Pipeline defect assessment basics

Chapter 13

Pipeline defect assessment

basics 1

Dr Phil Hopkins

13.1 Introduction

Structures are made from materials, and materials fail. A failure means a structure
cannot carry out its function or cannot meet its specification or expectations. These
structures can fail for a variety of reasons ranging from being overloaded beyond their
design limits, to developing a defect (such as a crack) during service.
Defects in structures can decrease their ability to function, and can lower their
strength; consequently, we need to assess the effect of any defect in a structure (see
Figure 13-1).
This Chapter covers the effect of defects in structures, and methods for assessing
these defects, but first we need to understand why structures fail, and what we mean
by a defect.

13.1.1 Why do structures fail?

Structures have to withstand forces; for example, tall buildings have to withstand
forces created by the wind, and road surfaces have to withstand the forces from passing
traffic. These forces can lead to failures.

1  This Chapter is based on the e-learns available at the Competence Club (https://competence.rosen-group.com/)
and is courtesy of Rosen Group.

383
Managing Pipeline Threats

What is stress in a structure? Why are structures sensitive to these forces? A force
is a push or pull on an object. The force causes the object to stretch or compress. This
stretching or compression creates a strain in the object. We can see this strain as defor-
mation of the object. Stress is created in the object by this straining: it is the resistance
to the straining. Stress is always within the object and is caused by the deformation.
These stresses can be so high, that they fail a structure. Obviously, structures cannot
withstand higher and higher stresses: they are limited by the strength of the materi-
als the structure is made from. These materials have a limiting strength. This limiting
strength is a material property called the yield strength or tensile strength, and when
this strength is exceeded, the material (and hence the structure) fails.

Cause of failures. We have all seen failures, ranging from broken hinges on the doors
in our house, to cracks in our highways. Structures fail due to a combination of:

• the forces (or loads) on the structure (the forces on a structure cause a stress in
the structure);
• any defect in the structure (a material or design defect will weaken the structure,
or elevate stresses, and accelerate failure);
• the properties of the materials that make up the structure (strength, shape, etc.);
• time the force operates (the longer a structure operates, the more likely it is to
fail (for example, due to repeated loads called fatigue));
• the environment the structure operates in (e.g., temperature, corrosion, or
irradiation, can, over time, weaken the material).

This Chapter is concerned with defects, such as dents and cracks, failing structures.

13.1.2 What is a defect?

We have many definitions, or alternative words, for a defect in a structure (see

Table 13-1). Most describe a defect as something that should not be there or exceeds a
limit that has been specified or agreed for that structure.
It is often easy to see defects in structures. Most of the homes we live in will have
cracks in the walls and if we own a bicycle, we will often experience a flat tire caused
by a hole in its inner tube (see Figure 13-2). These defects are not going to cause an
immediate disaster, but we do need to know if they pose a longer-term problem, and if
we need to replace or repair the defective component.

384
 Pipeline defect assessment basics

Table 13-1. Definitions of defect, etc. [1]

Term Definition

defect “imperfection of a type or magnitude exceeding acceptable criteria”.

imperfection “flaw or other discontinuity noted during inspection that may be subject to acceptance
criteria during an engineering and inspection analysis”.

flaw “imperfection that is smaller than the maximum allowable size”.

discontinuity “interruption of the typical structure of a material, such as a lack of homogeneity in its
mechanical, metallurgical, or physical characteristics”.

anomaly “a possible deviation from sound… material or weld”.

QR Code 13-1. Scan here for two case studies of pipeline failures caused by defects.

Figure 13-1. Defects in structures can lower their strength.

Figure 13-2. Defects, and their effects, are a common sight.

