Is use of a corrosion allowance on long distance pipelines still valid?

[LittleInch]

Before anyone replies "Of course it is", let me explain further.

The use of a Corrosion Allowance in pipelines carrying mildy corrosive fluids - typically I'm talking CO2 corrosion here - has been around since the dawn of pipeline codes when the result of any issue was to throw metal at it. This works and in reality is not a huge cost when looking ar relatively short pipelines, say <100km. However, for big inch long distance lines each mm represents millions of dollars. The corrosion rate is nowadays calculated using various algorithms backed by experimental data with various factors and arrives at a (far too exact) figure for uninhibited and inhibited corrosion rates over the life of the pipeline, often in the 2-4 mm range for say a 95% CI availability.

My issue has always been that corrosion like this doesn't occur as a whole scale removal of that amount of material, but in relatively small locations with pitting type corrosion. With the advent both of intelligent pigs to both detect and define any areas of metal loss, any corrosion found does not necesarily reduce the operating pressure of the pipeline when you use the calcualtions now available in B31.G and RSTRENG etc. These were not available when pipeline design codes were being written.

Thus where you make an insistence in the operating manual that CI is injected continuosly and regular inspection occurs, do you need to include a CA for the whole pipeline and instead concentrate it in specific areas or zones where it is either more probable (high temp, high pressure, low spots or locations where it is difficult to return to fix it later (road crossings, HDD etc, where a 0.6 DF might exist anyway. Why spend tens of millions of dollars now to protect small areas of pipeline in the future?

Clearly there is a risk balance to be made, but with the tools and calculations now available to deal with pitting corrosion as to whether it is a problem or not or provide additional external sleeving or other remedial measures, is the use of CA still valid in its full calculated state or can it be reduced to 50%, 25% or 0% of the inhibited rate. Your views most welcome.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[SJones]

Here's my answer to a similar question on a certain Linkedin forum:

In summary, given the process of corrosion allowance determination with corrosion models whose reliability is unknown, production data that could be a best guess, over optimistic assumptions regarding corrosion control efficacy, and so on, it really is a 'how lucky do you feel' game.

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.

 

[BigInch]

Was it ever? I know of a large number of pipelines where no corrosion allowance was used at all.
In any case, it could only be valid up to the point where other corrosion mitigation strategies, or repair/replace, become more economical, which probably isn't very far off.

Independent events are seldomly independent.

 

[racookpe1978]

In today's anti-spill anti-business anti-everything EPA regulations from the fed's down to every department of parks, enviro and health? Not having a corrosion allowance could not be defended after the oil/water/chemical/soap/sewage/hazmat stuff gets out into the woods/water/wetland/backyard/lumberyard/front yard of a politically-connected enthusiastic lawyer.

Having too small a corrosion allowance might be defensible if you have some standard. But none at all?

 

[BigInch]

If it were so important, it would be required by B31.4, B31.8, or CFRs.
Launchers and receivers ARE required by CFRs on regulated pipelines. Did the CFRs require a corrosion allowance? No. They wanted pig launchers on all pipelines. That is so corrosion can be adequately monitored, a pipeline segment's remaining life estimated and the segment can be replaced well before it ever leaks at all. If intelligent pig monitoring is part of your maintenance and failure prediction program, what use is a corrosion allowance, except to waste a great tonnage of steel.

Independent events are seldomly independent.

 

[LittleInch]

Thanks for the responses.

S Jones. I appreciate that there is a risk management issue here, my point is that the use of a blanket corrosion allowance is now a very blunt tool. I'm talking about export fluids here not well fluids so fluid composition shouldn't change quite as much as well fluids. The saving for a long pipeline could be tens of millions of dollars per mm, which should be able to pay for quite a bit of enhanced CI injection equipment. Targeted use of an allowance for sections either more vulnerable or difficult to repair should be used along with effective inhibition?

BI. Were those pipelines with no allowance ones where co2 corrosion had been calculated or just ones with no specific corrosion other than external?

