) than the WPS covers, and even if the qualifying weld was done on exactly the same dimensions as the first weld, it would have been a short coupon that will heat soak, slowing the cooling rate of the test weld compared to the real world one.