Conference PaperPDF Available

Problems Associated with Coiled Oilfield Tubing

Authors:

Figures

Content may be subject to copyright.
Problems Associated with Coiled Oilfield Tubing
Roderic K. Stanley
Houston, Texas, USA
Abstract
Studies over the past 5 years have indicated that the major
problems associated with coiled steel oilfield tubing are
corrosion, usage and handling damage, manufacturing
imperfections, and poor welding practices. This paper
presents industry results of these studies, and indicates how
damaging to coiled tubing (CT) operations these problems
can be. Further, it proposes the use of a combination of
non-destructive testing (NDE) and fatigue life modelling as
the appropriate solution in order to derate strings, or
sections of strings, and to provide for a classification of in-
service strings just as other oilfield tubulars are re-
classified.
NDE is currently performed for coiled tubing in an
haphazard manner, thus the progress in quantifying the
required NDE, under a proposed API recommended
practice will be discussed.
Coiled Tubing and Line Pipe
Coiled Tubing (CT) originated as small diameter butt-
welded tubing for well servicing use. The tube-to-tube
weld was a weak point, possibly because poor NDE was
used (double wall RT, and PT for the OD surface), and the
need for more consistency arose. By joining strips together
end to end and running the accumulated strip through an
ERW mill, long tubes can now be made. Typical lengths of
a CT string are 20,000 - 25,000 ft. Strip lengths are from
1200-3200 ft, depending upon the thickness, and welded at
an angle. Constant mass slabs are used to hot-roll the strip.
Table 1: Physicals for New Coiled Tubing
Grade Yield Str
(Min)
Tensile Str
(Min)
Max
Hardness
Elongation
Kpsi (MPa) Kpsi (MPa) HRC %
CT70 70 (483) 80 (551) 22 K1A0.2/U0.9
CT80 80 (551) 90 (620) 22 K2A0.2/U0.9
CT90 90 (620) 98 (689) 22 K3A0.2/U0.9
CT100 100 (689) 110 (758) 28 K4A0.2/U0.9
CT110 110 (758) 120 (827) K5A0.2/U0.9
As length requirements rose, three things happened:
(a) Thicker strips were used at one end (and occasionally at
the other to provide for corrosion at the bottom of a
string).
(b) Strips that were continuously tapered were introduced
so that the skelp-end welds could be made gauge-to-
gauge.
(c) Grades, defined from the initial yield strength, rose
from 55 kpsi to 70 kpsi, then on up to 80, 90, 100, and
110 kpsi. Table 1 (an extension of data from API RP
5C7)1 shows the current situation.
Concurrent with the development of higher grades of work
strings, it was found that the older strings could be retired
for service as hang-off strings, i.e. used to lower the annular
area of the OCTG tubing in gas wells, and that CT could be
classed a coiled line pipe (CLP). CLP in sizes from 0.500
to 6 5/8 has now been accepted by API under a specification
published in 19992. Grades are shown in Table 2.
Table 2: Physicals for New Coiled Line Pipe
Grade Yield Str Tens Str TensStr Max Elong
(Min) (Min) (Max) HRC %
Kpsi/MPa Kpsi/MPa Kpsi/MPa
X52C 52(359) 66 459) 110(758)
X56C 56(386) 71(489) 110(758)
X60C 60(414) 75(517) 110(758)
X65C 65(448) 77(531) 110(758)
X70C 70(483) 80(551) 110(758)
X80C 80(551) 88(606) 120(827)
Not
stated
but kept
below
HRC22
Computed
from
625000 x
A 0.2/U 0.9
There is currently no API specification for coiled tubing,
and API RP 5C7 is due to be updated.
Tolerances of Coiled Tubing Wall Thickness
CT is wrapped around a drum at the end of the production
line, hydrotested to at least 1.6Sytmin/D, and if necessary
spooled back down to remove manufacturing imperfections.
The API minimum and maximum wall thicknesses are given
Table 3 - Wall Thickness tolerances
Specified Wall tmin t max
t < 0.110 in. t - 0.005 in. t + 0.010 in.
t > 0.110 in t - 0.008 in. t + 0.012 in.
in table 3. The tolerances cover the possible changes in
wall thickness that occur during strip rolling and make-up
during manufacturing of tubing from strip. Removal of
imperfections, detected by eddy current inspection during
production, is permitted to remaining walls of 0.95t and
0.875t. When this occurs, the flaw-removal area should be
smoothed, leaving no transverse sharp-bottomed marks. In
the case of 0.875t, API requires agreement with and
notification of the customer.