385
Managing Pipeline Threats

13.1.3 Defects in pipelines

Pipelines can contain defects such as corrosion and gouges (see Figure 13-3). De-
fects can be caused during the manufacture of the line pipe (our pipeline is constructed
using tubular steel sections called line pipe), fabrication of the pipeline on site, or op-
eration of the pipeline.
Most of these defects are introduced during the operation of our pipeline by threats
to the pipeline; for example, buried pipelines can be damaged by earth moving equip-
ment that is not aware of the presence of the pipeline. The damage caused by this ex-
ternal interference can puncture, gouge, or dent the pipeline. The threats to a pipeline
are summarised in various publications (for example [2]). Table 13-2 gives a simple
summary.
These threats create differing types of defects in the pipelines; for example, cor-
rosion pits, dents and gouges in the pipe, and cracks in welds in the pipe. The threats
can also fail the pipeline; for example, Figure 13-4 shows the main causes of failures in
European natural gas pipelines (from 2007 to 2016).
It should be noted that some threats are on the increase: theft of product from
pipelines (by illegally tapping (drilling) into the pipe) is on the increase in many coun-
tries; for example, Figure 13-5.
Figure 13-4 and Table 13-2 show that pipelines may contain defects such as corro-
sion and cracks; therefore, we must ensure we detect these defects before they can fail
the pipeline, and then assess their significance when we detect them.

Figure 13-3. Examples of defects in pipelines. Photographs courtesy and copyright of Penspen Ltd, UK.

386
 Pipeline defect assessment basics

Table 13-2. Threats to a pipeline.

Threats

Corrosion (on the internal or external surfaces of the pipe).

External interference (impacts on the pipeline (e.g., from earth-moving equipment (onshore pipelines) or
anchors (subsea pipelines) causing defects such as dents, gouges, or punctures).
Material defects (in the pipe and its welds).
Cracks in the pipeline (e.g., caused by fatigue).
Natural environmental force/hazard (e.g., cold weather or floods damaging the pipeline).
Man-made environmental force/hazard (e.g., mining subsidence).
Design errors.
Equipment error (e.g., problems with valves, flanges).
Operation/operator (human) error.
Vandalism/theft/sabotage/terrorism.

External interference (dents, gouges, etc.)

Corrosion (internal or external)

Construction Defects (e.g., in welds)/

Material Failures (e.g., in the line pipe)

'Hot Tap' (errors while connecting into/working on

a pressurised pipeline)

Ground Movement

Unknown/Other (e.g., design error, lightning, etc.)

Figure 13-4. Causes of failures in onshore European gas pipelines [3].

Figure 13-5. The increasing problem of theft (taps) from pipelines (data from Italy)[4].

387
Managing Pipeline Threats

13.1.4 Assessment of defects: fitness for service

A defect normally means something not within a requirement in a standard (see

Table 13-1) so you might ask ‘Why assess a defect? All defects should be repaired?’ This
is a reasonable question, but an unreasonable solution, as all structures contain defects.
Nothing is perfect; therefore, defects in new structures are necessarily limited by quality
control levels. These quality levels are usually based on past experience of building the
structure and are the limits of good workmanship. When you buy a new house, you are
expecting good workmanship, but you would not expect perfection.
As a structure ages we find defects in excess of these original quality control levels
(e.g., Figure 13-2). A house is a good example of aging defects, as you cannot compare
a very old house with a new house. Old houses have many defects which would not
be accepted in a new house, but you would not repair every defect, or replace the old
house. You would evaluate the defects and determine if they are a threat to the safety
of the house.
We need to determine if these larger, or more numerous defects, need to be re-
paired or not. We do this by assessing the defects using engineering calculations. A
British Standard[5] gives the reasoning behind these assessments:

‘In circumstances where it is necessary to examine critically the integrity of new

or existing constructions… it is also necessary to establish acceptance levels for
the flaws revealed. The derivation of acceptance levels for flaws is based on the
concept of fitness for purpose. By this principle, a particular fabrication is consid-
ered to be adequate for its purpose, provided the conditions to cause failure are
not reached. A distinction has to be made between acceptance based on quality
control and acceptance based on fitness for purpose.’

‘Quality control levels are usually both arbitrary and conservative but are of con-
siderable value in the monitoring and maintenance of quality during production.
Flaws that are less severe than such quality control levels as given, for example,
in current application standards, are acceptable without further consideration.
If flaws more severe than the quality control levels are revealed, rejection is not
necessarily automatic. Decisions on whether rejection and/or repairs are justified
may be based on fitness for purpose, either in the light of previously documented
experience with similar material, stress and environmental combinations or on
the basis of an “engineering critical assessment” (ECA).’

