Weld repair of caustic vessel (NaoH)

[Alpha Charlie]

A caustic vessel 6mm thick (epoxy lined)remained in service for 40 years. 1.2mm uniform thickness loss was found in 2021 and two indications (cracks) were found in shell to shell weld. Vessel is originally not post heat treated.

I need suggestions for following maintenance activities;

1- Is it advisable to carry out welding inside vessel without heat treatment? (Vessel is originally not stress relieved)
2- Is it recommended to coat vessel again (Epoxy coating)?

 

[TugboatEng]

IF the vessel is still sound and the cracks can be end drilled then why weld at all? The epoxy coating can be re-applied. Consider some fiber reinforcement in the areas with cracks.

 

The first order of business is to determine the mode of cracking.
The leading candidates, based on the limited information provided, are stress corrosion cracking and fatigue (or more properly, corrosion fatigue).
In situ metallography is in order.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[bimr]

Your post has so little information that it is impossible to comment on it.

What is the caustic material?

How large is the tank?

What is the temperature and pressure?

What is the tank material?

What is the epoxy lining?

Generally, if the tank is 40 years old, you should consider a new one. The types of lining materials have improved since them.

 

[EdStainless]

Can you repair, yes. Should you repair, maybe.
You need to plan on stripping the entire lining (unless it is only a few years old).
Do a very detailed inspection, there may be a lot more issues hiding.
Do all of the repairs and hire pros to recoat it.
Now think about it, you could just buy a new normalized & PWHT vessel with a modern coating, or put a lot of work into this one and still have a repaired 40 year old unit.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

40 year old kind of decides it ... how many kinds of industrial equipment are expected to live longer than that?

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[weldstan]

The vessel has lost nearly a 1/16" corrosion allowance but you don't state age of coating nor do you specify whether cracks were on long or circ seams. Weld cracking would be more likely since the coating would most likely be non-uniform on them. But who knows how long ago the vessel was last thoroughly inspected. As Ed stated, you "still have a repaired 40 year old unit." In this case with probably over 1/2 its CA gone.

 

[Alpha Charlie]

First of all, I would like to thank you all for your responses.
Vessel diameter is 2200mm and cylindrical height is 2600mm. Vessel thickness of 6mm is already mentioned in my previous post. Corrosion allowance is 2mm so we can get along with 1.2mm thickness loss. Vessel design temp is 60 C while it is operating at atm pressure. This is the first time vessel being inspected.

We are doing MPI of all welds in the vessel. Until now 4 indications (cracks) of different lengths has been found all in welds. All these cracks are not visible in DPT. Vessel is not stress relieved and we found fallen coatings from multiple locations. I think built in weld stresses, non-uniform coating and service severity has lead to these cracks.

UFD and PAUT does not seem to fit here due to reduced thickness of vessel. We are not proceeding with any weld repair. I will appreciate if someone can suggest how to check cracks depth?

 

[EdStainless]

Have you done any structural analysis? What is critical crack length in this vessel?
You intend to leave the cracks?
I hope that your replacement is on order.
This requires some detailed risk analysis.
You can use UT to measure cracks. I am sure that you find that they are longer than you thought.
But like all NDT there are limits of sensitivity.
If UT tells you that a crack is a specific size, that is given the test conditions and setup.
That isn't an absolute value.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Alpha Charlie]

EdtSainless,

No structural analysis is performed yet. All cracks are present in one weld and max length of crack is 8mm as found in MPI. We would mainly be focusing on monitoring cracks growth in periodic inspection. We are trying to check crack depth by UT but it has limitations due to vessel reduced thickness.

What risk analysis you suggest and what crack size will be considered critical crack for such a vessel?

 

[Dhurjati Sen]

I have joined late and hope the vessel is still in Inspection stage.

Carry out RT of the portions where crack like indications have been observed to be absolutely sure.

It seems you are going ahead with repair and further use of the vessel, so you need to be very sure regarding the defects, if any, at all except the coating damage.

Regards.

DHURJATI SEN

 

Inspect away, but detecting fine cracks is not a particular advantage of RT.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[Alpha Charlie]

Yes agreed. These cracks are not detectable with RT.

 

[EdStainless]

You don't plan on grinding these out before recoating?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[weldstan]

Recommend using a rotary file or high speed fiberfrax grinding disc to remove cracks and recoat.

 

[Alpha Charlie]

Weldstan and Edstainless;

At one location, welding capping was grinded up to 1mm but crack appeared as such. Since we did not find cracks in other welds of the vessel so most probably these cracks are due to built in stresses at fabrication times.
I am more interested in finding cracks depth now but not successful so far.

 

[blacksmith37]

Did you consider caustic stress corrosion of low strength weld metal. I found it once where the wrong filler had been used.

 

[Alpha Charlie]

Yes considered it but how to identify low strength electrode now?

I have a question and if someone has any experience regarding it, please share your thoughts;

1- What if we weld repair the cracks and perform PWHT of all welds (Even though vessel is originally non-PWHT)? API 574 recommends stress relieving of the CS in case of weld repair.