Casing/Tubing design manual october 2005 Chevron

Подождите немного. Документ загружается.

Casing/Tubing Design Manual 4-17

April 2007

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

20,000

-60,000 -40,000 -20,000 0 20,000 40,000 60,000 80,000

Axial Stress (psi)

Triaxial, 9.625" OD, 0.545" Wall

1D API, 9.625" OD, 0.545" Wall

Triaxial, 7.0" OD, 0.73" Wall

1D API, 7.0" OD, 0.73" Wall

1.05 x API

1.10 x API

0.8 x API

1D API Yield - Neglects Axial Influence

Triaxial

Yield

1D API Yield

Triaxial

Yield

40%

Lower

50%

Lower

Figure 4-7. Casing (Triaxial) VME Burst Pressure and API (1D) Burst

Pressure

4.8.5 Burst Strength Based on Full Yield

An even higher tube burst pressure resistance than the API rating is also

available for some designs. This alternate performance is referred to as “heavy

wall rating” and is based on tube body full yield under multi-dimensional stresses

[Murphey, 1994]. The tube body full yield is referred to the tube body yield

through the entire cross-section of the tube. This tube full yield burst strength is

formulated as:

875.0

⎟

⎠

⎞

⎜

⎝

⎛

−

⎟

⎠

⎞

⎜

⎝

⎛

BAPI

(4-19)

The factor 2/

√3 accounts for the effect of the radial stress. The factor

()

1+ tD

results from the full-yield condition. The thicker the tube wall thickness, the more

significant the effect of the factor

(

)

1+ tD on the burst rating.

4.9 Tube Body Yield

The tube body yield has actually been used in the above to formulate the tube

VME burst strength and tube full yield burst strength. The general non-API

formulation that tests for tube body yield uses a multi-dimensional stress state.

The calculation is a combination of two classic formulations whose stress states

are superimposed:

• The Lame’ equations for a thick cylinder, loaded by internal pressure,

external pressure, and undergoing generalized plane strain.

• The bending equations for a beam undergoing small bending under

which it is assumed that plane cross sections remain plane.

4-18 Casing/Tubing Design Manual

April 2007

Given the following input variables:

• Internal pressure,

• External pressure, p

• Axial stress,

• Local curvature of tube axis, c

• Tube outside diameter, D

• Tube wall thickness, t

• API or specified minimum yield stress of tube material,

API

• Young’s modulus,

• Poisson’s ratio,

The effective stress that relates a multi-dimensional stress state to a uniaxial

stress state is calculated in the following manner.

Determine the radial, circumferential, and axial stresses in the tube body as

given by the Lame’ equations [c.f. Timoshenko and Goodier, 1970]:

()

ii oo

ioio

pr p r

pprr

rr r

−

22 2

(4-20)

()

−

pr pr

pprr

rr r

ii oo

ioio

22 2

(4-21)

σµ ε

ii oo

pr p r

−

(4-22)

where

is radial stress,

is circumferential stress,

is axial stress, r

internal radius,

is external radius and r is the radial coordinate, rrr

≤

≤ .

Note that the axial stress may appear as input to the problem, instead of being

calculated from more primitive terms as indicated above. This is typically the

case when the end conditions in the problem are expressed as forces rather than

constrained displacements. If it is necessary to calculate

, the additional input

variables are

, Poisson’s ratio, and

, the axial strain.

If the tubular is in a curved wellbore, or if the tubular is helically buckled, there

will be an additional axial stress component due to bending. Assuming that the

cross section of the tube remains plane during bending, the bending stress will

vary from zero along the axis of the tube to an extreme on the outer surface of

the tube given by:

Ecr=±

(4-23)

where the tensile and compressive bending stresses occur on opposite sides of

the tubular.

Casing/Tubing Design Manual 4-19

April 2007

The equivalent stress can now be calculated:

() ()()

σσσσσ σσσσσσσσ

θθθ

e r zb r rzb zb

=+++ − − +− +

(4-24)

and compared to the API or specified minimum yield stress of the tube material

Because of the close relation between the effective stress as defined above and

the von Mises yield criterion for metals, the ensuing safety factor is often termed

the von Mises safety factor.

4.9.1 Example Problem

At a certain depth in a well, the 177.8 mm, 43.16 kg/m (7 in., 29 lb/ft), N-80 tube

is subjected to an internal pressure of 15 MPa (2,176 psi), an external pressure

of 6 MPa (870 psi), and an axial force of 444822 N (100,000 lb). The tube is

being run through a portion of the well having a dogleg of 10 deg/30 m (10.16

deg/100 ft). Compute the yield safety factor.

