Strange crack in tank

[ricellis]

Can someone please help me out with this crack? I'd like to know a possible cause and suggested repairs. The crack is longitudinal and is beside a vertical weld in a 30m high 30 yr old precipitation tank. The crack is 3mm deep, and extends from 8mm to 78mm from the weld toe.
The crack is in the tank cylinder, about 1/3 up from the bottom

Thanks,

Richard

 

[unclesyd]

Comeback with more information about your problem.

What is the material of construction of the tank and the environment it sees, process, temperature, etc.

Was the tank insulated.

 

[ricellis]

The tank was designed to API 650 and contains hydrate slurry in sodium aluminate solution. It operates at 90 degrees C. The plate is unlined G250 steel, 30 yrs old, 25mm thick. There is minimal temperature fluctuations and minimal fluid flow. Fatigue is not expected. This is the first crack we've seen entirely in the parent metal, away from the HAZ. We've seen a lot of HAZ cracks due to stress corosion cracking. The only thing I can think of is a crack that started due to a delamination of the old steel in this area?

Richard

 

[metengr]

Are you sure you don’t have cracks in other locations of the tank? Did you perform supplemental nondestructive testing to confirm only a single crack? If you want to determine the specific cause of cracking in the tank wall, you could remove a section of the tank wall containing the crack and install a flush patch. This is a window repair where the cracked section of the tank wall is removed and a replacement piece of equivalent material is full penetration welded to the existing tank wall. The corners of the window opening should contain as large a radius as possible along with local nondestructive testing before and after welding the flush patch plate. This type of repair would allow for a metallurgical analysis of the removed material to confirm the cause of cracking.

Your second repair option is to extract a small boat sample of the tank wall (containing the crack) and remove the remaining portion of the longitudinal crack by local excavation (grinding). After the boat sample and the remainder of the crack are removed, locally weld restore the excavation and the area where the boat sample was removed, and perform nondestructive testing before and after welding is completed.

 

[kenvlach]

What is the pH?
I don't know your use, but I associate sodium aluminate with the precipitate from rinsing aluminum after caustic etching. If some NaOH is present, could be a caustic embrittlement issue.

The NACE caustic soda service chart recommends stress relief of carbon steel welds for use with 50% NaOH above 120[sup]o[/sup]F (49[sup]o[/sup]C). As [NaOH] decreases toward 0%, the safe zone (for non-stress relieved CS welds) rises to 180[sup]o[/sup]F (82[sup]o[/sup]C).
Chart is on page 18 in a PowerPoint presentation of D.H. Lister & W.G. Cook, Univ. of New Brunswick:

http://www.unene.ca/un1001/UN1001-SECTION-9-2006-March.ppt

 

[MOB1]

"Delamination of old steel" is a bit of a furphy.
More likely to be caustic stress corrosion cracking as identified in the HAZ areas previously.
It is sometimes worth investigating for other welds in the zone of the crack if it appears in a "strange" area. Previous weld repairs may have been carried out on the tank.

 

[MJCronin]

rice..

Consider posting your question on

http://www.ast-forum.com

I think it's a great forum......

-MJC

 

[klogerbmt]

You may consider having further metallurgical investigative work done. I would recommend having some in-situ metallorgraphy conducted on the cracked locations and have them evaluated by a metallurgist. The method is non-destructive and requires the field polishing and etching of small sections. This would require access to the area but could eliminate the removal of boat samples.