WHAT’S IN

YOUR COILED
TUBING?

Roderic K. Stanley, Ph D, I. Eng

Rkstanley@ndeic.com

ItRobotics, Inc., Stafford, TX International Oilfield Services

Coiled Tubing Resources Management

Icota Lunch and Learn, February 8, 2011

 A New Coiled Tubing
Assessment System
• Coiled tubular assessment systems are available from
itRobotics. Service available from Int’l Oilfield Services
• The system measures wall parameters and detects
imperfections.
• The inspection head of the system is a “clamshell”, so that
only certain regions of a CT reel can be inspected if needed.
• The effect of OD surface imperfections can be assessed by
the use of the Flexor TU06 fatigue model from the U of Tulsa.
They can then be removed as needed.
• The system meets the requirements of SR 37 in API 5ST
(Specification for Coiled Tubing) for “inspection after
hydrostatic test”. Thus it can be used to set baseline data at
a CT mill.
• The system can be used in the field, so that the degradation
of a string during its lifetime can be assessed.

ItRobotics, Inc
12841 Capricorn Street
Stafford, Texas 77477-3912
281-313-7900, 281-313-7901 fx Not Zone 1 yet, but
electronics are potted
 Head, Closed Nd-Fe-B
• Length counter wheel
magnets
• Permanent Nd-B-Fe magnets provide the
saturated longitudinal field induction in the
tubing.
• 4 “shoes” in stainless steel housings
contain multiple hall-effect sensors ring Signal
the tbg. cable
attach
• The tangential (Bx) and normal (Br)
external field strength is measured close to
the tubing surface to detect imperfections
and measure wall thickness.
• An eddy current lift-off technique (8
sensors) is used for diameter
measurement and ovality computation.
• The tube is longitudinally magnetized.
• If butt welds are needed, the 2 ends should
be demagnetized.
• Sizes 1.25, 1.50. 1.75, 1.00, 2.375.
• Size change is easy.
The tube diameter is
measured at 8 places, and
Head Dimensions
the ovality of the tube is
computed from these
readings.
Θ = Dmax – Dmin x 200
Dmax + Dmin
Rings of hall-effect sensors
between the poles of the
magnets measure the wall
thickness and detect
imperfections and defects.
The system can back over
suspected signals and
over-write the inspection
record as a check that
signals are real.
 Performing
Standardization
• Here the unit is being standardized on
a reference standard that meets the
ExxonMobil & API 5ST requirements.
• The standard contains 10% LID, LOD,
TID and TOD EDM reference notches,
1/32nd in. TDHs, & wall thickness
reduction.

MFL from OD flaw

MFL from ID flaw

 Computer Screen, 1.75 x 0.134-in. tbg

Speed
Weldline MFL

Notepad

360
degree
map
Wall loss Butt
1/32 TID TOD
Area Weld
Variable TDH
Lamps light when
Alarm MFL signal in octant,
Threshold LOD Max, Min and average make imperfections
LID
s wall plotted easy to find during
prove-up

Max, Min and

average OD plotted

Expandable Length Counter- this one is set for 25 ft Scales show tube dimensions

• As part of prove-up, any area of this “standard” screen can be expanded so as to “focus in” on each of the
signals. The vibration sensor also helps discriminate real signals from tubing vibrations. And the data can be
over-written to prove it was real.
 Ovality, Diameter, & 4 runs on a
Reference Standard
• This simulation shows OD, wall and
ovality at any location down the tube,
in real time.
• When the red button is pushed, the
diameter and wall thickness
measurements appear.
• X- and Y- axes show measured
dimensions.
• Wall thickness is the value computed
from the tangential hall effect signal
readings. (There is no contact with
the tubing, so none of the problems
that are associated with ultrasonics).
• Data are stored every 0.5 inch because
diameter and ovality do not change very
rapidly.
• Reference standard is rotated 90°
between runs so as to verify the
uniformity of sensitivity.
• This is commonly done in OCTG
inspection standardization.
 Length of flaw can be measured

Zoomed Signals from all

Digitized (every 0.5

MFL
mm) MFL signals
4 “shoes”

signals 1
• Areas of the maps can
be zoomed for more
detail, and possible
signal recognition. Transverse notch
Longitudinal
• This is done by stop- notch
ping the inspection, or
later on, depending Rotatable
upon where the ins- 3D MFL
pection is performed. map
• This is because this
part of the programme
takes a short while to
load

• By rotating the 3D map, the peaks of the longitudinal EDM

notch measure its length.
• There is MFL from the ends of LOD and LID notches.
• Observe the difference in signal shape between the longitudinal
and transverse notches.
Zoomed MFL 2 - Longitudinal EDM Notch
0.020-in. wide (API 5ST SR37)
• There is only MFL from the beginning
and ends of EDM notches
• There is MFL in at least 3 sensors.
• Note the regularity in the magnetic Ends of EDM
notch
noise all the way around tube. (I am not
sure what caused this; maybe magnetic
permeability variations.)