The calculations can be summarized as follows in Table 4-6.

Table 4-6. Calculations

Action Variable Metric English

Input variables

15 Mpa

78.54 mm

6 Mpa

88.90 mm

206843 Mpa

2,176 psi

3.092 in

870 psi

3.500 in

30x10

psi

Compute curvature

5.818x10

-3

rad/m 1.773x10

-3

rad/ft

Compute stresses at r

-15 Mpa

67.01 Mpa

81.62 Mpa

±94.52 MPa

-2176 psi

9,721 psi

11,835 psi

±13705 psi

Compute stresses at r

-6 Mpa

58.01 Mpa

81.62 Mpa

±106.98 MPa

-870 psi

8,415 psi

11,835 psi

±15514 psi

Compute effective stress

at r

ε−

166.09 Mpa

80.98 MPa

24,082 psi

11,747 psi

Compute effective stress

at r

ε−

171.79 Mpa

75.57 MPa

24,910 psi

10,962 psi

Compute safety factor

(

)

3.21

If the bending stress is non-zero, four calculations must be made (inner and outer

radius and positive and negative bending stress). In the absence of bending, the

effective stress,

will always be largest at the inner radius.

The API version of this equation is solely a check for yield under the

condition

===0 , and is a special case of the procedure recommended here.

4-20 Casing/Tubing Design Manual

April 2007

The axial stress,

, can be given in a variety of ways. Here, we assume it is

available from some other related calculation.

The safety factor is computed using the maximum of the computed effective

stresses. In this example, the maximum stress occurs at the outer radius on the

tensile side of the tube bend.

4.10 Tension/Compression Strength

Tube tension/compression strength (resistance) measures the structural

resistance of the tube cross section to an axial tension/compression load that

causes yield of the tube body. The axial tension/compression resistance of tubes

is based on the following principles:

• The axial tension resistance and compression resistance of tubes are

considered the same for the tube body, where the tube body starts to

yield under axial tension or compression load.

• The axial tension/compression resistance of the connection is usually

less than that of the tube body, because of the smaller cross-section

area of the connection critical plane(s). Proprietary connection vendors

rate their connections on axial tension and compression resistance

based on their connection design.

• For API connections, particularly LTC and STC connections, the axial

tension/compression resistance is less than that of the tube because of

thread jump-out risk.

Tube tension/compression performance is calculated using the procedure

recommended by the API (Latest Edition). Given the following input variables:

• Tube outside diameter, D

• Tube inside diameter, d

• API or specified minimum yield stress of tube material,

API

The tube tension/compression resistance is calculated by the following

procedure:

Compute the tension/compression load necessary to yield the tube body

according to the following formula:

()

dDF

API

−=

(4-25)

4.11 References

4. Bull. 5C3, Bulletin on Formulas and Calculations for Casing, Tubing, Drill

Pipe and Line Pipe Properties, latest edition, API, Washington DC.

5. Bradley, W. B.: "The Effect of Casing Wear on the Burst Strength of Casing,

Part 1, Joint Leakage," paper 75-PET-11 presented at the 1975 30

Annual

Petroleum Mechanical Engineering Conference, Tulsa, Oklahoma, 21-25

September.

Casing/Tubing Design Manual 4-21

April 2007

6. Bradley, W. B.: "The Effect of Casing Wear on the Burst Strength of Casing,

Part 2, Statistical Burst Strength of Worn and Unworn Casing Strings," paper

75-PET-27, presented at the 1975 30

Annual Petroleum Mechanical

Engineering Conference, Tulsa, Oklahoma, 21-25 September.

7. Krattus, J. R.: “Ferrous Alloys” in Aerospace Structural Metals Handbook,

Revised December 1973.

8. Kuriyama, Y., Tsukano, Y. and Mimaki, T.: "Effect of Wear and Bending on

Casing Collapse Resistance," paper SPE 24597 presented at the 1992 67

Annual Technical Conference and Exhibition of the SPE, Washington, DC, 4-

7 October.

9. Leslie, W. C.: The Physical Metallurgy of Steels, McGraw-Hill, New York City,

(1981).

10. Murphey, C. E.: "Tubing Design," appendix C, CPTC-DWPD Tech Memo 94-

02, Houston, Texas (April 1994).

11. Wu, Jiang: “Casing and Tubing Design Criteria,” ChevronTexaco ETC Tech

Memo #2002-24, Houston, Texas (September 2002).

