SPE 36346

SOClety of Petrolun Engneers

Gulf Coast Section
KCCMTA
International Coiled Tubing Association
Analysis of Coiled Tubing Welding Techniques
K.R. Newman, SPE, P. A. Brown, SPE, W.O. Van Arnam, SPE, CTES, L.C. and S. Wolhart, SPE, GRI
Copyright 1996, Society of Petroleum Engineers, Inc.
This paper was prepared for presentation at the SPElICoTA Roundtable held in
Montgomery, Texas. 26-28 February 1996.
This paper was selected for presentation by an SPE Program Committee following
review of information contained in an abstract submrtted by the author(s). Contents of
the paper. as presented. have not been reviewed by the Society of Petroleum
Engineers and are subject to correction by the author(s). The matenal, as presented.
does not necessarily reflect any posrtion of the Society of Petroleum Engineers. its
officers. or members. Papers presented at SPE meetings are subject to publication
review by Editorial Committees of the Society of Petroleum Engineers. Permission to
copy is restricted to an abstract of not more than 300 words. Illustrations may not be
copied. The abstract should contain conspicuous acknowledgment of where and by
whom the paper was presented. Write Librarian. SPE. P.O. Box 833836. Richardson.
TX 75083-3836. U.S.A.. fax 01-972-952-9435
Abstract
As the use of coiled tubing (CT) in workover services, drilling
services, coiled completions and coiled pipeline increases, the
need for improved field welding techniques increases. This
paper summarizes the results of a weld analysis research
project in which failed welds were analyzed, welding
procedures were reviewed and alternative welding techniques
were considered.
Introduction
A significant amount of research has been performed in recent
years to improve the understanding of the CT pipe life. Pipe
life prediction models are now being used to predict when the
pipe is reaching the end of it's fatigue life. This
understanding has improved the reliability of CT services.
The weakest points in a CT string are at the welds.
Significant improvements have been made by the CT
manufacturers to the welds made during the manufacturing
process. Instead of welding tube-to-tube or "butt" welds on
completed sections of CT pipe, the manufacturers are now
welding the sections of strip material together before it is
milled into a tube. The ends of the strips are cut at an angle or
a "bias" for this type of weld, thus these welds are commonly
known as "bias" or "CM" (continuously milled) welds. The
bias causes the weld to be distributed helically along the axis
of the tubing, and as a result increases the axial strength of the
weld. This strip welding technique has significantly improved
the reliability ofCT pipe.
In the field it is often necessary to join two sections of CT.
Obviously the strip welding technique cannot be used because
the CT is already a tube. Thus, butt welds must be used.
These butt welds are the most unreliable portion of the CT
string.
A research project was performed for the Gas Research
Institute (GRJ) to better understand CT welds. This GRl
project involved three major efforts:
Weld Analysis - A previous joint industry project (JIP)
had been performed by CTES in which nearly 400 CT
fatigue tests were performed to better understand the
fatigue life of CT welds. Many of the failed samples
from the JIP were made available to this GRJ project for
analysis.
Weld Procedures - Welding procedures for CT from
various sources were reviewed along with the complete
welding process. Recommended procedures were written
for four different welding scenarios.
Alternative Techniques - Ten alternative welding
techniques were examined to determine if they could be
applied to CT welding.
The results of this research are available through GRl in a
full report (references 1 and 2). This paper summarizes these
results. An ICoT A standards committee is developing an
industry standard for CT field welding which will incorporate
results from this research.
Weld Analysis
Three sets of experts were used to perform this analysis:
UT (The University of Tulsa) analyzed CT bias weld
samples and welds removed from the test at 50% and
75% of the expected life, to better understand crack
initiation and growth.
A TS (Acute Technology Services) investigated manual
and orbital weld samples to compare the radiographic
evaluation performed on the welds at the time they were
made with the fatigue testing results. A TS also
performed a metallographic and micro-hardness analysis
of several failed welds.
2 ANALYSIS OF COILED TUBING WELDING TECHMQUES SPE 36346
CTES expanded the analysis performed by UT and A TS
by performing failure analysis on very good and very
poor performing weld samples. The purpose of this
analysis was to try and categorize which causes of failure
were the most detrimental.
The UT analysis looked mainly at the cause of primary
and secondary fatigue cracking in CT. A heavy emphasis was
placed on the bias welded specimens. The investigation
included optical metallography, SEM fractography and
microhardness data. The major conclusions from this UT
analysis were:
Longitudinal seam weld flash is the major fatigue crack
initiation site in bias welded CT.