Racookpe. I'm not saying none at all, I'm saying that the allowance should be targeted and not just applied in a blanketed fashion, hence swamps, water crossings, wetlands, yes, but open countryside, easy to repair pipeline sections no. Given the cost of steel, the use of inspection tools and use of continuous inhibitor injections with certain sections with some extra spare must make sense in this day and age.

It would be nice to know if anyone else has tried this...



My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

Some (gas transmission) pipelines had up to 10% CO2 and were still relatively close to the gathering systems, with relatively high water drop outs. Others, farther into the transmission systems where they ran basically dry gas, with some condensate and occasional water dropouts, esp during cold winters. Others offshore, so basically within a long gathering system. Offshore we used no corrosion allowance and simply monitored the corrosion rates with coupon probes and adjusted the corrosion inhibitor injection rates accordingly. That was on more than 20 pipelines, some short, some a few hundred miles long, from 6 to 24" diameters. I installed them more than 20 years ago and I've never seen an incident report fingering any one of them.

In an unregulated onshore gathering system consisting of around 2000 miles of CS X52 pipe connecting 850 wells with high water content, high microbe action, up to 5% H2S from some wells, around 5% CO2 from others, relatively warm temperatures, the "strategy", before a corrosion inhibitor program was begun, was basically explode and replace. We blew up a lot of cactus. In much of the system there was no external CP either. After we started installing external CP and an inhibitor injection system, the number of blowouts decreased rapidly. The damaged pipelines (there were plenty) never showed evidence of all-around corrosion, only bottom of pipe, esp at lower spots in the system, localized pitting near bottom of pipes, corrosion along seams and at weld joints ... of questionable quality. External corrosion wasn't concentrated towards any specific area of the pipes and was likely to appear almost anywhere near damaged or peeled off or up coatings (the older coal tars and tape wraps).

The S**P pipelines didn't use a corrosion allowance either. Although they were for supposed clean refined products, a surprisingly large amount of water, sand and other contaminates entered the pipelines from tanks at marine terminals and refineries.

BTC Pipeline
As I recall, the BTC pipeline did not rely on a general corrosion allowance, preferring inspection and timely replacement to ward off predicted failures,

http://www.bp.com/liveassets/bp_int...denda_Content_BTC_ESIA_ADD_Ap-Preventions.pdf

Don't let me influence you. Here's some comments by NACE engineers.
NACE BBoard somewhat conflicting comments, but do mention a movement direction away from applying a general corrosion allowance, although some of their more conservative engineers do not seem to like it.. for whatever reason.

http://events.nace.org/library/corrosion/RiskAnal/dicuss-allow.asp

and...

http://www.eng-tips.com/viewthread.cfm?qid=159732

Independent events are seldomly independent.

 

[LittleInch]

BI - thanks for the info - I think you know where I'm aiming on this one.

I understand and respect the inherent conservatism of the hydrocarbon pipeline industry, but after 20 years or more of having smart pig technology and understnading more about how to categorise corrosion, calcualte remaming strength and new CI techniques, is this now time to roll back some of the unecessary conservatism and target use of metal more appriately?



My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[MikeHalloran]

John V. Lindsay 'saved' New York City from yet another of its recurring budget crises by firing all the bridge painters. Twenty years later, the West Side Highway fell down. Duh. He never got blamed for it. To add injury, the assertion that it is cheaper to replace a bridge than to maintain it has been echoed by many similarly empty heads.

When, not if, the MBAs gain complete control of the pipeline industry, they will fire or at least decimate the corrosion control staffs because "constant inspection never found anything of immediate concern", and pipelines will transition to "condition based maintenance" with no surveillance. From that perspective, surveillance is unnecessary because the pipeline itself will tell you when it needs maintenance, e.g. by exploding or relieving itself of its contents.

The MBAs will not be blamed for any pipeline failures.
They will point to LittleInch, for even breathing the idea of reducing or omitting corrosiion allowance, whether that happens or not.