12. Nester, J. H., Jenkins, D. R., and Simon, R.: “Resistances to Failure of Oil-

well Casing Subjected to Non-uniform Transverse Loading,” Drilling and

Production Practice 1955, (1956) 374-378.

13. Pattillo, P. D., Moschovidis, Z. A., and Lal, M.: "An Evaluation of Concentric

Casing for Nonuniform Load Applications," SPE Drilling and Completion,

(1995) 186-192.

14. Pattillo, P. D. and Rankin, T. E.: "How Amoco Solved Casing Design

Problems in the Gulf of Suez," Petroleum Engineer, (1981) 86-112.

15. Smith, M. B. and Pattillo, P. D.: "Analysis of Casing Deformations Due to

Formation Flow," Proc., Applied Oil and Geoscience Conference, Edmonton,

Alberta, (1980).

16. Timoshenko, S. P., and Goodier, J. N.: Theory of Elasticity, third edition,

McGraw-Hill, New York City, (1970).

17. Wu, Jiang: “Casing Burst Strength after Casing Wear,” paper SPE 94304,

presented at the 2005 SPE Production and Operations Symposium,

Oklahoma City, Oklahoma, 17–19 April.

5 Tube Connections

5.1 Introduction ......................................................................................................................5-1

5.2 Joint Strength...................................................................................................................5-2

5.2.1 Upsets.........................................................................................................................5-3

5.2.2 Low-Angle Load Flanks...............................................................................................5-3

5.3 Leak Resistance...............................................................................................................5-3

5.3.1 Teflon

Ring................................................................................................................5-4

5.3.2 Metal-to-Metal Seals....................................................................................................5-4

5.4 Internal Stresses ..............................................................................................................5-5

5.4.1 Cylindrical Threads......................................................................................................5-5

5.4.2 Torque Shoulder..........................................................................................................5-5

5.5 Other Considerations .......................................................................................................5-5

5.5.1 Internal Profile .............................................................................................................5-6

5.5.2 Large OD Connection Backout Torque........................................................................5-6

5.6 Chevron-Approved Connections ......................................................................................5-6

5.6.1 Chevron Connection Classification..............................................................................5-6

5.6.2 Well Conditions for Connection Selection....................................................................5-8

5.6.3 Connection Selection Example..................................................................................5-14

5.6.3.1 Step 1–Use Table 5-3 to Determine the Required Connection Sealing

Class ..............................................................................................................................5-14

5.6.3.2 Step 2–Use Table 5-4 to Determine the Required Structural Class......................5-14

5.6.3.3 Step 3–Use Table 5-1 to Select the Possible Casing Connections.......................5-15

5.7 Tabulate Summary on Standardized Approved Premium Connections .........................5-16

5.8 Connection Information..................................................................................................5-17

5.9 References.....................................................................................................................5-45

5.1 Introduction

Individual tube bodies are joined by threaded connections. These connections

can range from simple, inexpensive API designs available from a number of

vendors to proprietary design that may cost as much as the tube body.

The ideal connection would be transparent. That is, it would allow the tubular

string to behave as if it were one continuous cylinder. Unfortunately, connections

are more commonly detrimental to the performance of the tube in at least one of

several important attributes. The degree to which the tube integrity is adversely

affected is a major determinant in the cost of a connection.

The following sections discuss critical aspects of connection performance. The

intent is to explain the variety of performance behaviors that can be expected

and to offer some insight into the disparity in costs between different connection

designs.

In this introductory discussion, the datum is the API round thread design,

specifically as embodied in the LT&C connection. This design has a long history,

documenting both its strengths and weaknesses, and it is still in wide use

because of its low cost.

Casing/Tubing Design Manual 5-1

October 2005

The premise of this discussion is: “Given the low cost and ready availability of

API LT&C, what performance advantages are to be gained from a more

expensive connection design that would justify the additional cost?”

Following this introduction, a list of connections currently approved for Chevron

use is reviewed by presenting a brief data sheet on each design.

5.2 Joint Strength

Joint efficiency for casing connections is defined as the ratio of the parting load of

the connection to the parting load of the tube body

1,2

. Typically, joint strength is

related to the critical section area. The critical section area of a connection is the

cross-sectional area in the pin or box which, when combined with the axial force

at the same location, yields the largest stress. Because axial load transfer in a

connection takes place from the base of the pin to the base or center of the box,

it is usually one of these locations that has the largest stress and thus contains

the critical section area.

