Corrosion Allowance due to pitting in pressure vessels

[RonJeremy]

There are a number of threads in this forum ( thread338-159732 , thread338-171512 , and thread338-130588 ) where the same contributor consistently states 'with authority' that a corrosion allowance is not required if you don't expect uniform corrosion.

Now then, as a vessel engineer in oil and gas and petrochem, I would like to challenge that assertion, and throw it open for discussion. I believe that it is exactly localized corrosion such as pitting that the likes of ASME intend CA to address. And I believe that it is dangerous to pass it off as a FACT in several threads that you do not need a CA unless you expect uniform corrosion. I might be wrong: please let me know your views.

1. I hardly ever see vessels (with 1/8 " CA specified) that have been in service a few years that have had 1/8 " corroded material flake off uniformly over the inside of the shell. They usually have pitting or crevice corrosion, or local corrosion near to an inlet nozzle, or at a hot spot, for example.

2. Epoxy or rubber lined CS vessels always have a large CA applied (typ. 1/4 ") to mitigate for local corrosion due to pinholes in the lining. You will NEVER have uniform corrosion in a lined CS vessel, but you must always specify a CA, for the reason stated.

3. We normally specify internal attachments, wetted on both sides, to only have 1 thickness of corrosion allowance applied (i.e., not full CA on both sides), because it is not expected that you will have a worst case of corrosion attacking both sides of the internal part simultaneously.

4. It used to be the case that corrosion products were not deducted from the weight of a vessel. Recently, I have seen one software (PVElite or Compress?) quoting 'Corroded Weight' almost as if they think that the vessel is dissolving like an ice cube! Apparently, a 1/2 " thick vessel with a 1/8 " CA, will weigh 75% percent less at end of life than it did when new! Nonsense (normally).

I'm sure that there are people out there there who will say that "in my plant we get uniform corrosion...". OK, I'm not saying you don't: I'm saying that it is wrong to think of CA only applying to uniform corrosion, and that most of the time the corrosion will be local (may be pitting, or crevice, or may be in the weld metal or HAZ).

Discuss...

 

[SJones]

What 'the contributor' consistently, and correctly, states is that corrosion allowance is not a means of corrosion control for localised corrosion. I know that what I am about to quote is from a pipeline design specification, but the concept is the same:

From DNV OS-F101

A corrosion allowance is primarily used to compensate for forms of corrosion attack affecting the pipeline's pressure containment resistance, i.e. uniform attack and, to a lesser extent, corrosion damage as grooves or patches. Still, a corrosion allowance may also enhance the operational reliability and increase the useful life if corrosion occurs as isolated pits; although such damage is unlikely to affect the pipeline's resistance, it will cause a pinhole leak when the full wall thickness is penetrated. However, the extra wall thickness will then only delay leakage in proportion to the increase in wall thickness

In general, for carbon steel, the conditions that may give rise to pitting will also cause general corrosion and so a corrosion allowance is applied anyway. It just means that adding any extra metal to compensate for pitting only buys a bit more time. This is the tenet of using corrosion allowance in a lined vessel - buying sufficient time to find the pitting. Designers really then have to think hard about the economics of what they are doing: the cost of extra steel AND the lining AND the frequent inspection to find pitting versus the more expensive CRA vessel that might be virtually 'fit and forget'.

1.
This is probably more akin to 'patches and grooves' rather than pitting.

2.
See above.

3.
If a corrosion process occurs inside the vessel, and the corrodent contacts both sides of an item, why would only one side corrode? Nevertheless, internals are generally not pressure retaining so your policy may just result in more frequent change out of internals rather than loss of containment.

4.
Agreed, this is appears to be a bit fanciful but our pressure vessel colleagues may be able to elaborate.


Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[MikeMet]

Good discussion, to add;

We don't add corrosion allowance (CA) if we think pitting is an issue, we change the material of construction to a material that suffers general corrosion. Pitting corrosion rates are very unpredictable, which makes CA addition impractical. If pitting occurs, then evaluate by API 579, Fitness For Service, and fix so pitting does not reoccur.

1. You need to get out more. I have seen numerous vessels of carbon, stainless and nickel alloys that have suffered general corrosion. Typically we add CA to carbon and stainless, not nickel alloys. But the vessels did us up the CA and then we start working with FFS by API 579, usually re-rating buys us time until weld overlay or replacement.

2. The corrosion of carbon steel behind a liner is general corrosion that takes place in a localized area. Maybe semantics. But again a FFS evaluation can help. But CA is always added to lined vessels because there always seems to be a leak through the liner at some time.

3. Agree with Steve above

4. A corroded vessel will weigh less than it did when originally installed. Metal is gone, has to weigh less. I don't why anyone would need to know this during the initial calculations, but it would be good to evaluate if you are replacing a vessel and need it for lifting weight.

 

[EdStainless]

A lot of this depends on the materials of construction, expected corrosion mechanism, and risk of failure.
If you are using a SS, and there is a risk of pitting, and a leak would be bad, then adding thickness does you no good at all.
If there is reason to expect localized corrosion then you should address it with local design modifications.

I have seen scrapped heat exchangers that weighed half of the original weight. The only people that really care are the salvage guys.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube