Stress-Corrosion Cracking TG vs. IG

[SMF1964]

I have a 347 stainless steel superheater tube that is sensitized. The microstructure shows the grain boundaries even before I etch due to the really big carbides along them and under the electron microscope some of these carbides are niobium rich and others are chromium rich. No surprise.

The tube failed due to an axial, ID initiated, transgranular and branching crack that looks like the usual Chloride induced type of SCC.

The cracking is transgranular. The structure is sensitized. Can I make the argument that the cracking must have been present prior to the steel becoming sensitized because the cracking is transgranular and not intergranular?

 

[EdStainless]

No, you cannot make the argument of a prior exisitng crack.
I would dout that there was any prior defect.
CSCC in sensitised 300 stainless can have crack patterns that are of any type.


= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[SMF1964]

Thanks for the answer. I was hoping for the other answer, but I thank you, nonetheless. I guess today I'm the bug, not the windshield.

 

[metengr]

Most of my experience with failure analysis of austenitic stainless steel boiler tubes is TGSCC not IGSCC from entrapped condensate in tube bends and adjacent to external welded attachments.

 

[SMF1964]

This failure is about 8 feet above the bottom of the loop and is more than a foot away from the nearest attachment. This unit and its two "sister" units at another of our plants have a history of IGSCC (not TGSCC) in the division panel SH lower loops/U-bends, but no failures in the final superheater (where this tube came from).

 

[metengr]

Did you perform any explosive cleaning in this area, recently? Also, is there any tube swage transitions in this location?

 

[SMF1964]

Explosive Cleaning: Unknown, I will investigate. Failure doesn't look like one, but you never know... I'll let you know.

Swage transition: none present.

 

[SMF1964]

No explosive deslagging in this area.

 

[metengr]

SMF1964;
Well, you ruled out several possibilities that I could think of off-the-top of my head. Still thinking. I would rule out any low cycle fatigue because of the orientation of the crack; this type of crack normally occurs external and is circumferential versus axial in orientation.

Based on your 3rd post, where you described the location of failure, I would rather doubt this is even in-service TGSCC because you have no means for condensate being up this high in the loop during boiler off-line conditions. One other possibility is you had flooded these pendants with untreated water (which is still entirely possible) when they were installed. I always keep options open until I can rule it out. I have heard where replacement steam circuits were installed and squeezed with untreated water to satisfy the hydrotest or a pressure test installation requirement.

I don't suspect a ductility or creep rupture problem with the 347 material because you would see intergranular fissures versus transgranular cracking. One other item to consider is an ice plug from a freeze-up during a forced outage in winter (if you are located in this climate). I ran into this situation some time ago with one of our CE units that tripped on an extremely (bitter) cold day. All it took was one forced outage on a weekend while the boiler operator left the ID/FD fans running. The inside of the boiler was so cold we needed to bring in temp heat for repairs. We could not drain the SH tube assemblies at the time of the freeze-up and after about 6 months we started to see tube leaks caused by axial cracks that initiated on the tube ID surface. Some axial cracks were part thru-wall – all were transgranular at first. The cracks that remained after the freeze up propagated from fatigue until failure from stress rupture, over time. It took up about 6 years of operation to squeeze out the SH tubes that were damaged from ice plugs.

 

[unclesyd]

I would look at the possibility of Differential Thermal Fatigue in the tube. This type failure will mimic SCC. I would look for striations in the fracture surface.

 

[SMF1964]

metengr - I would expect a freeze-plug related failure in austenitic stainless steel to still contain microvoid coalescence as the fracture morphology - this crack shows a transgranular fracture mode with no microvoids visible.

unclesyd - would this DTF failure mode produce the extensive branching that I see in this failure? I see thermal fatigue failures a lot in our oil gun tips and waterwalls (from excessive use of water cannons for deslagging), and that cracking is strictly a single crack plane or multiple cracks that join into a single crack plane, not a crack that splits into multiple branches. Can you comment?

 

[unclesyd]

I seen extreme branching if DTF in several failures involving Austenitic SS.
The worst case of cracking with branching was in a tail gas line from a H2 plant where water was being injected by means of desupheating quill. This cracking spiraled down the pipe for about 10 pipe diameters. The cracking highly branched off multiple and lengthy main cracks. The cracking was caused from the desupheating quill being installed pointing to the wall versus being actually aligned. These cracks were internal and had penetrated the pipe wall when found.

We have continual cracking in one of our process where we cycle from ambient to 255C every 20 minutes in 304L SS vessels. This cracking is normally a single crack following some phantom stain lines. At times for unknown reasons this cracking will take the form a highly branched crack form, resembling SCC. These cracks are external and will penetrate the shell if not addressed.

Metals Handbook Vol 10 (Failure Analysis and Prevention) page 540 in the 8th edition has some information thermal fatigue.

Just another thought, have you seen any signs of carryover around the failures. In case we had failures in some SS piping coming from a heat recovery unit where EDTA/NTA and NaOH were involved.