In-service Use
Tubing is wrapped on a working reel, with the heavy end on
the bed-wrap. In service, it is wound over a guide-arch,
inserted through a blow-out preventer (BOP) stack into a
well and reeled, often under pressure, into the well until the
lower end reaches the working area. Numerous types of
service occur: nitrogen injection, acid injection, hammering,
drilling (without tube rotation) through bridge plugs,
drilling side-wells from an existing well into a new
production zone, and electric-submersible pump
applications. In some cases, there may be a wireline or an
armoured copper cable pulled into the tubing.
Figure 1: Rig operations with coiled tubing.
In many situations, the CT is worked several times through
a zone; thus a section of it is wound repeatedly off of, and
back on to the reel. During such processes, the tubing
acquires fatigue because on the reel, it may be at 2-4%
strain (i.e. way beyond the elastic limit), and the process of
straightening and re-bending the tubing adds to the total
fatigue. This fatigue may be estimated using certain
theoretical fatigue life (TFL) models.
Development of Fatigue Models for Tubing
Models3 have been developed which use the slow strain rate
parameters that can be measured from controlled cycling of
new tubing samples. These quantities are derived from fits
to the stress-strain curves in the plastic region, as
determined from the Manson-Coffin relation. They predict
that the accumulated fatigue in tubing depends upon the
following:
(a) Outer diameter (D) and wall thickness (t).
(b) Grade, as expressed by the strain rate parameters.
(c) Internal pressure.
Development of Fatigue Models for Welds in Tubing
An industry project developed de-rating procedures for each
type of weld found in CT. These are:
(a) Gauge-to-gauge skelp-end welds.
(b) Step taper gauge skelp end welds.
(c) Tube-to-tube welds.
As can be imagined, experimentally testing a set of welds
would show a great deal of variance. Such welds had been
radiographed (RT) in strip and tube form, using a 2T hole in
a No. 10 penetrameter, and also the outer surface of the tube
samples was inspected by liquid penetrant (PT). These
methods are insensitive to (1) volumetric problems such as
(a) large grains, which can lead to brittleness, and (b) planar
lack of fusion, and (2) inner surface welding anomalies such
as (a) excessive weld beads and (b) surface cracks. This
experimental approach, however, lead to different amounts
of derating for the weld types listed above.
Figure 2: Measured Fatigue cycle life over a 48 inch
bend block for a 1.75" x 0.175" 80 grade tube, skelp-end
welds in the tube, and a 0.175-0.156 in tapered skelp
weld.
Figure 2 shows typical cycle lives (to failure) for CT-80
grade 1.75" tubing with and without skelp-end welds, and
with a 0.156-in. taper weld. Well-made skelp-end welds
cycle well, but skelp end welds made from one gauge to
another do not cycle well at all.
In-Service Problems
Users relying purely upon fatigue models began to pose the
question "why does the tubing fail at 20% of it rated life?"
and the answer is generally that at the locations in question,
it has failed at 100% of its life. One then asks the question "
how did this particular location get to 100% so quickly,
especially when adjacent regions exhibit considerable
remaining life?" The answer is usually that there is a
transversely oriented wall problem of some type that does
not cycle well4,5,6. The exception to this is the unfused weld
seam that comes open under the effects of internal
corrosion, pressure, and bending. Unfortunately, many
MEASUR ED FATIGUE CYCLES. 1.75" OD X 0.175
TBG W ITH 0.175-0.156 SKELP W ELD
0
100
200
300
400
500
600
TUBE BIAS W ELD TAP ER W ELD
CYCLES TO FA ILURE
events can occur to damage the tubing, (even on the first
job) and provide regions at which the model just does not
make the correct prediction.
In order to systematically approach the problem of
returned failures, an in-depth study of all failures was
performed, including those already completed by
metallurgical labs. The results are summarized in Fig. 3
using a simple classification scheme that had to include the
potential for NDE7,8,9. Problems with tubing have now been
generally determined to include the following:
Figure 3: Classification of CT Failures over a 9-year
period
(a) Undetected internal corrosion, especially at the 6
o'clock position, i.e. corrosion in and around the edges
of the pool that forms under gravity when fluids drip
down the inside of the tubing wall. Since most tubing
contains an internal unremoved weld flash, and exhibits
variable ID, the passage of wiper pigs does not always
remove all the fluid. Surface tension also causes a
layer to stick to the inner surface, and thus in a
considerable volume of fluid can trickle down to the
low point. (A 20- ft diameter tube section with ID =
1.5" had 3600 in2 of surface area. If 0.001-in. were to
adhere to the ID under surface tension, than 3.6 in3 of
fluid is available.)
