Reversing Austenitic stainless sensitisation 2

[robsalv]

I've had a good look around the forums, but my question is not really been answered.

Is austenitic stainless sensitisation fully reversible?

My experience, limited but growing, doubts that it is.

Thoughts? Experiences?

TIA.

Rob

 

[unclesyd]

It can be reversed with a full or quench anneal. The temperatures are above 1850°F and a rapid quench. The quench is normally water in a tank or as a heavy mist or spray. The temperature varies with each type stainless and it's thermal history.

 

[robsalv]

What sort of soak time are we talking about UncleSyd?

 

[EdStainless]

Depends on the degree of sentization.
More accurately it depends on the size of the carbides. If they are small and well dispersed (almost never) then I would think about 15 min at 1850F or 5 min at 1950F. If the carbides are large and heavy then you may be looking at 20 min at 1900F to get good re-solutioning of the carbides.
These times/temps are only aprox, and the apply after the entire part is at the soak temp.

How heavy are the cross sections? What C level? How sensitized? How did they get sensitized?

There are some risks here, surface oxidation and grain growth are the first two that come to mind. Make sure that the cure isn't worse than the illness.

= = = = = = = = = = = = = = = = = = = =
Corrosion, every where, all the time.
Manage it or it will manage you.

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

 

[robsalv]

Edstainless - we do have a situation with a new vessel and it's 32mm thick SE heads.

The knuckle region showed many visual cracklike indications. Blending removed most of the macro cracks, but metallographic replicas found intergranular cracking and extensive sensitisation at the knuckle. A spot near the centre of the head showed only some sensitisation. Can stainless be sensitised only at the external skin level. We're running some ID replicas to find out.

We're considering our options. The head may require replacement purely from the IG cracking point of view, but assuming it's a skin type effect, perhaps the head is recoverable if we grind enough and also turn the sensitasation around.

The vessel is not in a corrosive service, but it's in cryogenic service, so even if we could clear the IG cracked layer, unless we can turn the sensitisation around, I'm not willing to sign off on the vessel going into service. Sensitisation compromises the low temp properties.

The other complicating thing is that all the document trails indicate that we should have a clean head. The original plate was clean according to the cert. The pressing and subsequent annealing were to the code. But I still have a crap head... so much for paperwork.

Cheers

Rob

 

[robsalv]

FYI, I've just had confirmed metallographically that the head is completely cactus.

The inner surface of the knuckle region exhibited both sensitisation and inter granular cracking. That means the knuckle region is cactus the whole thickness through.

How on earth could the hot forming and annealing have been stuffed up that badly?? So much for all the paperwork. Now it's up to commercial to determine who pays for the repairs.

Rob

 

[stanweld]

Poor temperature control during forming and heat treating can be a major problem. It has been my experience that many furnaces have not been surveyed and properly calibrated for years. It would also appear that contamination of the surfaces may have occured between forming and heat treating.

 

[EdStainless]

Hot work may have finished at too low of a temp.
Anneal heat up rate may have been too slow/not enough soak time/not rapid enough cooling rate.

Sensitized stainless is very sensitive to CSCC. If this is in cryo service then you may be getting some condenstion on the surface. It wouldn't take much to crack heavily sensitized material.

= = = = = = = = = = = = = = = = = = = =
Corrosion, every where, all the time.
Manage it or it will manage you.

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