Over
24,000 miles of pipeline have been laid on the Outer Continental Shelf (OCS) in
the Gulf of Mexico since 1948. Over the years, much of this pipeline has been
abandoned or removed, but as of June 1997, there were still some 17,000 miles
of active pipe. Pipe-laying activity has been up and down over the years,
somewhat mirroring the "boom and bust" cycles of the oil and gas
industry. Some 1,222 miles are over 30 years old, and 5,952 miles have
celebrated a 20th anniversary. Obviously these 5,000-plus miles of pipe would
be considered at higher risk from an integrity standpoint than the 11,000 miles
younger than 20. The mere fact that these old lines are still in operation
reflects well on the skills of the corrosion control community (Figure 1).
FIGURE
1.
Active Gulf of Mexico Pipelines: Mileage vs. Age
EXTERNAL
CORROSION CONTROL OF OFFSHORE PIPELINES
All offshore pipelines are protected
from seawater corrosion in the same way. The primary corrosion control system
is pipeline coating. This is supplemented with cathodic protection (CP) to
provide protection at coating defects or "holidays." In the Gulf of
Mexico, the pipeline coatings used until the early to mid-1970s were either
asphaltic/ aggregate, "Somastic"-type, coatings or hot-applied coal
tar enamels. Since then, the trend has been to use fusion-bonded epoxy powder
coatings. In the earlier days, the trend in cathodic protection (CP) was to
rely on impressed-current systems. In the 1960s and early 1970s, zinc bracelet
anodes attached to the pipe were widely used. Since then, more efficient
aluminum alloys have surpassed zinc as the preferred material for offshore
galvanic anodes. There are, however, still some operators using impressed
current systems and some using zinc anodes.
BRACELET
ANODES
Virtually all new pipelines
installed in the Gulf of Mexico are equipped with aluminum
bracelet anodes. There are two basic types, square shouldered and
tapered.
The square-shouldered anodes are
typically used on pipe that has a concrete weight coating. When installed, the
anodes are flush with, or slightly recessed inside, the outside diameter of the
concrete.
The tapered anodes are designed to
be installed on pipelines with only a thin film corrosion coating. The whole
idea is to protect the bracelet anodes during the pipe-laying process. The
anodes are particularly at risk from mechanical damage when the pipeline
travels over the stinger on the back of the lay barge.
Even with these tapered designs, non-weight-coated pipelines
still sustain anode damage, which can in turn cause coating damage. Several
methods are being used to combat this problem. The use of cast-on polyurethane
tapers is gaining popularity, and mounting both halves of the bracelet on top
of the pipe is a common technique when pipe is laid from a reel barge and the
anodes have to be attached offshore (Figures 2 and 3).
Six-inch pipe reeled on the
barge Chickasaw
Tapered bracelet anodes
installed on top of pipe
DESIGNING
CP SYSTEMS FOR OFFSHORE PIPELINES
When designing a cathodic protection
system for a pipeline, the corrosion engineer has to consider the following
variables, all of which will have an impact on the final anode alloy and size
selection:
• Design life required - (minimum is
20 years)
• Pipe diameter length and to-from information
• Geographic location
• Type of coating
• Pipe-lay / installation method
• Water depth
• Burial method
• Product temperature
• Electrical isolation from platforms or other pipelines
The smart cathodic protection
designer will look early on at the intended pipe installation method, as this
will have a direct impact on the amount of coating damage one may expect (there
is also a risk of having anodes detached during the lay process). In all
pipeline design guidelines, the conservative approach is advised. For example,
the majority of early Gulf of Mexico (buried) pipelines were designed on the
basis of 2 mA / ft. of bare steel and 5% coating failure. In essence, this
means taking 5% of the total pipeline surface area, and applying 2 rnA / ft. of
cathodic protection current to it. This may sound reasonable, until one looks
at what 5% bare means:
On a 40 ft. joint of 12 in. pipe, 5%
bare coating would have 2 square feet of bare steel, or to express it another
way. 4 linear feet of pipe would have the coating gone from 180° of the
circumference. This is an extremely conservative figure. As a result, the early
pipeline system designs would appear to be very conservative.
PIPELINE
INTEGRITY
When considering the role of cathodic protection (CP) in
pipeline integrity we should investigate what causes offshore pipelines to fail
and leak. If all the failures of pipelines in the Gulf of Mexico were counted
and tabulated, the findings would probably show the general trend expressed in
Figures 4 and 5. (These graphs are based on studying a limited sample of
failure reports from two oil companies.)
Causes of offshore pipeline
failure
Causes of offshore riser
failure
Since external corrosion is only
responsible for a very few of the documented pipeline failures, we could
truthfully say that, in general, the combination of CP and coatings is doing a
good job.
However, we must not be led into a
false sense of security. The only reason the external leaks have not started in
earnest is that the old systems were unknowingly over-designed. Thus, a 25-year
design life has effectively turned into 30, 35 or even 40 years.
There is a practical limit on how
long sacrificial anodes will last, and it is based on the auto-corrosion rate
of the anode material. If we were to assume that pipeline systems are all good
for at least 30 years, then there should be several thousand miles of pipeline
with depleted CP systems (Figure 1). The question, then, is why are we not
seeing more external failures?