(b) Undetected corrosion on the bed-wrap from external
storage in coastal areas. The attack here tends to be
from salt-bearing rainfall, and the surface tension
which holds fluids between the adjacent wraps. Such
areas are almost constantly refreshed with water, and
remain at high humidity.
(c) General wall loss from letting the tube sit in the acids
commonly used in well-servicing operations, e.g. 15%
HCl.
(d) Corrosion pitting from letting the tube sit in the acids
commonly used in well-servicing operations, e.g. 15%
HCl.
(e) Corrosion pitting from wet CO2. Here the corrosion
typical of wet flowing CO2 in OCTG occurs.
(f) Outer surface damage, especially sharp-bottomed
transversely oriented, from spooling or other operations
which leads to rapid fatigue accumulation during
cycling. Such damage includes that from all parts of
the injector and BOP system that can contact the
tubing. Even ball bearings from injector chairs have
been found as tubing wall damage.
(g) Weld-line pinholes parting after only a few jobs,
(whether attacked by corrosive fluids or not),
(h) General wall loss from wear against other materials,
such as the OCTG or bare sides of the well. One
problem encountered here is when the tubing goes into
sinusoidal buckling, and the areas in contact with the
sides of the OCTG wear rapidly.
(i) Necking: Decrease of wall thickness due to excessive
tensile force.
(j) Diametrical growth, which occurs in areas of high
cycling at relatively high internal pressure.
(k) Ovality, which also occurs in areas of high cycling, put
at lower pressure.
(l) Skelp-end weld cracking. This has been found to
occur in brittle sections of this weld.
(m) Surface cracking under tension caused by exposure
to H2S.
(n) Poor Butt Welds. The quality of field butt-welds is
difficult to quantify by RT and PT alone.
(o) Elongated gouges on the outer surface caused by
dragging the tubing past some type of constriction in
the well or injector.
Figure 3 represents our attempt to classify known problems
into broad categories based upon their origin, in order to
determine cause and responsibility. The major categories
are further discussed below.
Corrosion
With sufficient data, and the problem of corrosion caused
by acidic hydrotest water eliminated by requiring that the
hydrotest fluid have a pH of 8-9 (i.e. raising it by adding
NaHCO3 and an inhibitor), corrosion caused by operations
arises from improper cleaning or inhibition of the tubing.
Washing out tubing with an alkaline solution to remove ions
such as Cl-, and then inhibiting the ID surface from attack
by O2 is clearly important. Further, drying the ID surface
with warm dry air and then flushing the air out with N2 and
capping the ends to stop O2 ingress represents a good
preventative measure.
In the case of attack from well servicing acids, use of acid
inhibitors is essential, but does not always appear to be
guaranteed to protect the tubing. For the outer surface,
COILED TUBING FAILURES 1987-98
0
10
20
30
40
50
60
70
80
90
Corrosion
Rig Operations
Manufacturing
OD wear
Tensile
Butt Weld
OCCURENCES
removing the dried well fluids and adding a relatively thick
coating of a corrosion resistor, especially to those wraps at
the top of a string that never seem to leave the reel, has been
found to be effective. Users often cut and discard a length
(e.g. 200-ft) from the bottom of a string after acid jobs, so
that the preserved section at the top eventually finds its way
in to service.
Finally, the effect of working tubing in high H2S is now
being studied, and with the high forces found in the tubing
walls, the tubing is found to experience transverse surface
cracking which may grow rapidly under cyclic loading.
The corrosion bar in figure 3 may then be split into
several sub-divisions in order to determine the relative
amounts of each contributing to failures.
Operationally Induced Surface Damage
Many failures occur when sharp-bottomed gouges are
cycled. In tests with transverse EDM notches of depth 5%
and 10% of the tube wall, it was shown that 90% of the
measured fatigue life could be lost (Fig. 4) by leaving such
small defects in the tubing10. However, by taking such
surface defects out, most of the fatigue life returns. For
these data, the removal area was contoured, as it would be
when removing imperfections from OCTG. The length of
the removal area is now the subject of study, and it appears
that in the high strain regime, removal of the root of the
imperfection is critical, but that the length of the removal
area may be relatively small in length along the tube.