Mike Halloran
Pembroke Pines, FL, USA

 

[LittleInch]

Mike, you don't like MBA's do you?....

I take your point, but there is another aspect which has happened to me when because there was a corrosion allowance, some idiot decided to cease CI for x years to "use up" the allowance. Of course the pipe didn't play ball and leaked crude after x-2 years eventually needing full line replacement.

Making it clear that you need these things in operation leaves little wiggle room and I'm not trying to eliminate it altogether but target extra metal properly.

But I appreciate the comment and there is something in it.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

A general corrosion allowance equivalent in the airline industry would be to make each and every passenger wear a parachute.

Independent events are seldomly independent.

 

[MikeHalloran]

I like MBAs ok, sliced thin and served with a nice chianti.

...

Okay, seriously, I associate their arrival with change, and not in a good way.

I think this is part of the curriculum:
Very soon after they start, they pick a fight with one of the senior staff, usually a covert leader, e.g. me.
It's often over some technical issue that they don't quite understand, and are incapable of understanding.
They always win, and the chosen victim gets to find another job.

I don't know what they do next.




Mike Halloran
Pembroke Pines, FL, USA

 

[waterpipe]

Corrosion in oil & gas pipelines is definitely different and more complex than water pipelines but this CA has been a common issue in the water sector. Though corrosion and its consequences are far less serious in water transmission line, quite often pipeline specifications ask for extra 3 mm CA in wall thickness calculations. I’ve seen a pipeline protected by epoxy interior lining and 3-layer HDPE exterior coating with a well-designed CP system, which HOWEVER had extra 3 mm wall thickness as CA along its more than 200 km length! Add to this the fact that the maximum allowable design stress (hoop stress) was limited to only 50% of SMYS. I’ve always wondered is it justified to have this “blanket” CA (nice term by littleInch) while any corrosion defect would be in form of local pits, checks, etc.?

The point is that even for pipelines much less susceptible to corrosion issues, there is a tendency not to let go the old habits.

AWWA M11 1989 version says:
“At one time it was a general practice to add a fixed, rule-of-thumb thickness to the pipe wall as a corrosion allowance. This proved to be an irrational solution in the waterworks field, where standards for coating and lining materials and procedures exist. It is preferable to design for the required wall-thickness pipe as determined by the loads imposed, then select linings, coatings, and cathodic protection as necessary to provide the required level of corrosion protection.”

However, they’ve changed the tune in 2004:
“At one time, it was a general practice to add a fixed, rule-of-thumb thickness to the pipe wall as a corrosion allowance. This was not an applicable solution in the waterworks field, where standards for coating and lining materials and procedures exist. The design should be made for the required wall-thickness pipe as determined by the loads imposed, then linings, coatings, and cathodic protection selected to provide the necessary corrosion protection.

To me, it seems that this 3 mm CA myth is something brought to the water pipelines from oil & gas industry and engineers (where there had been no lining) and has remained within water pipe specifications without being questioned by anyone. The question is when someone dares to say "The king has no clothes".

 

[BigInch]

Actually I think water pipelines are likely to have much more problems from fluid corrosion than oil or gas, provided the petroleum products do not contain H2S, or CO2 and, because of that, not sure, but I think the general corrosion allowance practices came from the water pipeline industry.

I hate Windows 8!!!!

 

[SJones]

It's a value thing. EN 1325-1 defines value as

"The relationship between the contribution of the function to the satisfaction of the need and the cost of the function." It then adds some notes:

"Note 1 The term of (sic) value is also used when factors other than cost such as reliability, weight, availability of resources and time are considered;

Note 3 This definition concerns the value for a specific user (value can be different for the (sic) different users.........."

Need is defined as:

"What is necessary for, or desired by, the user"

Function is defined as:

"Effect of a product or of one of its constituents"

It's down to users to differentiate between 'necessary' and 'desired, and to work out exactly what is the function of a corrosion allowance'

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.