Weld Root Corrosion in Production Pipework

[RoyB]

We are suffering from very severve weld root corossion in LTCS A333 gr6 production pipework. The latest to fail has only been in service for 18-24 months. The concensus of opinion is that the 1% nickel in the welding consumable is the prime cause.
The pipework should no signs of corossion, but the welds coroded to the point of leakage.
Any ideas appreciated, we are currently replacing the pipework with like for like so will probably be re-visiting the problem in 18 months time.

 

[INTEGRITY]

RoyB
In the past, we have had problems with preferential root weld corrosion, in API 5L crude oil production pipework. After around 1 year of investigation in 80 wells, we found that the most possible cause was a combination between the fluid erosional velocity and the CO2 presence. The problem was minimized adjusting some variables, such as: CO2 corrosion inhibitor selection with high "shear stress", increasing corrosion inhibitor dosage, increase the chemical injection system availability. And in some cases, we replaced the pipework for a largest diamater one. Additionally, END procedures (Gamma-Ray and TOFD) were developed to allows us a close monitoreo, that us of data to make for fitness for service analysis.

Regards,

 

[garciaos]

In addition to comments from "Integrity", you can use a ceramic composite as "Belzona" or "Thortex" to protect the welds root from PRWC. We used that, in some wells and it worked very good

 

[aybee]

sound like you are using a consumable which is not suited to the job, dont know why you want one with nickle anyway A333 is an impact tested version of what is essentaliy A106B or very close I think( dont have the specs to check)so I suggest you investigate another consumable or use a normal low hydrogen E7018 type consumable and PWHT the weld to get the lower impact value allowance.
it sounds like knife edge corrosion at the HAZ due to diferential cathodic attack started by the slight change in metallurgy Weld - base metal