Figure 4: Effect on CT life of a 10% transverse notch
removed and left in tubing. Vertical axis is cycles to
failure on a fatigue machine.
Such damage is caused when slip rams are closed on the
tubing.
Section Removal
Early attempts to NDE-inspect in-service CT have revealed
all of the above imperfection conditions. Two types are
section removal occur as follows:
(a) Defect condition requires analysis: Here 4 samples
for fatigue and 2 for tensile testing are needed, along
with the area required for metallurgy.
Figure 5: Samples needed for analysis
The 4 x 8-ft. samples provides for measurement of the
remaining cycle life on a fatigue machine, and when
used in comparison with the initial life model, provides
a good estimate of whether the tubing should be tube-
to-tube welded back together and returned to service.
The 2 x 4-ft. samples provide for 8 tensile samples, for
which the stress-strain curve can be compared with that
of the original curve for the material. All samples
provide the ability for visual inspection, measurement
of diameter, wall thickness, ovality, pit depth, material
removal, etc.
(b) Tubing Section needs to be removed: Worn and
heavily cycled sections of a tube may be removed and
replaced with a tube-to-tube welds, if it is advantageous
in cycle life to do so. In effect, is the life of a new
tube-to tube weld higher than that of the worst part of
the worn section? Currently, this can be investigated
from the TFL model, using the results of the tube-to-
tube weld cycling project.
In many cases corroded sections are removed. Typically, a
section from the top of the string that has suffered OD
corrosion, is removed, and replaced with a butt-weld that
might never come off the reel. Mid sections have been
removed or replaced. Lower sections are removed and
discarded.
Other Operational Damage
Tensile problems account for two sets of problems as
follows:
(a) Necking: Here the tubing is pulled at a force in excess
of the local tensile strength, which may be somewhat
variable along the string. This occurs when the string
gets stuck, and the risk is taken that the tubing still has
the original specified wall, and there has been no
damage. Attempting to pull at the original yield
strength in a low-wall region can cause section of the
tubing to be operated at much higher force, and here the
wall begins to neck down as the crystal lattice planes
slide over each other.
(b) Ballooning: Here the tube expands radially due the
high internal pressure and cycling.
FLAW REMOVAL- CYCLE DATA, 1.75 x 0.134
CT80
0
100
200
300
400
500
600
700
800
VIRGIN PIPE 10% REMOVED 10% NOTCH
Both conditions are extremely serious and warrant early
detection.
Manufacturing Defects
All out-of specification conditions in new tubing represent
defective conditions. These include the following:
(a) Strip Defects: Strip mill gouges and roll-ins have
always been a problem as they lower the wall thickness
below the specified value.
(b) Unfused or Improperly Fused Electric Weld11: Here
the weld is not totally fused, and the condition is
virtually impossible to detect on in-line manufacturing
NDT. In the case of cold (paste/pasty) welds, even
ultrasound may require special processes to detect the
condition. Most "penetrators" would also not be
detected by eddy current NDT, but would be found by
UT. Such defects may be only 1-5 mm long, and the
smaller ones may not break the ID or OD surface. The
opening they
(c) Unfused or Improperly Fused Skelp-End Weld:
Well made skelp-end welds in gauge-to-gauge material
should exhibit 85-90% of the cycle life of the tube
material. Cracks and unfused areas in skelp-end welds
have been the cause premature failure.
Industry Studies on Surface Defects
Continued studies12 of fatigue cycling of new tubing with
machined or impressed defects has revealed that it might be
possible to establish relationships between the fatigue life of
new tubing with that of tubing with simulated "defects" as
follows:
(a) The nature of the imperfection: i.e. whether there is
metal removed or not. This would distinguish between
gouge and pit-like imperfections, and rolled in material
such as nuts and injector bearings.
(b) The dimensions of the imperfection: i.e. the relative
importance of length, width, depth, remaining wall, and
the curvature of the root of the defect upon the
remaining fatigue life. This would have some impact
upon manufacturing NDE and "receiving NDE" in
determining the need to remove a surface imperfections
at either stage. It will also impact the quantities
measured during in-service inspection, and assist in
setting inspection requirements for de-rating in-service
strings.
