why coiled tubing fails and how to avoid failures in your well

1

SPE Distinguished Lecturer Program

The SPE Distinguished Lecturer Program is funded principally through a grant from the SPE Foundation.

The society gratefully acknowledges the companies that support this program by allowing their professionals to participate as lecturers.

Special thanks to the American Institute of Mining, Metallurgical, and Petroleum Engineers (AIME) for its contribution to the program.

Society of Petroleum Engineers Distinguished Lecturer Programwww.spe.org/dl

2

Why Coiled Tubing Works and How to Make Sure it Works on Your Job

... in Your Well

Why Coiled Tubing Fails and How to Avoid Failure on Your Job

... in Your Wellor

Steven M. TiptonThe University of Tulsa

Society of Petroleum Engineers Distinguished Lecturer Programwww.spe.org/dl

3

The Oilfield is a Rough Place

4

Outline• Introduction / Mechanics Primer• CT Fatigue Research

– TU Coiled Tubing Mechanics Research Consortium

• Surface Defects• Analytical Modeling

– FlexorTU5• Inspection

– 3D laser imaging

5

Background• 1989 - developed CoilLIFE – Schlumberger

• 1994 - JIP to revise CoilLIFE and develop CT test machine

• 1995 - developed coiled tubing fatigue module forNOV-CTES (Achilles)

• 1996 - 2000 University of Tulsa JIP’s to studyCT mechanical and fatigue behavior

• 2000 – present: University of Tulsa Coiled TubingMechanics Research Consortium

6

What is “Coiled Tubing”• Exactly what it sounds like• Continuously Milled Tubing• Rolled from Flat Strip

– Resistance Welded Seam– HSLA (High Strength Low Alloy)

• 55-120 KSI– CRA

• Chrome Alloys, Titanium

• Diameters: 0.75” – 4.5”• Thicknesses: up to 0.25”• Lengths > 30k ft

– 10K – 20k ft most common

Courtesy Quality Tubing

7

Typical CT Rigup

Coil Tubing

Reel ~ 10-20K ft

“Gooseneck”

Guide Arch

Injector

What is it used for?

• Anything “jointedtubulars” are used for!

• Drilling

• Completion

• Servicing

Courtesy ICoTA

8

Coiled Tubing Units

9

Coiled Tubing

2009 SAE Baja Vehicle

10

Bending Strains

Bottom material in COMPRESSION

Top material in TENSION

“Strain” defined as % material stretchesin Tension or shortens in Compression

11

repeats each tripspool

straight straighthole

arch archspool

Bending Cycles

time

stra

in

12

Bending Strains Lead to• ULTRA-Low Cycle Fatigue

– e.g., lives ≈ 10 – 100 cycles to failure• Normal Low Cycle >103 Cycles to Fatigue• Engine Parts >106 Cycles to Fatigue

• Ballooning– >30% even when pressure is <50% yield

• Permanent Elongation– 6-10 ft per 10,000 ft trip, even for axial

loading well below yield

13

Huge Cyclic Plastic Strains (>2%!)

Bending Strains

“Ratcheting”Incremental diameter growth >30%

Lives < 10 cycles!

Even with pressure below

50% of yield

Wall Thinning

bending axis

14

Permanent Elongation

15

Bending Strain Distribution

Strain

16

Strain

εx,m = r/R

r

R

Bending Strain Distribution

17

Bending Stress Distribution

StressSy

-Sy

18

Bending Stress Distribution

StressSy

-Sy

19

Causes Permanent Elongation

StressF

20

CT Fatigue Research

CoilLIFEFatigueTestMachine

21

CT Fatigue Research


22

CT Fatigue Research


23

TU CT Fatigue Test Facility

24

Accelerated CT Fatigue Test Machine

25

5 4 3 2 1 1 2 3 4 5 6 7 8NUMBER OF FAILURES

NUMBER OF FAILURES

Fatigue Crack Development

26

Fatigue Crack Development

MOST fatiguecracks initiateat the inner surface

Fatigue behavior of CTdefies engineering logic!