Grinding marks perpendicular to the applied fatigue stress
are origination sites for numerous primary and secondary
fatigue cracks.
Jigs and fixtures used to prepare and align the weld joint
are capable of contributing to reduction in fatigue life.
Fatigue cracks were usually found to start on the inside
surface of the CT. These cracks cannot be detected with
techniques which examine only the outside surface, until
after the crack has propagated through the wall and a
failure has occurred.
The A TS analysis concentrated on manual and orbital
welds, looking primarily at radiographic interpretation and its
ability to identify weld fatigue failures. The analysis included
optical metallography and microhardness traverse testing.
Contrasts between the current industry radiographic
interpretation with an accepted industry standard ANSI B31.3
were drawn. The major conclusions from this A TS analysis..
were:
Defects identified by radiography in many instances had
no relative relationship to actual fatigue performance.
Current CT industry practice for radiographic
interpretation is far more severe than many industry
standards used in related industries.
The location of many primary fatigue cracks in manual
and orbital welds coincides with the front of the chill
blocks.
The CTES analysis combined the results from the UT and
A TS analysis with the analysis of additional welds. This
analysis included optical metallography together with both
optical and SEM fractography. The major findings by CTES
included the following.
Identification of causes of primary fatigue crack initiation
ranked from most influence to least influence as follows:
1. Insufficient material due to bum through, suck
back or excessive grinding.
2. Planar weld defects including lack of penetration
and lack of fusion.
3. Major rounded defects including gross porosity.
4. Severe grinding marks
5. Cap and root pass reentrant angles
6. Internal Flash Removal Transition
7. Fine Grinding Marks
8. Corrosion pit (few used CT samples were
included in this study)
Current radiography practice identified discontinuities
that did not lead to fatigue failures and failed to identify
some that did lead to failure.
The method for dressing the longitudinal seam weld and
the shape of the transition to non removed flash was
important to the fatigue life of the weld.
Welding Procedures
The number and variety of manual welding procedures
currently used on CT was found to be large. Whereas the
industry had a belief that all welding procedures were similar
in their requirements, this study discovered a wide variation in
application of the GT A W (gas tungsten arc welding) process.
In some cases, selection of parameters or welding variables
considered critical to weld integrity was left to the welder
performing the task.
Five manual welding procedures from CT manufacturers
and service companies were reviewed in this project.
Numerous experts were interviewed for their input. Based on
this input and analysis of the welding process, three welding
procedures for manual welding, and one welding procedure
for orbital welding were developed (see Appendix). The
terminology for the weld geometry is defined in Figures 1-3.
The autogenous welding procedure is recommended for
CT with thin wall thicknesses. This procedure requires a very
high skill level for manual welders and high helium gas
mixtures.
The high penetration V -groove welding procedure is
recommended for thicker CT. Again, a high skill level and
high helium gas mixture is required.
The low penetration welding procedure requires the least
welder skill to perform and only argon gas. This will
probably be the best suited procedure for use in the field.
The orbital welding procedure is recommended if an
orbital welder is available.
Alternative Techniques
The practice for butt welding CT utilizes gas tungsten arc
welding (GT A W) in either the manual or automatic operating
mode. The ability of the GT A W process appears to be limited
SPE 36346 K.R. NEWMAN, P.A. BROWN, W.O.VAN ARNAM, S. WOLHART 3
--------------------------------------------------------------------------------
in producing welds with fatigue lives in the range of plain
pipe. As the industry trend toward larger diameter and thicker
walls continues, the time and cost of GT A W will be less
favorable. The following is a list of ten welding processes that
were considered for welding CT. The processes are !inted in
order of assessed capability for CT welding, with the most
capable listed first. Factors considered in the ranking include
the ability to make sound welds, expected mechanical and
fatigue properties, ease of completing reliable welds,
applicability for service company district application and
relative cost.
I. Plasma Arc Welding (PAW) - Higher welding speeds at
lower currents. High integrity welds with fewer passes.
Very skilled welder or automatic equipment required.
2. Laser Welding Minimum heat input reduces
metallurgical effects in heat affected zone. Joint must be
precisely positioned. Equipment very expensive and not
developed for CT.
3. Homopolar Welding - High speed, minimum heat, no
filler metal weld. Should be very good for CT. No
standard equipment exists. Current equipment costs are
very high.