(c) How the imperfection arose: i.e. in the case of wall
loss, is the imperfection caused rapidly, or by longer-
term corrosion which might also affect the surrounding
tube wall in some as-yet undetermined way. Typically,
what is the effect of passing acid, etc, on the tube wall?
(d) The effect of 'grade': i.e. how does the effect of the
various chemistries used by manufacturers to establish
'grade'? also affect both fatigue life of new perfect
material, and also material with surface imperfections.
Theoretical Life Models
Refinements to theoretical life models have proceeded as
follows:
(a) Tensile updates: Adjustment of existing models to
continued fatigue cycling of NEW tubing by the CT
manufacturers. This would change the constants used
in the models, but not accomplish much for the
problems associated with in-service CT, since it is not
known how the stress-stain changes with age, other
than to assume there is a 10% loss of yield strength
during the life of the tubing.
(b) Adjustment for uniform wall loss due to acid flow
through the tubing. It may well occur that uniform
flow of acid through the ID of a tube removes a
uniform layer of steel from the ID, but the real problem
is that of the effect of the acid flow on the steel. Again,
the drip-down problem presents a more serious case.
Experience shows that these adjustments do not work, but
rather that assessment of the tubing at certain intervals, is
effective in maintaining string life. No current life model
can cope with anything other than unflawed tubing.
(c) Adjustments for Surface Defects: A current JIP has
provided for an attempt to assess the effects of surface
defects. An earlier fatigue model has been adjusted in
order to provide for the loss of fatigue life caused by (i)
the loss of wall, and (ii) the topography of the defect in
the wall. This model is experimental.
The Need for NDE
All of the conditions described earlier can be determined by
NDE, and can therefore be accounted for. Standard NDE
methods are applicable for the following inspections:
(a) Skelp Inspections: Skelp manufacturers need to
provide defect-free strip. In reality, this does not
appear possible, and reject rates rise for strip suppliers
as more restrictions are placed on them. However, in
the case of strip destined for work-strings, equipment
exists for the strip to be 100% full body inspected.
(b) Strip-Edge Inspection: Strip should be inspected for
all possible surface and volumetric defects. Visual
methods do not work for internal imperfections. Strip
edge 'quality' is critical in determining the ability to
weld them together. Here cleanliness of the strip edges
is next to godliness.
(c) Skelp-End Weld Inspection: The nature of weld
defects is such that skelp-end welds should be
inspected by both RT and shear wave UT.
(d) Weld-Line Inspection: Ultrasound is the only ins-
pection method that should be considered for weld line
inspection if all defects are to be found. What is now
true for ERW OCTG is also true for CT and CLP.
(e) Receiving Inspection: It is customary in OCTG
tubular inspection to have levels of inspection that
include an inspection of the product after hydrostatic
test. This is based upon that fact that in some
circumstances, imperfections may grow during
hydrostatic testing from a situation in which they pass a
UT inspection before hydrotesting to one where they
fail the same test after hydrotesting.
(f) Final Inspection: In many cases, the "Final Ins-
pection" is the "Receiving Inspection" for the customer.
The purchaser must determine the need for a final
inspection at the plant.
(f) In-service Inspection: The uncertain degradation of in-
service CT provides many reasons to inspect the tubing.
Users want an assessment of the tubing that is in-line
with pulling the tubing from the well, and this has been
found to be quite difficult. Excellent NDE tools have
emerged for in-service CT inspection that are based
upon the need to find volumetric flaws. (i.e. the need to
find weld-line imperfections is not addressed in these
tools and in any event could not be distinguished by
them.
What NDE Can Provide For Coiled Tubing
It is clear from the above that conventional NDE can
provide substantial information that will lead to improved
quality, i.e. better and more accurate knowledge of the
tubular. For this reason, NDE devices have been designed
that will detect the following:
(a) Wall Thickness: We define "wall thickness" as the
wall that generally occurs in the absence of pitting and
other small imperfections. This is tmin, or the lower
values as stated above. In the case of newly-
manufactured CT this does not matter. In the case of
in-service CT, the ambient wall and defect wall signals
can be separated electronically. This is done because
both ultrasonic wall and magnetic wall measurements
scan for the back wall (UT) and average back wall.
(MR-magnetic reluctance method)
(b) Outer Diameter: Two outer diameters are defined. The
maximum (Dmax) and minimum (Dmin ) diameters at any
location can be measured and used to compute the local
increase/decease in diameter (necking/ovalling/bal-
looning) and ovality. This can be done by either the
eddy current stand-off method, or by sliding a contact
over the part that is connected to a LVDT.
