Page 166 - Pipeline Risk Management Manual Ideas, Techniques, and Resources

Leak volume 7/143
I-in.-diameter hole as the leak size. However, this would not Materials
adequately distinguish between a 36-in.-diameter pipeline and
a 4-in.-diameter pipeline. While a 1-in. hole in either might When different materials and various likely failure modes
cause approximately the same spill or release (initially, at least), are to be included in the risk analysis, the spill size factor of
we intuitively believe that a 36-in.4iameter pipeline presents a the leak impact factor can be adjusted. Although such an
greater hazard than does a4-in.-diameterpipeline, all other fac- adjustment is intended primarily to address widely different
tors being equal. This is no doubt because a much greater materials often encountered in a single distribution system, it
release can occur from the 36-in.-diameter pipeline than from can also he used to address more subtle differences in pipelines
the 4-in. line. of basically the same material but operated under different
The hole size is determined by the failure mode, which in conditions. For example, a higher strength steel pipeline
turn is a function of pipe material, stress conditions, and the usually has slightly less ductility than Grade B steel and, when
failure initiator. combined with factors such as changing stress levels and crack
As an extreme example of failure mode, an avalanchefail- initiators, this raises the likelihood of an avalanche-type line
ure, is characterized by rapid crack propagation, sometimes for break.
thousands of feet along a pipeline, which completely opens the An important difference lies in materials that are prone to
pipe. Main contributing factors to an avalanche failure include more consequential failure modes. A large leak area is usually
low material toughness (a more brittle material that allows characterized by the action ofa crack in the pipe wall. A crack is
crack formation and growth), high stress level in the pipe wall more able to propagate in a brittle material; that is, a brittle pipe
(usually at the base of a crack), and an energy source that can material is more likely to fail in a fashion that creates a large
promote rapid crack growth (usually a gas compressed under leak area--equal to or greater than the pipe cross-sectional
high pressure). area. This problem is covered in more detail in a discussion of
In many applications, a risk assessment model does not fracture mechanics in Chapter 5.
attempt to distinguish among likely failure modes-a worst The brittleness or ductility of a material is often expressed
case scenario is often assumed for simplicity. Distinguishing in combination with its strength as a material toughness or
between leak types adds a degree of complexity to the model; fracture toughness. Important material factors influencing
however, the added information that is provided can be useful in toughness in pipeline steels include chemical composition
some cases. In a general sense, the leak size probabilities can (percentage of carbon, manganese, phosphorus, sulfur, silicon,
somewhat offset an otherwise higher consequence event. For columbium, and vanadium), deoxidization practices, cold
example, a smaller diameter line more prone to large breakage work, and heat treatments [65]. The challenge of gauging the
can equal the consequences ofa larger line that is prone to small likelihood of a more catastrophic failure mode is further com-
pinhole leaks. plicated by the fact that some materials may change over time.
A spill size probability distribution can be developed from Given the right conditions, a ductile material can become more
an examination of past releases. This is further discussed in brittle.
Chapter 14. Material toughness is an important variable in the potential
We will assume, if only for the sake of simplification, that a for certain failure modes. Even in the same material, slight dif-
larger hole size leads to a larger leak and that a larger leak has the ferences in chemical composition and manufacture can cause
potential for more severe consequences. Of course, under the significant differences in toughness. The most common
right circumstances, a large- or small-area failure can be equally methodused to assess material toughness is the Charpy V-notch
consequential in the pipeline system. For example, amore ductile impact test. This test has been shown to correlate well with frac-
failure that allows only a minor pipe wall tear can leak undetected ture mechanics in that test results above certain values ensure
for long periods, allowing widespread migration of leaked prod- that fatigue-cracked specimens will exhibit plastic behavior in
uct. A more violent break of the pipe wall, on the other hand, may failure. Charpy-Izod test results for some common pipeline
cause a rapid depressurization and quick detection of the prob- materials are shown inTable 7.3.
lem. See also the discussion of leak detection in Chapters 7 and The ASTM has reported [7] that the tensile stress behavior of
1 1 for more details regarding possible leak volumes. steel is not well correlated with its behavior in notched impact
Because of the many different materials and conditions that tests such as the Charpy test. In other words, acceptable ductile
may need to be compared when studying some pipeline sys- behavior seen in tension failures sometimes becomes unaccept-
tems, a consideration can be included to allow for higher or able brittle behavior under notch impact failure conditions.
lower anticipated incidents of large openings in a pipe failure. Therefore, specifying minimum material behavior under ten-
One intent is to make a distinction between pipes more likely sile stress will not ensure adequate material properties from a
to fail in a catastrophic fashion. This is highly dependent on fracture mechanics standpoint. Impact testing or some equiva-
pipe material toughness. Where pipe material toughness is con- lent of this is needed to ensure that material toughness proper-
stant, changing pipe stress levels or initiating mechanisms will ties are adequate. Until the last decade or so, material toughness
govern. or material ductility was not normally specified when pipe was
Figure 7.5 shows a model of the interrelationships among purchased.
some of the many factors that determine the type of pipeline The rate of loading and the temperature are important param-
leak that is likely. Initiating mechanisms that promote cracks eters in assessing toughness. The likelihood of brittle failure
are more likely to lead to a large leak than are mechanisms that increases with increasing speed of deformation and with
cause a pinhole-type leak. (See Chapter 5 for further discussion decreasing temperature. Below a certain temperature, brittle
on fracture mechanics and crack propagation.) fracture will always occur in any material.