In truth, the answer to that
question is that we probably are seeing a higher external corrosion leak rate
than we have at any time in the past. But when will it peak? The pitting rate
of steel in seawater on a well-coated pipeline in the absence of cathodic
protection anodes could vary between 0.01-0.05 inches per year. Thus, it could
take anywhere from 5 to 25 years to pit through an inch of steel. This amount of
loss could be sufficient to cause a pipeline failure. Higher corrosion rates
can be generally expected when the pipe coating has a combination of large
damaged areas and adjacent pinhole defects, and when the pipe is exposed to
seawater rather than mud. There is also a particular risk of microbiologically
influenced corrosion (MIC) on buried lines with bitumastic-type coatings and
depleted cathodic protection.
WHAT
IS THE RISK?
On pipelines in excess of 30 years
old, the risks are quite high. If the cathodic protection systems have
depleted, then corrosion will begin at numerous sites all over the pipeline.
Unless detected and retrofitted, the first leak could be the end of the
pipeline, as the next several hundred won't be far behind. There are only so many
clamps that an operator can afford to install before economic concerns dictate
pipeline replacement or abandonment. Given the cost of laying pipelines
offshore today, many of the lines will never be replaced, and this could result
in early deaths of the oil and gas fields they service. Other old lines are the
critical links between the new deep water fields and the shore-based markets.
Loss of these lines will present an interesting and unenviable dilemma for
operators.
WHAT
IS THE ANSWER?
There are three basic strategies
that a pipeline owner can adopt:
1. Survey the pipeline cathodic
protection system.
2. Retrofit the cathodic protection anodes on pipelines of a certain vintage.
3. Do nothing (and hope that the laws of electrochemistry will ignore your pipeline),
essentially ignoring the problem.
CATHODIC
PROTECTION SURVEYS
Close-interval cathodic protection
surveys are the most logical strategy, but strangely enough, operators in the
Gulf of Mexico survey very little. When a survey is actually run, it is usually
of little value because the method used (trailing wire) inherently produces
erroneous data.
There are accurate survey systems available; these either
involve physically contacting the line at intervals or utilizing remotely
operated vehicles (ROV's) (Figure 6) to track the pipeline and carry reference
electrode arrays above the pipeline at known locations (a typical plot from
such a survey is shown Figure 7). This type of survey will let the operator see
the condition of the line and make informed decisions regarding retrofitting.
Work-class ROV Challenger equipped
for pipeline survey. Photo courtesy of Sonsub Inc.
Detailed pipeline CP inspection
plot. Green trace is current density, red trace is potential. Downward green
spikes indicate anode locations; upward spikes reflect coating damage.
In addition to the corrosion data
shown, the survey will also yield important information on the precise location
of the pipeline and the depth of burial below the seabed; these data points can
be crucial when designing the eventual anode retrofit.
RETROFIT
ANODES ON PIPELINES OF A CERTAIN AGE
Retrofitting the cathodic protection
system with supplemental anodes would only make sense if the line in question
is very old and the required additional life were significant. The cost to
perform a pipeline cathodic protection inspection will run anywhere from $2,000
to $6,000 per mile, and that cost may be eliminated if the decision to retrofit
is made. There will only need to be a post-installation survey, once the new
anodes are laid.
Of course, retrofitting pipeline
cathodic protection systems offshore is not always a simple matter, especially
when lines are deeply buried. Often, the retrofit program will need an up-front
survey to find the pipeline - so why not survey it first?
DO
NOTHING
Very often this decision is made
based on the following logic: "If I know I have a problem, I will have to
take care of it; if I don't survey the pipeline, I will not have to find out
whether or not I have a problem." This logic sounds like the chronic
smoker who dares not visit the doctor for fear it will be discovered he has
lung cancer! A surprising number of operators follow this logic.
SUMMARY
In summary, it must be concluded
that cathodic protection plays an absolutely vital role in pipeline integrity
offshore. Cathodic protection is cheap and reliable, with an outstanding track
record of success in offshore applications. But cathodic protection systems
have a finite life and unprotected steel has a very short life in seawater.
Check your cathodic protection if the pipeline is more than 25 years old.
About
the Author
Jim Britton has worked in the corrosion control industry
since 1972 and has been primarily involved in offshore and marine projects
since 1975. He holds a bachelor's degree in corrosion technology from the
United Kingdom. His work brought him to the United States in 1982. In 1986, he
founded Deepwater Corrosion Services Inc., which specializes in engineering and
manufacturing retrofit cathodic protection systems for offshore assets. Britton
is a consultant for major oil companies worldwide. He has published a variety
of articles and has been a guest lecturer at colleges and universities
throughout the US. He has been an active member of NACE International since
1979.
Source:
Source:
http://stoprust.com/technical-papers/26-offshore-pipeline-integrity/
Nice posting.Thank you for sharing this post such a useful information.
ReplyDeleteSteam Pipe Insulation
Cathodic Protection System
Good classification Thanks for sharing.
ReplyDeletePURIT the rust / Corrosion Prevention oil manufacturer has an elaborate range. You can check it out for more information