Weld Cracking in Rich Amine service

[Mohammad Saad]

Hi, hope you all are doing well.

In the refinery that I work in, we have an Amine Regenerator unit that strips out H2S from Rich Amine (DEA). The top 3 shell courses are of SA-516 with SS304 cladding. We are having a leakage at the 4th shell course whose thickness was reduced from 16mm to a range of 7~10mm, and small patch where the thickness was around 3mm about 7 months ago. Belzona 1251 and FRP was applied but there is a leakage now. The exact position cannot be determined for now because of the belzona and FRP. The management has decided to reduce the pressure to zero and weld a patch plate over the entire effected area.

My question is, if this happens, how much chances are there for amine stress cracking or H2S cracking considering we have to keep the unit running for 2 months only (we have a complete Turn Around then, the entire shell course is planned to be changed in the TA). We are short on time so we cannot flush the vessel before welding. We are trying to study if the cracking would be a critical problem or not for the next 2 months.

 

[bones206]

You will probably need to hire a consultant with experience in this industry.

 

[JStephen]

Agreed that this is probably a more involved question than you'll find good answers for online.
Sounds like a Fitness For Service/ Corrosion Engineering/ Process Safety issue all combined.
There is a "Boiler and Presure Vessel Engineering" forum here, you might check there also.

https://www.eng-tips.com/threadminder.cfm?pid=794

 

[JD P.E.]

I work in this area of pressure vessel mechanical integrity and do fitness for service analysis. Environmental cracking (such as sulfide stress corrosion cracking) cannot accurately be predicted and SHOULD NOT UNDER ANY CIRCUMSTANCES try to be predicted. It’s simply unpredictable. You cannot establish crack stability because you cannot predict future behavior (unlike fatigue).

A couple of other points. I routinely work in environments where equipment is pushed to the limits - however, I have not seen belzona / FRP combination used to try to prevent leaks in this service. Maybe this is common , but I would think the leak tolerance for this service is 0. Maybe I’m wrong.

Realistically if management has reduced operating pressure to 0, the stresses should be low minus residual stresses. So effectively you have an atmospheric storage tank - if you can stand a pinhole leak/crack over the next 2 months then run it until then. If not, you have limited options at this point.

Edit - I see you mean that you will reduce pressure to weld the patch plate on and then pressure up to keep running. In that case, yes I believe you are at risk of sulfide SCC of carbon steel welds even if it’s just 2 months. Like I said before, you can’t really quantify a prediction but carbon steel will crack in this service. You really need to gain entry to this to make a better repair to keep running. I am assuming this is not in USA?

Also, if the thinned area you mentioned is below 50% T-min you are automatically in a level 3 fitness for service analysis or even disqualified and required to repair if you are under ASME FFS / API 579.