Appears as a “pinhole” failure

cross section through tubing wall

27

120 ksi, 1.5" x 0.134" x 48"R

0

50

100

150

200

250

300

350

400

450

500

0 2000 4000 6000 8000 10000 12000 14000 16000

Pressure (psi)

Cyc

les

to F

ailu

re

CT Life Prediction Routine

50% (mean) prediction

95% confidence prediction

28

• Factors that influence life:– Diameter– Pressure– Wall Thickness– Bending Radius

• Spool• Gooseneck

– Material Grade

CT Life Prediction

70 – 120 ksi

Larger diameter – larger bending strains

Thinner walls – higher tangential stresses

Higher pressure – higher tangential stresses

Smaller radius – larger bending strains

29

Effect of Tubing Diameter

30

Gooseneck Radius of Curvature

31

Material Selection

32

Fatigue Life Estimation

σf’, b , εf’, S=f(σh, Sy’, Δεx)E’, K’, n’

LCFTesting

Constant PressureFixtureTesting

FittingRoutine

Plasticity / FittingRoutine

MATERIAL DATA SETLCF Data

CT Fatigue Model

εx

εx

time

ComplexField Loading

time

P

Coil Tubing Parameter

time

D

inputinput

output

33

0

100

% L

ife E

xpen

ded

Distance along String

Fatigue Life Profile

• Implemented with String Management Software:CT Life Prediction

warning level

• Fatigue eliminated as a primary failure mode!

34

What About Defects?

35

Means of Imposition

• Milling (cut into surface)• Impressing• Corrosion Pits (chemical milling)• EDM – Different shapes possible

a b fd ec g h i

dw

x

36

Means of Imposition

Milled Ball Impressed Ball

37

Influence of Surface DefectsLow Pressure Impressed vs Milled 1/8" Balls

0

0.2

0.4

0.6

0.8

1

1.2

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45

d/t

%D

efec

t Life

to B

asel

ine

Life

Impressed (t=0.156)

Milled (t=0.156)

Impressed (t=0.204)

Milled (t=0.204)

scatter

Depth / Wall Thickness Ratio

Rat

io o

f D

efec

t Life

to B

asel

ine

Life

2.375” 80 ksi tubing, tested on 72” radius fixture

38

Repairing Defects

• What happens after a defect is discovered?

– Scrap / Replace string (expensive)

– Cut and splice (welding –expensive)

– Grind repair

39

Repairing DefectsOne Option: Butt Welding

Unfinished inside of the butt weld constitutes a severe structural discontinuity!

40

Butt Welding Fatigue Data

0

10

20

30

40

50

60

70

80

90

% o

f Unw

elde

d Sa

mpl

e Li

fe

Manual Welds:average = 38%std. dev = 18%

Orbital Welds:average = 41%std. dev = 11%

Manual Average Orbital Average

41

Grind Repairing

replaced withsmooth, gradual thickness change

Abrupt, localized discontinuity

Technique is important and grind repair guidelines have been developed

42

Fatigue Life Predictive Model

No defect:229

Cut defect:76

Repaired:129

Impressed:214

43

Defects Must be FOUND

• Visual Inspection• Magnetic Flux Leakage

– Most popular approach for CT inspection– MFL can only DETECT defects– Can’t provide any information about defect

size or severity• 3D Laser Imaging for NDE

– Complete geometry of defect is captured digitally

44

3D Laser Imaging System(Low-Cost)

Longitudinal Milled Defect

45

Commercial Laser System

Sawcut

Impressed ballMilled ball

Milled Cylinder

46

Impressed ball Machined ball

Color-coded Digital Data

47

w

d

Ap

Automated Defect Measurements

rmax

rmin

48

Color-coded Digital Data & Photographic Image

Useful to distinguish cut or impressed

49

Sour and Corrosive Environments

•Similar effects: CT to Pipe•H2S

•HIC (Hydrogen Induced Cracking)•SSC (Sulfide Stress Cracking)

•SCC (Stress Corrosion Cracking)