388
 Pipeline defect assessment basics

In summary:

• fitness for service, or fitness for purpose assessments are quantitative

engineering evaluations that are performed to demonstrate the structural
integrity of a component containing a defect[6];
• these evaluations are often called an engineering critical assessments (ECAs),
and are detailed in international standards (e.g., [5, 6]); and,
• these evaluations use theories and models based on fracture mechanics.

The evaluation will tell us how much the strength of a structure is reduced by the
presence of a defect. We will cover fracture mechanics later in this Chapter.

13.1.5 Perspective on defect assessment

Defect assessment methods provide a quantitative measure of the acceptability of a

defect in a structure – they will calculate the stress the structure will fail at, due to the
defect, and we can then decide if this stress is too low. The methods are both useful and
powerful, but they must not to be used in isolation: they are part of an overall process
for the management of defects, and the wider integrity management plan for the struc-
ture or system[5]. This process needs to address many factors, including[5]:

• the cause of the defect, and remedial action to prevent further occurrences or
growth;
• whether a previous inspection failed to detect this defect, and if the defect has
not been detected, what are the reasons;
• the previous history of the structure and whether it can explain the occurrence
of the defect (its nature, location and size);
• whether an inspection indicates that the defect has grown, and the observed
growth is consistent with assumptions about loading and defect growth rates,
after allowing for uncertainty in the inspection results;
• the implications for other structures of the same or similar design and whether
modifications to the structure, or a change in the service conditions, might be
required; and,
• whether there is a pattern of the defect being detected in other structures of the
same design.

Clearly, there is a need to conduct a full engineering evaluation of the cause and
effect of any defect detected in a structure and manage the whole process. The next
section gives an example of what this management of defects entails.

389
Managing Pipeline Threats

13.1.6 Example of defect management

Cracks can be created during the manufacture of materials and products, and the
fabrication of structures. This Chapter is concerned with assessing the significance of
defects, such as cracks, but the assessment is just one part of the management of a de-
fect. This section briefly explains how pipeline operators manage pipelines containing
cracks.
Cracks can be present in pipelines. They could be present in the line pipe used to
construct the pipeline, or introduced during the construction of the pipeline, or during
operation.
Pipelines operate under a safety management system (for example [7]). These sys-
tems identify and manage risk and improve safety. A key element of a safety man-
agement system is the integrity management program (for example [8, 9]), where the
policies, processes, and procedures to prevent failures are documented.
These systems give an overall process to manage a wide range of defects in a pipe-
line, but it is better to have a special process for some defects that are both common
and dangerous. This is the case for cracks in pipelines: a specific management process
is available[1].
Crack management involves:

• identifying the effect of the cracking on pipeline integrity;

• conducting a risk assessment on the pipeline to determine the segments of the
pipeline at most risk; and,
• establishing reassessment (by inspecting the pipeline or testing the pipeline)
frequencies.

This management will require[1]:

• determining the necessary pressure reductions needed in a cracked pipeline;

• presenting suitable methods for assessing the condition of the cracked pipeline;
• establishing criteria for inspection and testing, to detect the cracks and prove
integrity;
• calculating the remaining lives of cracks remaining in the pipeline; and,
• developing preventive and mitigative measures.

This process shows that defect assessment is just one part of defect management.
We will now start to explain how we assess defects, but first we need to explain the
basis of defect assessment methods. This basis is made up of:

• structural mechanics (the effect of loads and deformations on materials); and,

• fracture mechanics (the effect of a defect such as a crack on materials).

390
 Pipeline defect assessment basics

13.2 Structural mechanics

All structures are subjected to loads and deformation. Structural mechanics is the
study of stresses (loads) and strains (deformation) in engineering structures.

13.2.1 Early days and the strength of structures

We have many magnificent, beautiful, ancient engineering structures (e.g., Egypt’s

pyramids, or the Roman viaducts) that have survived for thousands of years. We know
these structures must have been designed, but we do not know how. We suspect it was
mainly by trial and error, or design and failure… but the results are impressive. Remem-
ber, these early pioneers drew with quills and ink, worked under candlelight, and had a
life expectancy of 30 years….
Our ancestors understood the importance of the strength of a structure: the fa-
mous Roman architect Marcus Vitruvius (70 BC - 15 BC) wrote that an architect
should focus on three themes when designing a building:

• firmitas (strength);
• utilitas (functionality); and,
• venustas (beauty).

