MT Indications after Stress Relief

[metalman8357]

Welding plain carbon steel with the FCAW process. The component is a 12' diameter x 2" wide ring fillet welded on both sides to the face of a large plate. We use dry powder to MT the welds before stress relief and then we use wet MT fluorescence after stress relief. We stress relief at 1130F for 8 hours. We are noticing that after stress relief about 80 MT linear indications are popping up at the toes of the weld all the way around the ring. Stress relief should help to relax residual stresses, not create indications.

Any ideas?

 

[metengr]

Is this weld undercut or actual cracks at the weld toes? Have you explored the depth of the MT indications? What was the preheat before welding?

 

[weldtek]

I realize this is after the fact but why would you use a more sensitive inspection method after PWHT? The goal should be to never, never introduce a new inspection method or technique, after PWHT. I even go to the extreme that I don't want to use a different technician for the examinations that are performed post SR. Of course sometimes it's unavoidable but I continually stress to my inspectors - no changes allowed in examinations performed before and after heat treat.

Sorry, that doesn't help your immediate issue, have you attempted to condition the areas by lightly grinding followed by WFMT?

 

[weldstan]

Differential heating of the ring & plate or moisture (water) entrapment could lead to stress high enough to initiate cracking.

 

[weldstan]

Delayed hydrogen cracking occurring between initial MT and PWHT could also have caused cracking.

 

[metalman8357]

Weldtek,

Is there something in your inspection methods that would not allow you to use a more sensitive technqiue after PWHT? We are inspecting to MIL-STD-2035 and as far as I'm aware there isn't any stipulation for this. This would be good ammo to have because as it stands now our inspectors are allowed to move freely between dry powder and WFMT. As you can see this is a huge problem. It woudl be much easier to weld repair before PWHT rather than when the part is in the final fabricated condition.

 

[Guest102023]

Given the thicknesses involved I think stress relief cracking could be a possibility. A donut weld represents the very worst case for residual stress.

Have a chemical analysis done on the cracked base metal, and search for Nb, V, Zr and Ti.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[weldtek]

I'm not familiar with the spec MIL you reference but the approach I recommend, and that I take for every job is that any NDE required after PWHT must be done prior to that heat cycle, and, I make every effort to ensure it is done, as it will be done later.
Any new variable introduced in the second inspection is an opportunity for an unwelcome surprise.
I also extend this principle to include imposing tighter tolerances in the first inspection. As an example, if RT is required before and after PWHT I don't allow certain indications that are at, or close to, the limit allowed by code. It isn't just a concern that the stress cycle will cause the indication to grow, but over the years I've found that even slight variations in technique can cause an indication to appear differently in radiographs that are taken later.

When an operator dependent inspection method such as UT is required it's a good idea to use the same technician for both examinations. You certainly want to use the same procedure for the second inspection. Illustrating, no one* would recommend using the conventional shear wave technique prior to and follow up with phased array and TOFD after.

* almost no one

You summed it up well in your last sentence.
It woudl be much easier to weld repair before PWHT rather than when the part is in the final fabricated condition.

 

[weldstan]

metalman 8357,
Did you mean 12 feet ("12'") in diameter or 12 inches in diameter?

 

[Duwe6]

Buff on these indications; don't be bashful. Use a 'Tiger Pad' rotary sanding disk, overlapping layers, 40 or 60 grit; dull the pad by using it on some steel for about 2-3 minutes first [new Tiger Pads are very aggressive]. If they don't buff out, the part is scrap - nothing to loose.

Keep an AC mag yoke on hand, buff off about 3-4 mils of thickness at the linear indication, then redo the MT. If it is 'heat check' it will have gone away, it is less than 2-mils deep initially, usually in the microns. If it is undercut, it will blend out. If it really is cracking, it will have significant depth >3 mils. Keep buffing and MTing until you find the bottom of them. Please come back with this data and I'll bet my favorite hat that one of us will be able to diagnose the problem and offer a 'fix'.

 

[JR97]

Sorry for late post.

Indications could be non-relevant from permeability differences. My experience is Wet UV inspection will produce permeability indications more readily than dry visible.