(c) Detection of Pitting and Cracking: The detection of
pitting and transverse cracking is performed by
magnetic flux leakage inspection (MFL). Accurate
determination of defective conditions by MFL is not
available because of what can be quite large changes in
magnetic permeability around defects. Thus MFL will
determine the presence of imperfections, but one
should never rely on this technique for defect
assessment. Further, in the case of highly strained CT,
one would not wish to assume that results taken by
MFL on unstressed tubing would be valid for seriously
stressed tubing, especially that which was not only in a
state of serious strain, but which had been subjected to
several cycles of high tension and compression.
(d) Detection of Highly Cycled areas: In most cases, eddy
current scans for outer surface defects also can scan for
changes in AC permeability and electrical conductivity.
Electromagnetic NDE can therefore possibly provide
the locations of the onset of seriously fatigues areas.
This can be of great importance when it is considered
that the length of tubing constantly changes with the
working of the tubing.
(e) Measurement of Fatigue: Since eddy currents are
particularly responsive to the electrical conductivity (σ)
of a part, and the amount of fatigue can be taken as the
accumulated sub-microscopic damage to the crystal
lattice structure (when new), it may be that non-
destructive measurements of changes in σ provide a
way of assessing fatigue. This may be easier in the
case of non-ferromagnetic materials, but there is the
possibility that actual measurements of accumulated
fatigue may be possible for ferromagnetic materials,
and thereby supplement the results of TFL modelling.
(f) Seam Weld-line Inspection: Seam welds in CT and
CLP can be inspected with UT. One major problem is
the presence of the internal flash column, and a second
is the variation of tube wall thickness along a string of
CT. CLP is, of course easier to inspect, since the wall
remains constant, and for the larger sizes at least, the
internal flash column is removed.
(g) Tube-to-Tube (CT) and Pipe-to-Pipe (CLP)
Inspection: Modern methods need to be determined for
the inspection of such welds in these products in order
to reduce the reliance on radiography and liquid
penetrant inspection.
(h) Training and Discipline: Since good NDE should be
performed under a competent Level III, the training and
discipline involved with performing NDE at level I and
II assists in providing for quality control of tubing
manufacture. All CT inspectors should be trained and
certified under an ASNT type of scheme.
Summary
In this paper we have shown the kinds of problems
associated with manufacturing and in-service use of coiled
tubing. All of these problems lend themselves to some form
of assessment by NDE. In fact, NDE has been responsible
for retiring CT when needed, and keeping CT in service
when it was felt that there was nothing inherently wrong
with the tubing. The effects that are related to fatigue
cycling and relatively high internal pressures make the
assessment of in-service CT not only extremely interesting,
for also a challenge.
The following paper discusses the current state of NDE
for coiled tubing.
Nomenclature
ASNT American Society for Nondestructive Testing.
ASTM American Society for Testing and Materials
API American Petroleum Institute.
CLP Coiled Line Pipe
CT Coiled Tubing
EDM Electro-discharge Machine
EMI Electromagnetic Inspection (Oilfield usage for
longitudinal, transverse flaw inspection, wall
thickness measurement, and eddy current grade
comparison )
ET Eddy Current Testing
ETCE Energy Technology Conference and Exposition
ID Inside Diameter
JIP Joint Industry Project
MFL Magnetic Flux Leakage
MT Magnetic Particle Testing
NCT New Coiled Tubing
NDT/E Nondestructive Testing/Evaluation
OCTG Oil Country Tubular Goods
OD Outer diameter
PT Liquid Penetrant Testing
RT Radiographic Testing
TFL Theoretical Fatigue Life
UCT Used Coiled Tubing
DOuter Diameter of tube
tSpecified wall thickness
tplate Specified thickness of plate (strip)
tav Average wall thickness of tube
tmin Minimum specified wall in unflawed areas
SySpecified minimum yield strength
σElectrical conductivity
References
1. API Recommended Practice RP5C7 "Recommended
Practice for Coiled Tubing Operations in Oil and Gas
Well Services," 1St Ed, Jan 1997.
2. "Specification for Coiled Line Pipe," American
Petroleum Specification 5LCP, pub. API December
1999.
3. "Multiaxial Plasticity and Fatigue Life Prediction in
Coiled Tubing," S. M. Tipton, in Fatigue Lifetime
Predictive Techniques: 3rd Volume, ASTM STP 1292,
pp. 283-304, 1996.