His ideas of beauty were based on symmetry and proportions.

The modern development of structural mechanics began with an increasing de-
mand for the structural analysis of domes in medieval cathedrals. The best examples
are the dome of Santa Maria del Fiore in Florence (1415 -1436), and the dome of St
Peter’s Basilica in Rome (planned in 1506, but completed in 1590). The only available
design method at that time was to build a small-scale model of a dome; however, this
method was not very reliable, because of the scale effects.
What are scale effects? We can easily build small models of structures. It is easy
to build a small boat to model a large boat. It is easy to model a large building with a
small-scale model, but these models will not use the same materials, the same thick-
nesses, the same joining processes, etc., and can only present a model that will not rep-
resent the full-scale structure. A good example was designing obelisks: small obelisks
were simple to build, difficult to fail, and lasted many years. When an engineer wanted
to create a much taller, heavier obelisk, it was very difficult to build, fragile, and would
fail after a few years. The scale effect was evident….
The main motivation in the early development of structural mechanics was to un-
derstand the strength of materials, and therefore control the fracture of materials used
in new and larger buildings. There was an increasing demand for bigger buildings (cas-
tles, cathedrals, bridges, etc.), and these bigger buildings (higher, wider, deeper, and
heavier) created huge loads on the supporting structures.

391
Managing Pipeline Threats

Many great engineers and scientists were attracted to designing and understanding
structures; for example, Michelangelo (1475- 1564) was involved in the design of the
Basilica dome in Rome. Galileo Galilei (1564-1642) had an interest in many things,
including the strength of materials (see Figure 13-6). He investigated the breaking
loads of rods under tension and correctly concluded that the load was:

• proportional to the cross-section area of the rods; but,

• independent of the length of the rods.

This does not seem a major finding today, but hundreds of years ago engineers were
relying on intuition, experience, models, and trial and error to build bigger structures.
They did not have structural theories, or detailed data on the strength of materials.
Their work was primarily heuristic.
The world had to wait for Leonhard Euler’s theory of column buckling (1757),
Thomas Young’s work on the strength of materials (1807), Louis Navier’s theory of
beams (1826), and Otto Mohr’s work on calculating forces on a structure (1874). These
works allowed engineers to design and build bigger and better structures.

13.2.2 The effect of defects

Before Galileo, Leonardo da Vinci (1452-1519) had sketched in his notebooks a

test of the tensile strength of a wire and investigated metal failure (see Figure 13-6). He
observed a fact: the strength of a material was dependent on the presence of defects.
Leonardo da Vinci was the first engineer to document an investigation into the
fracture of materials. He was assessing the strength of materials, using long iron wires,
and (correctly) assumed that the strength of a wire was constant (for a specific length
of wire tested). He tested iron wires of various lengths, loaded by baskets of sand, and
was surprised to observe that the strength of the wires was inversely proportional to
the length of wire: short wires failed at higher loads than long wires. Da Vinci correctly
concluded that the long wires had more chances of containing a defect, than the shorter
wires. The presence of these defects lowered the wire’s failure load: long wires failed at
low loads, short wires failed at high loads.
Defects in a structure greatly decrease strength, so, for a structure to be strong –
and withstand high loads – it must also be defect-free. Unfortunately, nothing is per-
fect, and all structures/bodies contain defects: the secret of a strong structure/body is to
keep defects to a minimum.

392
 Pipeline defect assessment basics

13.2.3 Can giants exist?

Galileo is thought to have merged:

• his own thoughts of cross-sectional area being related to strength; and,

• Leonardo da Vinci’s observation of small defects weakening a structure,

to note that giants (should they exist…), would be very fragile creatures. Why?
Well, we can expect a giant’s strength to be greater than ours, since the cross-sectional
areas of their skeleton and muscles would be larger, but their bones would be long and
wide, giving a greater volume (their bone volume would be huge compared to ours).
Also, their weight (and the loads the bones must support), would also increase;
therefore, the chances of defects in their bones would greatly increase, as would the
loads on their bones. This means a simple fall may break their giant bones. Conclusion?
Do not be afraid if you are confronted with an aggressive giant. Just push him/her
over….