4. Amorphous Bonding - Very high strength weld made at
low temperatures. Should fatigue very well. Currently an
experimental process. Further development required
before commercialization.
5. Friction Welding (FRW) and Inertia Welding - Would
provide good weld quality. Requires the CT on one side
to be rotated.
6. Upset Welding (UW) - Quick, uncomplicated fusion
welding technique. Few reliable nondestructive tests of
weld quality.
7. Gas Metal Arc Welding (GMA W) and Flux Cored Arc
Welding (FCA W) - Requires less skill and is faster to
perform than GT A W. Poor quality weld deposit for CT
in horizontal position.
8. Thermite Welding (TW) - Compact, inexpensive and fast
process. Metal properties difficult to predict. Internal
plug required in CT. May have potential for quick field
welds to allow completion of the current task. It would
have to be replaced later with a better weld.
9. Flash Welding (FW) - Not good for tapered joints. Low
yield strength in weld material.
10. Electron Beam Welding (EBW) - Very good for CT
welding but very expensive equipment.
Each of these processes is discussed in more detail in
reference 1. Further work is proposed to test some of these
weld processes on CT butt welds.
Conclusions
There is significant room for improvement in CT butt welds,
botr, :n the application of the current GT A W technique and
possibly in the application of other welding techniques.
Current radiographic procedures for determining CT butt weld
quality are not adequate in determining potential for fatigue
failures. Grinding marks created in preparation for welding
are a source of fatigue crack initiation.
Acknowledgments
The authors would like to thank the Gas Research Institute
(GRI) for funding this work. We are also grateful to Professor
Steve Tipton of the University of Tulsa and A TS for their
contribution to this work. We are also grateful to the
participants in the Weld JIP for providing the weld samples
analyzed in this study. The participants in the Weld JIP were
Agip, Amoco, BP, Exxon, Halliburton, Nowsco, Precision
Tube Technology, Baker Hughes INTEQ, Schlumberger
Dowell, Shell, Southwestern Pipe, and Statoil.
References
I. Newman, K., Brown, P., Van Arnam, D.:" Slimhole and Coiled
Tubing Standards, Phase I - Weld Technology," GRI Final
Report number GRI-95/0500.1, December 1995
2. Newman, K., Brown. P .. Van Arnam, D.:" Slimhole and Coiled
Tubing Standards, Phase I - Weld Technology," Supplemental
Documents, GRI-95/0500.2, December 1995
4 K.R.Newman, P.A.Brown, D.V.Arnam,S.Wolhart
Figure 1- Square Butt Joint or Square Groove Weld
Root Face and
Groove Face
Weld Face
Weld Toe
Root Gap
zero gap typical Root Surface
Centerline of CT
Figure 2 - V-groove Weld With Land
Angle /
\
Groove __________ . /
I
_ Angle
\\ I / Depth of
Face Reinforcement
\ I / Bevel
I /
I J---."-------.
Groove Face
Root Gap
1/16
th
inch
typical
or Land
Figure 3 - V-groove Weld Without Land
Feather
Edge
Root Gap
1/16
th
inch
typical
Reintnr"Q,rne,n
Centerline of CT
Centerline of CT
SPE 33346
SPE 36346 K.R. :">'EWMA:">', P.A. BROWI'I, W.D.VAN ARNAM, S. WOLHART 5
Appendix
Autogenous Low Penetration High Penetration Orbital
WELDING PROCESS GTAW GTAW GTAW GTAW
Type Manual Manual Manual Automatic
1.0 JOINT
1.1 Groove Design
Joint Design Used Square Butt Single V-Groove Single V-Groove Single V-Groove (3)
Included Angle None 40/20per side 60-90
0
/30-45per side 60
0
2 1/2
G
Root Opening None 0.0625" max 0.0625" max 0"
Root Face Dimension 0-0.134" 0 0-0.020" 1 116" or Less
1.2 Backing, Retainers, or Chills
Backing or Retainer None None None None
