Hydrocarbon Processing - October 2022 - 60C

Maintenance and Reliability
rectly and accurately. However, the model
was not designed to assess structural integrity
of the entire vessel, such as inside
the vessel, skirt, vessel support, etc.
Views of a cross-section of the PSA
vessel model for the hydrotest load case,
hoop stress, von Mises stress and maximum
principal stress are shown in FIG. 2.
The locations for the extraction of the
stresses for the FFS assessment of the
vessel for weld locations are N02 (top
nozzle), N01 (bottom nozzle), H01 (top
head weld), H02 (middle body weld),
H03 (bottom head weld), V01 (upper
vertical weld) and V02 (lower vertical
weld), as illustrated in FIG. 3.
The through-thickness stress distributions
of the welds selected for the assessment
are shown in FIG. 4. The stress
component was chosen depending on
the crack orientation with respect to the
weld location (e.g., a hoop stress component
for transverse cracks in circumferential
welds and an axial stress component
for transverse cracks in seam or vertical
welds). These results have been summarized
in TABLE 4 in terms of membrane
and bending stresses, which are required
for the fracture mechanics assessments.
FFS assessment. The FFS assessment
and remaining life assessment for the
crack-like defects found in the welds were
carried out in accordance with guidance
given in API579 and BS7910. The assessment
included two major components:
* Determining the criticality of the
defects using a failure assessment
diagram (FAD)
* Calculating the remaining life of the
most critical defect from each vessel.
Flaw geometry, location and orientation.
The most critical defects were selected
from the inspection reports based
on the defects' height and the ligament.
The assumptions made for the FAD assessment
were:
* The lengths of defects were the
same as the weld size (as shown in
TABLE 5), i.e., wall thickness as the
flaw lengths were not available.
* The defects are located at the
highest stress area within the weld.
* Subsurface or embedded defects
are characterized as elliptical cracks,
and surface defects are characterized
as semi-elliptical cracks.
The generic geometry of the elliptical
FIG. 4. Through-wall stress extraction path for cracking at the weld location: (A) top nozzle, N02;
(B) bottom nozzle, N01; (C) top head weld, H01; (D) middle body weld, H02; (E) bottom head
weld, H03; (F) upper vertical weld, V01; and (G) lower vertical weld, V02.
Hydrocarbon Processing | OCTOBER 2022 60C
embedded flaw and semi-elliptical surface
flaw used in the FAD assessment are
presented in FIG. 5.
In this assessment, the lengths (2c as
indicated in FIG. 5) were considered having
the same size as the weld size, and the
crack depths (labeled 2a as an embedded
crack, labeled " a " as a surface crack) were
taken from the inspection report. Flaw locations
and orientations were as specified
by the inspection report. The flaw orientation
was assumed normal to the weld.
Primary and secondary stresses. The
primary stresses used for this assessment
come from the " hydrotest " loading condition
shown in TABLE 4 of the FEA results.
The PSA vessels were subjected to
(PWHT);
post-weld heat treatment
therefore, secondary weld residual stresses
in accordance with BS7910 for PWHT
conditions have been applied. For flaws
that are aligned parallel to the weld, the
residual stress can be assumed as 30% of
the yield strength of the parent material
at room temperature. For flaws that are
aligned transverse to the weld, the residual
stress can be assumed as 20% of the
lesser of the yield strength of the parent
and weld materials at room temperature.

Hydrocarbon Processing - October 2022

Table of Contents for the Digital Edition of Hydrocarbon Processing - October 2022