QR Code 13-2. Scan here for more understanding of structural mechanics.

Figure 13-6. Galileo’s and da Vinci’s tests.

393
Managing Pipeline Threats

13.3 Fracture mechanics

All structures will contain defects; for example, Figure 13-7. Fracture mechanics
is the study of the effect of defects on structures subjected to forces and deformations.
Our early scientists and engineers such as Galileo and da Vinci were interested in how
defects affected a structure’s performance. Early studies into fracture mechanics were
limited, as there was no great demand for it – if a structure cracked, or failed, you simply
replaced it, or ensured the next design of the structure was thicker and stronger. Also,
very small defects such as cracks could not be seen or detected, so little could be done
to control them.
The 20th century witnessed bigger, longer, heavy, and faster structures, such as
ships, being built. Additionally, new structures, such as jet aircraft, were put into ser-
vice. These increases in size, weight, and speed put greater forces and deformations on
the structures, leading to failures, and a need to understand why these structures failed,
particularly from cracking.
Cracks were originally thought to be small and insignificant, and not able to be a
threat to these new, large, elegant machines. This was not the case. Ships and aircraft
experienced failures due to cracks. Two famous examples were:

• The Liberty ships (see Figure 13-8) were ships built in the USA, during the
Second World War. They were used to transport goods across the Atlantic to
Europe and bring the region liberty. These ships experienced serious fractures:
cracks ran through the ship’s hull, causing dramatic failures. The cracks in the
ship’s hull propagated rapidly and were not arrested by the hull steel, as the steel
had a low toughness (see later).
• The world’s first passenger jet aircraft was de Havilland’s Comet, Figure 13-8. It
first flew in 1949.

◊ Passenger aircraft in the 1940s were often uncomfortable converted slow

wartime bombers, and all were noisy (because of piston engines and
propellers), with unpressurised cabins (meaning no air to breathe at high
altitudes, so they had to fly low where there is plenty of air), and this low-
flying meant they had to fly through, rather than above, the weather with
all its discomforts and reduced speeds. The Comet was high flying (12 km,
avoiding bumpy and noisy weather), with a pressurised cabin, comfortable,
with many windows, quiet (jet engines), and it was fast (500 mph (807
km/h)).
◊ Three fatal crashes of Comet aircraft in 1953-54 were caused by cracks
growing from corners of the square fuselage windows. These square
corners served as stress risers, accelerating crack formation, and the cracks
subsequently grew in a fuselage stressed by the necessary pressurisation

394
 Pipeline defect assessment basics

during high altitude flight. This pressurisation of the fuselage is repeated

many times for passenger comfort: it is pressurised at the start of each
flight and depressurised when the engines shut down. This repeated
increase and decrease of pressure causes fatigue. Fatigue is caused by
repeated stresses on a structure and assists cracks to grow in size.

Defects such as cracks create very high stresses in a structure and can cause a ma-
terial to fail below its yield strength. Clearly, the high stresses created at a crack tip are
a problem.

Figure 13-7. Crack in a pipeline.

Figure 13-8. Examples of cracks causing failures (left, a Liberty ship, right, the Comet aircraft).

395
Managing Pipeline Threats

13.3.1 Applying fracture mechanics

In structural mechanics, we compare a remotely applied stress on a defect-free

structure, to the yield strength of the material of the structure. If applied stress>yield
strength, I assume my structure is overloaded, and liable to fail (see Figure 13-9). In
effect, we are comparing load (applied stress) with material resistance (yield strength).
It is similar in fracture mechanics: we need a measure of the local stresses at the
crack tip caused by the remotely applied stress. We need a measure of the material’s re-
sistance to these local stresses (see Figure 13-9). We will discover that the resistance to
these stresses is provided by the material’s toughness: the higher the material’s tough-
ness, the more resistance to the presence of a crack. And finally, we need an equation
that links this load and resistance, to give us a failure condition (see Figure 13-9). These
equations are published in standards (for example [5, 6]), and they allow us to assess
defects in any structure.
Therefore, to be able to use fracture mechanics to predict the behaviour of the de-
fect, we will need information about (see Figure 13-9):

• the defect (e.g., its size);

• material properties (e.g., toughness): and,
• the applied stresses on the structure.