Gauss

• End signals indicate the length of the

Ends of 0.50-in LOD-notch notch, in this case 0.5-in.
• Recall that the magnetic field is
parallel to the notch but we still have
a 40 G swing peak-to-peak.
• 3 sensors (blue green red), have
detected MFL from LOD notch (upper
degrees trace).
1/32nd Through-drilled Hole

Hole is picked up on more than 1 sensor

One or more of these

will show in red when
a MFL signal passes
through that octant,
making internal flaws
easy to locate
Peak-to-peak
amplitude is 80 G

Window length is 15 ft
 Butt Weld
• Location of butt Typical Butt Weld
welds is critically MFL signal

important in used Very large 500

tubing Gauss signal

• These make
excellent markers.

Variability of MFL
around the weld
• The BW is picked up all the way
around the pipe.
• Its 4-peaked structure is typical of
these welds
• This MFL signal shows that a trans-
verse crack in a BW might be hard to
find, and they need to be reinspected
with Ultra-sound for cracks . (UT is
single wall, RT is double –wall)
 Signals Taken on New Tubing
• The following
signals were
collected on
new coiled
tubing.
 Roll-in Near Seam Weld

Signals from 75 sensors on a seam weld inspection unit with the flux density, ( B )at 45 deg to
the longitudinal seam weld and to the axis of the tube.
Skelp End Weld 19350-ft. along Tube

Seam weld

End

Seam Weld

Start

Recognizable by variable MFL at 45 deg around tube

 Variable Long Seam Weld Noise
• Data from
12,600-ft of a
string
• High seam
weld MFL
noise from
internal flash
column on
MFL
several strips. from Δt
• One step-
taper visible.
• OD is show-
ing a little
ovality in the
last 2 (thinner-
wall) strips.
 One Source of MFL Noise
• A 3D look at one indication
at 1298-ft. in this string,
has a peak-to-peak
amplitude of 140 Gauss.
• This is part of the prove-
up procedure and aids the
observer in flaw
recognition.
• The cause was a piece of
“spume”, found by mill
RT.

Shape of signal indicates whether MFL comes from

excess metal or a pit.
 New Tubing Stored for 3 Years.
Almost full length round-bottomed pitting
was found on the OD surface of a new
string that had been stored in Houston for
3 yrs and indicates the need for a good
maintenance programme since current
protective coatings do not appear to last
long. The pits have a maximum depth of
0.018-in. (8.5% of measured wall, 0.204-
in).

OD surface pitting in new tubing at 13,085-ft.

The 3-D picture enables determination of

location and thus determines where to scan
with ultrasound or radiography if they were to
originate from mid-wall or ID imperfections.
MFL Image of a row of pits in the OD of stored pipe.
 Analysis of Deepest Pit
• To assess the possible effect of : Flexor TU 6.1 calculations. New CT
this OD pitting, Flexor TU was Pit
used to generate the trips to
crack initiation and failure, and Trips to… none present removed
the 95% confidence level in the
initiation calculation for this Initiation of crack 187 176 181
grade (CT110) and size, an Failure 222 209 215
internal pressure of 3 kpsi, and a
42-in. radius reel (See Table ). 95% confidence 151 88 119
• The pits were not removed, and
this small wall loss was not
considered by the owner to be
sufficient to seriously affect the
fatigue performance of the tubing,
which is in line with the results of
research performed at the U of
Tulsa. Measuring the dimensions
of the worst pits and using them
in Flexor indicates the difference
in calculated fatigue life that
these flaws can create.
• The wall thickness data for this string are shown with the vertical scale expanded. There are 3
continuously tapered sections, and the skelp-end welds can be seen because of their MFL.
This weld MFL provides markers for determining length changes.
 Chafing Marks and Gouges
• Chafing Marks: Larger and heavier
CT can be chafed against the sides of
the storage reel, as shown here.
• This one was 0.013-in. deep and
removed by mill staff prior to shipping
the string, with the final tubing wall
thickness meeting API wall thickness
tolerance.