•Residual Stresses•Accelerates Corrosion

Presenter

Presentation Notes

Hydrogen Induced Cracking (HIC) or hydrogen embrittlement is a brittle mechanical fracture caused by penetration and diffusion of atomic hydrogen into the crystal structure of an alloy. It occurs in corrosive environment under constant tensile stress, similar to stress corrosion cracking (SCC); however, cathodic protection initiates or enhances HIC but suppresses or stops SCC. The cracks are usually non-branching and fast growing, and are more often transgranular (through the grains) rather than intergranular (through the grain boundaries). HIC occurs in high strength steels when atomic hydrogen dissolves in the crystal lattice of the metal rather than forming H2 gas. In the oilfield, the presence of H2S gas often leads to sulfide stress cracking (SSC), which is a special case of hydrogen induced stress cracking. Hydrogen may enter a metal surface by the cathodic reduction of hydrogen or water: 2H+ + 2e- → 2HAdsorbed (acidic waters) �2H2O + 2e- → 2HAdsorbed + 2OH- (neutral waters) Normally, the adsorbed hydrogen at the surface recombines to form hydrogen gas: 2HAdsorbed → H2 However, recombination poisons such as sulfide (S2-), prevent hydrogen gas from forming and the adsorbed hydrogen moves through the metal, thereby weakening it. Hydrogen sulfide (H2S) is especially aggressive in promoting hydrogen damage because it provides not only the sulfide poison, but hydrogen ions (H+) as well. Sulfide stress cracking (SSC) occurs in high-strength drill pipes, casing, tubing, and sucker rods. Like stress corrosion cracking (SCC), cracking may not occur below a threshold stress, however, increasing strength and applied stress, increasing H2S concentrations and increasing acidity (decreasing pH) increase SSC susceptibility. As opposed to SCC, decreasing temperature also increases SSC susceptibility. Time to failure is minimum at room temperature. The ramification is that, steels become most susceptible to SSC near the surface where the highest strength is required to carry the weight of the string. Increasing the wall thickness of the tubular can reduce the applied stress thus allowing the use of lower strength steels, but strength must be balanced against the applied load at the top of the joint due to increasing weight. High strength casing may be used deeper in the well where temperatures are higher. In SCC, failure initiates at the crevices on the metal surface, usually in the pits. Thus, SCC susceptibility of steels is related to its susceptibility to pitting. Whereas SSC generally initiates at impurity inclusions in the metal, hence it is dependent on the hydrogen absorption characteristics of the metal. Microstructure of steel also influences the SSC susceptibility. Quenched and tempered steels have better SSC resistance than normalized and normalized and tempered steels. Acceptable hardness limits for many alloys in sour service are described in the National Association of Corrosion Engineers (NACE) Specification MR-01-75. For SSC resistance, the hardness of carbon and low alloy steels must be maintained below 22 Rockwell Hardness C (HRC). Tubulars based on AISI 4100 series low-alloy steels are acceptable up to HRC 26. Higher alloyed steels may have higher hardness levels. A special case of hydrogen damage is known as hydrogen blistering. Hydrogen blistering occurs when hydrogen atoms diffuse into the steel, and hydrogen gas nucleates at internal defects and inclusions, forming voids which eventually generate enough pressure to locally rupture the metal. Hydrogen blistering is occasionally observed in the oilfield in sour systems.� Courtesy New Mexico Tech HTHA (High Temperature Hydrogen Attack)

50

Manufacturing and Materials defects/flaws

•From 7 to 20 reels per 1,000 reels (~last 3 yrs)

–0.7 to 2% –Only Those Reported

•Failures occur that are not fully investigated

–Especially later in CT Life

Courtesy BJ Services

‘Cold’ Weld

51

Conclusions• Fatigue is inherent to the use of CT• Large CT Bending Strains Cause

– Fatigue, diameter growth, elongation• Scatter is Associated with Fatigue Data • Fatigue Software can Predict Influence of

– Material, diameter, wall thickness, bending radius, internal pressure, statistical scatter, defects, grind repairs

• Field Monitoring Software used to Assess Every Section of Tubing Along the Entire String – Avoids failures in the field

52

Conclusions

• Defects can have first order influence (cut)– Or negligible influence (impressed)

• Repairing by Grinding is a viable option• Software helps to quantify defect severity

(and effectiveness of a grind repair )• 3D Laser Scanning

– Holds the potential to detect defects– Capable of measuring every detail about their

geometry

53

ConclusionsTo Avoid Failures on Your Job

.... or In Your well:• Control the Factors that Contribute to

Fatigue Damage Accumulation– Minimize tubing diameter– Maximize spool and

gooseneck sizes– Select material most

suited for application– Minimize tripping– Minimize tubing movement

with high pressure

54

Conclusions• Avoid Defects

– Eliminate sources of defects - handling, injector debris…

– Monitor/Inspect– Environmentally

Induced Cracking• Storage corrosion

– Repair shallow defect using grind repair techniques

55

Why Coiled Tubing Fails and How to Avoid Failure on Your Job

... in Your WellSteven M. Tipton

The University of Tulsa

[email protected]

918-850-0252

www.coiledtubingutulsa.org

Why Coiled Tubing Succeeds and How to Assure Success on Your Job

... in Your Well

56

Society of Petroleum Engineers Distinguished Lecturer Programwww.spe.org/dl 56

Your Feedback is ImportantEnter your section in the DL Evaluation Contest by

completing the evaluation form for this presentation or go online at:

http://www.spe.org/events/dl/dl_evaluation_contest.php

why coiled tubing fails and how to avoid failures in your well

Documents