We input this information into an equation that predicts if the defect will fail, or
not (see Figure 13-9).
Organisations, such as British Standards Institute (BSI) and American Petroleum
Institute (API) have produced documents (BS 7910[5], and API 579[6]) that give a full
fracture mechanics methodology. They give the necessary equations, procedures, etc.,
and help with the necessary assumptions and safety factors. API 579 considers the
following systematic approach to defect assessment:

• defect and damage mechanism identification (cause?);

• applicability and limitations of assessment procedures (can I assess it?);
• data requirements (have I the data?);
• assessment techniques and acceptance criteria (what method will I use?);
• remaining life evaluation (how long can the defect survive?);
• remediation;
• in-service monitoring; and,
• documentation.

396
 Pipeline defect assessment basics

QR Code 13-3. Scan here for more understanding of fracture mechanics.

In fracture mechanics, we need a measure of the local stresses at the crack tip, caused by the remotely applied
stress. Then, we need a measure of the material's resistance to these local stresses.
The resistance to these stresses is provided by the material's toughness: the higher the material's toughness, the
more resistance to the presence of a crack.

Figure 13-9. Applying fracture mechanics.

397
Managing Pipeline Threats

13.4 Defect assessment in pipelines

13.4.1 The early days

Most oil and gas pipelines were built between 1950 and 1980, and this was a period
when fracture mechanics was still developing, and defect assessment was relatively new.
Some pipelines were showing signs of age in the 1960s (e.g., corrosion), particularly in
the USA.

13.4.2 The pioneers

The research organisation Battelle in Columbus, Ohio, USA did the first major
fracture mechanics work on pipelines. Fracture mechanics methods had been available
since the 1950s, but in the 1960s they were still new and not well-developed. Methods
did exist in the 1950s for brittle materials, and thick-walled structures (such as nuclear
pressure vessels), but these methods were not directly applicable to thin-walled struc-
tures such as pipelines, which were made from relatively tough and ductile materials.
Additionally, the pipeline industry did not measure the fracture toughness (in terms
of quantitiative measures such as stress intensity factor (K), J-integral ( J), or crack tip
opening displacement (CTOD)) of its pipelines, to allow it to use the various fracture
mechanics methods. The only toughness measure the pipeline industry had was the
Charpy toughness, which is a qualitative measure of toughness, and not easily used in
fracture mechanics.
Battelle and other workers conducted full scale tests on defective pipeline sections
to overcome limitations in fracture mechanics for pipelines, and the absence of quanti-
tative measures of toughness. They developed semi-empirical (a combination of theory
and experiments) methods to predict the failure pressure of defects such as corrosion
and cracks in pipelines. Battelle produced fracture mechanics equations to predict how
defects would behave, but they had to use the Charpy toughness as their toughness.

13.4.3 The problem with Charpy

The pipeline industry still uses Charpy toughness and does not have fracture
toughness data in terms of a quantitative measure of toughness (i.e., K, J or CTOD).
Charpy is not a quantitative fracture toughness; therefore, it has to be correlated with
quantitative fracture toughness measures (e.g., the stress intensity factor, K).
This Charpy-K correlation is a particular problem when we assess cracks in pipe-
lines. API 579 and BSI 7910 use recognised correlations, but these correlations intro-
duce inaccuracies. The toughnesses used in these standards may be only one third of

398
 Pipeline defect assessment basics

the actual toughness. This leads to low predictions of failure stress for cracks, and the
problem of multiple repairs.
These correlations are not a problem when we assess corrosion. Corrosion is a
blunt defect and is not sensitive to toughness, and its failure is mainly governed by the
strength of the pipe. This is not the case for a crack – the failure of a crack in pipelines
can be governed by the toughness. This means that corrosion assessment is not a prob-
lem: the fracture mechanics models we use are reliable.