4. “Defects in Coiled Tubing,” Proc. 4th World Oil Conf.
On Coiled Tubing, Feb 1997, pub. by Gulf Publishing,
Houston.
5. “Failures in Coiled Tubing," R. K. Stanley, 5th Intl. Conf. On
Coiled Tubing and Well Intervention, Feb 1997, pub. Gulf
Publishing.
6. “An Analysis of Failures in Coiled Tubing,” R. K. Stanley,
IADC/SPE Paper 39352, Proc. IADC/SPE Conf, Dallas, TX,
Mar 1998.
7. “Nondestructive Inspection Tools for Coiled Tubing,” R. K.
Stanley and K. Varner, Proc. Brit. Inst. Of NDT Conf. Sept
1998.
8. "Testing of Coiled Oilfield Tubing - an Update," Roderic K.
Stanley, Materials Evaluation 58(8), pp. 970-975, August
2000.
9. "Coiled Tubing Failure Statistics Used to Develop Tubing
Performance Indicators," Willem van Adrichem, Paper
54478, Proc SPE Con. On Coiled Tubing, Houston May
1999.
10 "Results of Recent Inspections Performed on Coiled Tubing,"
Roderic K Stanley, SPE Paper 54484, Proc. SPE/Icota Conf,
Houston, TX, May 1999.
11 API Bulletin 5T1, "Imperfection Technology."
12 Joint Industry Project on Surface Defects, University of Tulsa
Second Year Report, 2000.
Notes: paper given at Second Pan American Conf on NDE,
Houston, Tx June 2001.
Article
Full-text available
The testing of coiled oilfield tubing is considered to be an important requirement prior to and during offshore servicing. One interesting method is that of wall thickness measurement of ferromagnetic tubes using a noncontact direct current magnetic technique. This paper covers some results obtained with this for carbon steel coiled tubing.
Article
Coiled tubing is being used increasingly in the oil well drilling and servicing industry. Continuous steel tubing of structural dimensions (up to 89 mm or 3.5 in. in diameter) is wound onto a large-diameter reel for repeated deployment into and out of a well bore. The bending strain range associated with each wrap-unwrap cycle can exceed 3% with lives well below 100 cycles. During constant internal pressure fatigue testing, tubing has been observed to grow in diameter by as much as 30%. This paper describes an analytical model to predict the fatigue behavior of coiled tubing subjected to variable pressure service conditions. The approach utilizes standard low-cycle fatigue data but requires additional experimental results from constant pressure fatigue testing. The algorithm is based on estimates of biaxial ratcheting from an incremental plasticity model using a hybrid associated flow rule, a modified kinematic hardening rule with multiple von Mises yield surfaces, and a specialized limit surface concept. An empirical damage parameter was formulated based on constant pressure fatigue data using mean and fluctuating von Mises equivalent strain components occurring throughout the life of a section of tubing. This parameter is used with the Palmgren-Miner definition of cumulative damage to track damage that is accumulating nonlinearly under constant or variable pressure histories. Modifications to standard incremental plasticity components and implementation assumptions used to apply the model are presented and discussed. The predictive capability of the model is demonstrated relative to data generated under constant and variable pressure histories.
Article
This paper outlines and discusses failures in carbon steel coiled tubing (CSCT), along with their causes, and proposes methods for de-rating used coiled tubing that might be applicable to an industry recommended practice. P. 515
Recommended Practice for Coiled Tubing Operations in Oil and Gas Well Services
API Recommended Practice RP5C7 "Recommended Practice for Coiled Tubing Operations in Oil and Gas Well Services," 1 St Ed, Jan 1997. 2. "Specification for Coiled Line Pipe," American Petroleum Specification 5LCP, pub. API December 1999.
Nondestructive Inspection Tools for Coiled Tubing
  • R K Stanley
  • K Varner
"Nondestructive Inspection Tools for Coiled Tubing," R. K. Stanley and K. Varner, Proc. Brit. Inst. Of NDT Conf. Sept 1998.
  • Stanley
Stanley, Materials Evaluation 58(8), pp. 970-975, August 2000.
Coiled Tubing Failure Statistics Used to Develop Tubing Performance Indicators
"Coiled Tubing Failure Statistics Used to Develop Tubing Performance Indicators," Willem van Adrichem, Paper 54478, Proc SPE Con. On Coiled Tubing, Houston May 1999.