• A second example of multiple chafing

marks.

• It has now become standard practice to

remove such shallow marks by buffing,
since removal maximizes the remaining
fatigue life at that point.
 Signals from Used Tubing
• The notched reference standards are
used to set up (standardize) the
instrument on NEW tubing at the mill, or
as “receiving inspection” at a client’s yard.
• When inspecting USED tubing we would
normally set up on a 1/16th-in. TDH or
partial hole, and a 10% wall loss area.
• We can classify various sections of tubing
according our own or customer
requirements.
 First, Why We Did Not Use UT
• UT requires very clean smooth entry & exit
surfaces, so serious cleaning is required.
• Surfaces of used tubing are generally
covered with rust or other materials that do
not transmit ultrasound.
• Even the small amount of rust shown also
affects the eddy current lift off sensor
response.

• Recall that there

is always the
potential that the
rust may contain
naturally
occurring
radioactive
scales (NORM).
NORM).
 Results of Field Acid Job
• This 1.25-in. string has done an acid job in
LA, & lost some wall thickness at the bottom
end. We took out some OD gouges to about
10% t so as to raise the fatigue life. We
recommended that the bottom 2000-ft be
removed since wall had fallen from 0.156-in.
to 0.122-0.125-in.

Three views of imperfection MFL signal found near to seam weld at 12500-ft. and removed by
sanding.
 OD Gouges at 11,630 ft & 12.468-ft
• Gouge close to seam weld.
• These gouges may be caused going
into the hole.
• A very small scratch all the way
around the tube.
• Note that seam weld noise is very low
here

A 2nd gouge near to the seam

weld.
The seam weld noise appears to
come from regularly- spaced
small spume mushrooms on the
ID.
 2.375-in. CT100 string
• This used string showed surface
gouges of maximum depth 0.015-in.

Longitudinal rows of pits were

also found.

Ovality as much as 8% was

measured on this string.
 Dents in Same String
• String also showed dents that
were caused by pressure
against spokes in service
reel.

• MFL from dent

 Bias Weld MFL Signal in Field
• Bias and butt welds make
good markers, and can be
used to determine real
lengths and local stretch.
• The first skelp-end (bias)
weld off the reel makes a
good origin of length
coordinates
• This magnetic noise from
these welds was measured
on the previous job
• Often there is more magnetic
noise nearer to the 45º
corners of these SE welds.
 Signal Prove Up
• Finally, signals are proved up.
• Conventional dimensional (caliper, UT Compression
wave, ruler, pit gauge) measurements for OD
surface imperfections.
• UTCW for ID pitting
• UT Shear Wave for 2-Dimensional (crack-like)
Compression wave unit
defects
• And, of course, the expanded 3D MFL signals
themselves.
• When assessing new tubing, the mill can perform
Radiography (RT) if needed.
• All of these prove-up techniques have been laid out
in earlier SPE and Icota publications, and are
Phased array shear wave
standard procedures for assessment of tubing
 Imperfection Assessment
• We have given a lot of thought to what an imperfection does to
tubing when it is bend-cycled and is accumulating fatigue.
• Some CT fatigue models exist that do not take into account the
effect of imperfections and defects.

So we have
added Flexor TU
(from the
University of
Tulsa) into the
laptop. This
programme
contains the
detrimental
effect of
imperfections
and defects, so
ID cracks – black
that quantitative
light decisions
regarding
removal can then
be made.
 Imperfection Removal for CT
• We can assess the severity of OD surface imperfections using Flexor
TU06.
• We can remove the imperfections so as to leave the minimum specified
wall (API, Client) in tact (new tubg) or help to raise the remaining life
(used tubg).
• We can perform compression wave wall measurements to check
remaining wall thickness after flaw removal.
• We can also perform shear wave inspection on bias, butt and seam
welds.
• We provide an inspection report that can be used as the basis for future
inspections.
 Conclusions
• A CT inspection and assessment system is available from
itRobotics, and in service at IOS-PCI.
• The system measures wall thickness and OD, computes ovality,
and detects imperfections and defects at specific locations on
tubing.
• The system also detects welds, which act as excellent markers
• Some prove-up can be performed using the MFL images, and
some will be done when the tool is stopped for a second look at
the tbg.
• ….and this is just the beginning!
• This MFL programming is retrofittable to conventional OCTG
MFL units