ranges of corrosion rates 1

[Chumpes]

dear all

what would be the maximum tolerable corrosion rate to be controlled by application of a corrosion allowance ? (for pressure purpose, general question not liked to a particular item)

0.1 mm / y ?
0.25 mm / y ?
other ?
any ? (reference ?)

Higher corrosion rates may induce non-uniform attack for which application of a corrosion allowance could be not sufficient.

 

[EdStainless]

It partially depends on what happens and what it costs when it fails.
And partially on how good your CR estimates are.
0.1mm/yr is about 0.005"/yr, this is often considered minimal.
Much equipment is built using 4-5x this amount and has good 20 yr service life.
Depends on your goals.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Chumpes]

thanks Ed
yes, it supposes that it is possible to provide reasonnable estimation of corrosion rates, which can be somewhat tricky ! especially when corrosion rates are high above 1mm/y

So, can I understand that you would allow up to 0.4-0.5 mm / y maximum if the application allows it ?

I am asking because i keep on being challenged by people that believe it is acceptable to have very high corrosion rates of carbon steel (lets say 1 mm / y) with hudge corrosion allowance (6mm) and short design life (5-6 years).

In my opinion, generalized corrosion should preferably be controlled at the first order by selection of a material with reasonnable resistance (CR < 0.1-to-0.25 mm/y 'ich), the extra corrosion allowance providing an extra safety margin and extra service life.

 

[weldstan]

The 6mm CA can and has been used with a short term life expectancy. It depemds on the overall costs of the equipment, its replacement costs or refurbishment costs versus a more expensive alloy. Also the expected life of the process system must be considered.

 

[SJones]

"How long is a piece of string?" is an equivalent question.

Steve Jones
Corrosion Management Consultant

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

All answers are personal opinions only and are in no way connected with any employer.

 

[EdStainless]

Problems with a high corrosion rate are:
1. the higher the rate the more likely that there is variability (local temp, pH, velocity) so if 1mm/yr is average are there places that are 3mm/yr?
2. If heat transfer is important (heat exchangers) then the CA material will mess with heat transfer balances.
3. If the plant life is only a few years then fine, but if you are talking a plant where some equipment has 20+ year life I find it very hard to believe that replacing this unit every 5yr is really cost effective.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Chumpes]

yes Ed, I totally agree, severe generalized corrosion can no more be considered as homogeneous.
Thanks for most of your answers, especially Ed's.

So, in absence of valuable reference i will keep my own philosophy for corrosion allowance vs. corrosion rates and design life

@SJones : I am sorry if you were told that all strings were having the same length, people probably lied to you in order not to hurt you ! (joke=) but your answer is less pointless than it looks because it allowed me having a good laugh !

 

[SJones]

There are no strings on me. Glad you saw the humour

The starting point is the comment made by Ed: how much do you trust your corrosion rate prediction? Your answer should be: without any reliable field verification measurements, not much at all. After that, there are so many other project specific variables to juggle - life cycle cost, risk acceptance/aversion are a couple of examples, that it is impossible to generalise down to a rule of thumb.

Steve Jones
Corrosion Management Consultant

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

All answers are personal opinions only and are in no way connected with any employer.

 

[Dhurjati Sen]

Dear Chumpes,

Now-a-days, in RBI software, corrosion rates are assigned based on historical data. This is okay only to that extent where the corrosion is uniform and general, however it is absolutely of no use where the corrosion is non-uniform, like pitting, crevice, erosion etc.

Still, people are trying to assign pitting corrosion rates, which is quite funny.

Even they try to assign an external corrosion rate (say in an insulated piping) which is even funnier.

Regards.

DHURJATI SEN

 

[brimmer]

Typically we don't design corrosion allowance to account for non-uniform attack. In this scenario it is impossible to determine corrosion rates, or how long that rate might last. You would typically try to mitigate pitting attack by other methods (material, operation, chemical mitigation, etc.) as opposed to letting this occur and depending on corrosion allowance.

 

[dik]

From a recent webinar from SkylineSteel... on ABC...

Dik

 

[Chumpes]

yo
for information, i finally found some engineering guidelines on acceptable corrosion rates for reasonnable design life in the refining industry :

they range from 0.076 mm/y to 0.3 mm/y (3 mpy to 12 mpy), depending on the company and on the item that is considered (heat exchanger tubes, shells, internals...)

regards

 

Higher corrosion rates may induce non-uniform attack

It doesn't work that way. Without knowing the corrosion mechanism(s) and where they operate, you could be living dangerously.

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

 

[Chumpes]

It doesn't work that way. Without knowing the corrosion mechanism(s) and where they operate, you could be living dangerously.

Stating the obvious is sometimes unecessary but thanks anyway.
I was talking about corrosion mechanisms gathered under the designation : "generalized corrosion", which is a more or less uniform loss of metal thickness.
As you probably already know, there are plenty of mechanism(s) of "generalized corrosion" in aqueous or dry phase that have been more or less well described, and more corrosion mechanisms that still require more knowledge, but i limited the discussion to damages that induce a "more-or-less" uniform attack of the metal surface.

My point of view is that living dangerously would be to believe that corrosion rates can be precicely calculated, and that very high rates of "generalized corrosion" could remain strictly uniform : i leave it to you and keep my safeguards.

 

[SJones]

I finally found some engineering guidelines on acceptable corrosion rates

Would you care to be specific and name names of the said engineering guidelines?

Steve Jones
Corrosion Management Consultant

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

All answers are personal opinions only and are in no way connected with any employer.

 

[Chumpes]

UOP PROPER DESIGN AND OPERATION OF NHT CFE EQUIPMENT
(max 0.127mm/y for HE tubes)

SAUDI ARAMCO SAES-L-133 Corrosion protection requirements for piping
(0.076 mm/y is a pecular limit from which material upgrade can be foreseen)

ENOC EDS-A-9 Selection of metallic materials (0.25mm/y
[For most equipment corrosion rate of greater than 10mpy is considered high enough to change materials])

Alireza Bahadori, Corrosion and Materials Selection, A Guide for the Chemical and Petroleum Industries
[Where corrosion rate is more than 0.3mm/y or the total corrosion over the design llife exceeds 6mm, alternatives should be evaluated]

Litwin, Process design guidelines - materials of construction
[Nace allows a corrosion rate of 20mpy, Litwin prefers not to exceed 10mpy]

Total, GS RC CORR 006,
[low corrosive service is CR < 0.125mm/y and corrosive service is 0.125mm/y < CR < 0.3 mm/y]