13.4.4 Assessment of corrosion, gouges, dents. and cracks in pipelines

Corrosion. Pipelines inevitable corrode as they age. The Battelle workers proposed an
equation to predict the failure pressure of corrosion in pipelines (see Figure 13-10). The
equation only needs inputs of the corrosion length and depth, pipeline radius and wall
thickness, and the strength of the line pipe steel.
These types of equations have been adopted in standards; for example, ASME
B31G[10]. Figure 13-10 is an acceptance curve obtained from ASME B31G, for a spe-
cific pipeline geometry and strength. The equations given in the standard can be used
to construct these acceptance curves, which allow a simple and quick method for de-
termining acceptance.
Note that acceptance curves are not failure curves:

• If the defect is below the acceptance curve, it is acceptable and does not need
repairing. This acceptable defect will not fail.
• If the defect is above the acceptance curve, it is not acceptable and needs
repairing. This unacceptable defect will not necessarily fail the structure, as the
acceptance curves contain safety factors.

Failure curves do not contain safety factors – if the defect is above the failure line,
it will fail the structure.
There are many methods for assessing corrosion in pipelines, but the most popular
methods are detailed in ASME B31G[10]. ASME B31G is the benchmark standard for
assessing corrosion but is limited to internal pressure loadings only. It cannot be applied
to pipelines subjected to external loads; for example, from ground movement.

Gouges. Pipelines are usually either buried below ground or are subsea. They can be
damaged by earth excavations, anchors, etc. This can lead to damage such as gouges
or dents. Therefore, a gouge is surface damage to a pipeline, usually caused by foreign
objects removing part of the pipe wall.
Gouges can be assessed using similar equations to the ones used with corrosion
(see Figure 13-10). Failure curves for gouges can be developed using these equations
(see Figure 13-11).

399
Managing Pipeline Threats

Figure 13-10. Example of a corrosion assessment equation, and an acceptance curve

(Photograph courtesy and copyright of Penspen Ltd).

400
 Pipeline defect assessment basics

Figure 13-11. Example of failure curve for gouges in pipelines under internal pressure.

Figure 13-12. The failure assessment diagram (FAD) in API 579 and BS 7910, used for the assessment
of cracks.

401
Managing Pipeline Threats

Dents. Pipelines can be dented during construction and/or operation. These dents can
be introduced by accidental impact from earth moving equipment when the pipe is
buried, anchor damage when the pipe is subsea, etc.
The pipeline industry has always adopted a pragmatic view on dents in pipelines
and accepts that dents will be present in operational pipelines. Pipeline standards de-
tail acceptance levels for dents; for example, Table 13-3 shows the size/type of dents
requiring removal or repair or an engineering evaluation.

Cracks. Cracks can be created during the manufacture of our line pipe (for example, by
poor steel making processes), or its construction (for example, by poor welding). These
cracks are inadvertently ‘built into’ a pipeline. Quality control, with its inspection and
testing, can eliminate most of the manufacturing and construction cracks: they need to
be detected and repaired, as cracks can cause pipeline failures.
Cracks can also be created during pipeline operation. Corrosion, and fatigue can
create and grow cracks in a pipeline; therefore, we can sometimes find cracks in pipe-
lines, and they will require an assessment.
Cracks are usually assessed using published procedures. The two most popular pro-
cedures - API 579 and BS 7910 - use fracture mechanics methods. The standards use a
failure assessment diagram (FAD) to calculate the failure stress of defects in a structure
(see Figure 13-12). The FAD was originally developed by the nuclear industry in the
1970s, but it now used in many other industries.

Table 13-3. Dent acceptance levels for operational pipelines[11].

Dents requiring removal or repair or an engineering evaluation

Dents with a depth greater than 6% of the pipe external diameter [0.250 in. (6.4 mm) in depth for a pipe
diameter 4.5 in. (114.3 mm) and smaller].
Dents containing gouges, grooving, scratches, cracking or other stress riser.
Dents containing metal loss resulting from corrosion or grinding where less than the minimum wall thickness
permitted by the component specification remains.
Dents that affect pipe curvature at a girth weld or a longitudinal seam weld.

402
 Pipeline defect assessment basics

The FAD needs three main inputs (data) (see Figure 13-12):

• crack size (length, depth, shape);

• applied stresses on the cracked structure;
• material properties (fracture toughness, and yield/ultimate tensile strength).