Copper or Brass Chills On Each Side On Each Side On each Side On Each Side
2.0 BASE METALS
2.1 Base Metal Coiled Tubing Coiled Tubing Coiled Tubing Coiled Tubing
Material Specification ASTM A606 ASTM A606 ASTM A606 ASTM A606
Type or Grade Grade 4 Modified Grade 4 Modified Grade 4 Modified Grade 4 Modified
2.2 Groove Joint Thickness
Base Metal 0.0625" - 0.134" 0.0625" - 0.134" 0.0625" - 0.134" 0.0625" - 0.134"
Deposited Weld Metal 0.0625" - 0.134" 0.0625" - 0.134" 0.0625" - 0.134" 0.0625" - 0.134"
Tubing Diameter Range 1" & over 1" & over 1" & over 1" & over
3.0 FILLER METAL
3.1 Filler Metal
Specification Number None AWS A 5.18 AWS A5.18 AWS A 5.18
AWS Classification None E70S-2 or E70S-6 E70S-2 or E70S-6 E70S-6
Size None 1/16" & 3/32" 1 116" & 3/32" 0.035"
F-Number None 6 6 6
A - Number None 1 1 1
Maximum Bead Thickness None 0.0625" to 0.268" 0.0625" to 0.268"
3.2 Supplemental Fillers
Supplemental Filler Metal None None None None
Consumable Inserts None None None None
Fluxes None None None None
4.0 POSITIONS
4.1 Welding Position 5G 5G 5G 5G
4.2 Progression Uphill Uphill Uphill Uphill and downhill
5.0 PREHEAT
5.1 Minimum Preheat 70 70 70 70
5.2 Maximum Interpass
Minimum Interpass 50 50 50
Maximum Interpass 200 200 200 200
Preheat Maintenance None Required None Required None Required None Required
6.0 POSTWELD HEAT TREATMENT
6.1 Temperature Range (OF) None Required None Required None Required None Required
6.2 Time at Temperature None Required None Required None Required None Required
6 ANALYSIS OF COILED TUBIl'iG WELDING TECHNIQUES
7.0 GAS
7.1 Shielding Gas
Percent Composition 75%He - 25%Ar
Purity Welding Grade
Flow Rate 15 to 30 CFH
Gas Cup Size No 4 to 6
Gas Lens Optional
Gas Cup to Work Distance Approx 1/2"
7.2 Supplementary Gas
Purge Gas None
Backing Gas None
Trailing Gas None
8.0 ELECTRICAL CHARACTERISTICS
8.1 Current Direct Current
Pulsed Current Permitted
8.2 Polarity Straight
8.3 Tungsten Electrode
Size 3/32"
Type AWS E WTh-2
8.4 Operating Parameters
Root Pass
Amperage 50 - 110
Volts 8 - 15
Travel Speed (ipm) 2-4
Fill Pass(es)
Amperage Not Required
Volts Not Required
Travel Speed (ipm) Not Required
9.0 TECHNIQUE
9.1 Single or Multipass Single Pass
9.2 Single or Multipass Electrode Single
9.3 Backgouging None
9.4 Initial Cleaning (1 )
9.5 Interpass Cleaning Clean Wire Brush
9.6 String or Weave Beads Stringer
Oscillation None
9.7 Peening None
(1) Solvent (MEK, Ethanol) cleaning wire brush, light sanding
(2) Maximum weave 2 times the diameter of the filler metal
75%Ar - 25%He
100%Ar (alt.)
Welding Grade
15 to 30 CFH
No 4 to 6
Optional
Approx 1/2"
None
None
None
Direct Current
Permitted
Straight
3/32"
AWS E WTh-2
50 - 95
8 - 10
2-4
80 - 95
8 - 10
2-4
Single < 0.125"
Multi> 0.125"
Single
None
(1 )
Clean Wire Brush
Stringer
None (2)
None
(3) Single v-groove for wall greater than 0.109" wall: Square butt 0 to 0.109" wall
75%Ar - 25%He
Welding Grade
20 to 40 CFH
No 4 to 6
Optional
Approx 1/2"
None
None
None
Direct Current
Permitted
Straight
3/32"
AWS E WTh-2
50 - 110
10 - 15
2-4
90 - 135
10 - 15
3-4
Single < 0.125"
Multi> 0.125"
Single
None
(1 )
Clean Wire Brush
Stringer
None (2)
None
SPE 36346
75%He - 25%Ar
Welding Grade
35 to 40 CFH
No 6 to 8
Used
1/2"
None
None
None
Direct Current
Synchronized Pulsed
Straight
3/32"
AWS E WTh-2
80 - 120
7-9
4-6
80 - 120
7-9
4-6
Single < 0.125"
Multi> 0.125"
Singie
None
(4)
Clean Wire Brush
Both
Joint Width
None
(4) Internal flashing from longitudinal seam removed by chipping or grinding, internal and external surfaces lightly sanded, weld
surfaces cieaned with acetone or comparable cleaner.