The standards detail all the necessary equations to determine if the crack is accept-
able. These calculations are in two parts:

• one calculation (focussed on fracture - Kr) requiring the fracture toughness of

the cracked material; and,
• the other calculation (focussed on overload (collapse) of the material around the
crack - Lr), requiring the strength of the material around the crack.

The results of these two calculations are then plotted on the FAD, to determine if
the crack is acceptable or unacceptable (see Figure 13-12).

QR Code 13-4. Scan here for more understanding of pipeline defect assessment.

13.5 Summary

This Chapter has covered how we assess defects, such as corrosion, gouges, dents,
and cracks in a structure.
We have been introduced to fracture mechanics, and discovered it is the applica-
tion of structural mechanics to the problem of predicting the behaviour of a crack in
a structure. Fracture mechanics defines the local conditions under which a defect such
as a crack will extend, by using more global parameters such as load and geometry. We
have discussed these parameters that model the conditions at the crack tip (K, J, and
CTOD), and have discovered that when these values of K, J, or CTOD reach a critical
value (the material’s fracture toughness, Kc, Jc or CTOD), the crack fails the structure.
The analogy is the stress in a defect-free structure reaching the yield strength of the
structure’s material, and failure ensuing.
It is important to note that defect assessment is just one part of the management
of defects in a pipeline. This management includes inspection criteria, repair methods,
etc. We must never treat defect assessment in isolation – it is part of our safety man-
agement system.

403
Managing Pipeline Threats

Finally, we will end with some advice. We can assess many defects in pipelines,
and there are many documents/software to help us in these assessments, but whatever
defect we assess, we need to remember… the 5 Rules of Defect Assessment:

• always understand the cause of any defect you are assessing;

• understand and use the best assessment practices;
• use all relevant data (e.g., inspection data, operations records, maps, etc.);
• check calculations, inputs, outputs, and assumptions; and,
• always appreciate the consequences of any failure of the defect.

And… we have one final piece of advice. You will be dealing with pipelines con-
taining defects. You may have to assess a pipeline containing many thousands of de-
fects. But remember… sometimes the pipeline contains so many defects… its condition
is so poor… it is beyond defect assessment. Advice? Never assess a mess….

13.6 References
[1] Anon., ‘Recommended Practice for Assessment and Management of Cracking in pipelines.
American Petroleum Institute. API 1176. July, 2016.
[2] Anon., ‘Pipeline Risk Modelling Overview of Methods and Tools for Improved Implementation’,
Pipeline and Hazardous Materials Safety Administration, USA. Draft 1. May 9, 2018.
[3] Anon., ‘Gas Pipeline Incidents: 10th Report of the European Gas Pipeline Incident Data Group
(period 1970 – 2016)’, EGIG. Doc. number VA 17.R.0395. March 2018.
[4] http://www.ukopa.co.uk/wp-content/uploads/2016/06/June-2016-Newsletter-Final-Version.pdf
[5] Anon., ‘Guide to methods for assessing the acceptability of flaws in metallic structures’. BS
7910:2013. British Standards Institution. UK. 2013.
[6] Anon., ‘Fitness For Service’, API 579-1/ASME FFS-1. American Petroleum Institute. June, 2016.
[7] Anon., ‘Pipeline Safety Management Systems’, ANSI/API Recommended Practice API 1173. July,
2015.
[8] Anon., ‘Managing System Integrity of Gas Pipelines’, American Society of Mechanical Engineers.
New York, USA. ASME B31.8S-2018. Supplement S of ASME B31.8. 2018.
[9] Anon., ‘Managing System Integrity for Hazardous Liquid Pipelines’, American Petroleum Institute.
API Recommended Practice 1160. Third Edition. 2019.
[10] Anon., ‘Manual for Determining the Remaining Strength of Corroded Pipelines’, American Society
of Mechanical Engineers. New York, USA. ASME B31.G-2012. 2012.
[11] Anon., ‘Pipeline Transportation Systems for Liquids and Slurries’, American Society of Mechanical
Engineers. New York, USA. ASME B31.4-2016.

404