For the API round thread design, the critical section area is normally at the base

of the pin. Because the round thread profile is cut into a non-upset pipe body

(except for EUE tubing) and the very base of the thread form is a pull-out region

for the machining tool, the critical section area of the connection will be less than

the cross-sectional area of the tube body. The joint efficiency of API LT&C,

therefore, is less than one hundred percent.

Compounding the deficiencies of the round thread design is the fact that the load

flank angle of the thread profile permits relative movement between the pin and

coupling. If the radial component of this relative movement equals the height of

the thread, then the connection can jump out prior to the fracture of either

connection element. Jump out is characterized by gross plastic deformation in

the connection, and, therefore, is not as reproducible as fracture. Whenever

possible, connection designs (or diameter/thickness ranges of connection

designs) resulting jump outs should be avoided.

To counter the two problems in the API round thread design of (1) low efficiency

and (2) jump out over a wide range of diameter/thickness ratios, proprietary

thread manufacturers have offered varying solutions.

Tubing connections are rated relative to minimum yield, rather the than minimum

ultimate tension or failure, used for casing connections. For tubing connections, joint

efficiency is the ratio of the axial load to cause yield in the connection to the axial load

required to cause yield in the tube body. For casing connections, joint efficiency is the

ratio of the axial load to cause ultimate tension in the connection, or failure by some

alternate mechanism such as jump out, to the axial load required to cause ultimate tension

in the tube body.

Almost without exception the load referred to in the definition of joint strength is axial

tensile load. The behavior of threaded connections in combined loading situations is too

complicated to admit a definition of efficiency based on a multi-dimensional load state.

5-2 Casing/Tubing Design Manual

October 2005

5.2.1 Upsets

For thread designs that do not involve thread profiles that traverse the entire wall

thickness of the tube, one means of increasing joint efficiency is to increase the

thickness of the tube body in the vicinity of the connection. The thread form is

then cut on a “thicker tube” resulting in an efficiency that, while lower than the

strength of the upset, is greater than the strength of the majority of the tube body.

Two disadvantages to the implementation of upsets are:

1. The upsetting process involves severe forging of the end of the tube and

must be followed by heat treatment. Although seldom a problem in modern

upsetting facilities, the possibility of undesirable stresses following upsetting

does exist.

2. Upsetting limits the number of times a thread design can be recut on a tube

without having to re-upset the tube end.

5.2.2 Low-Angle Load Flanks

To minimize jump out, most proprietary threads lower the angle of the load flank

of the thread. The amount of this alteration can vary from threads which copy the

API buttress thread form (3° load flank) to threads that actually go beyond 0° and

possess a negative load flank angle. This latter design is particularly prevalent in

clearance connections where the overall thickness of the connection dictates that

the threads must withstand the maximum possible load.

The statements above, of course, pertain primarily to tensile loading. Often the

stab flank of a thread has quite a high angle (on the order of 45° in some

offerings) to benefit stabbing the pin into the box on assembly. This high stab

flank angle is conventionally counteracted by a shoulder at some point in the

connection intended to carry the compressional loading. As an alternative to such

a shoulder, some vendors actually cut both flanks of the thread with negative

angles.

5.3 Leak Resistance

Permeating the round thread design are two spiral leak paths, one following the

thread root and one following the thread crest. One intended purpose of thread

lubricant is to block these paths when small ductile metal particles in the lubricant

grease carrier deform during assembly. This concept is flawed by the following

possibilities:

• There is no guarantee that the metal particles will bridge at any point in either

of the potential leak paths. That bridging does not always occur is evident

from the number of leaks in older wells where the grease carrier is no longer

present.

• The grease carrier has temperature limitations. At atmospheric pressure, the

grease carrier commonly used in API-modified thread compound will

vaporize at approximately 138°C (280°F).

Casing/Tubing Design Manual 5-3

October 2005

• The grease carrier is typically a petroleum-based product and can be leeched

away by contact with wellbore hydrocarbon fluids.

• Heavy metal particles used to promote the bridging in of many conventional

lubricants pose environmental and health hazards.

• Inter-thread movement during the assembly and subsequent application

loads, particularly hanging tension, can redistribute both thread lubricant and

the contact loads on the thread flanks.

• This problem is even more critical in API buttress connections, where the

clearances are potentially greater and the shape of the thread profile

enhances redistribution of the lubricant.

To combat these problems, the common solution is to ensure that at least one

location in the inter-thread area contains a positive barrier to fluid extrusion. The

nature of this barrier can assume different forms as described in the following

sections..

5.3.1 Teflon

Ring

Both API and proprietary vendors offer designs modified by a groove in which is

inserted a Teflon

ring. Teflon

rings may not, however, be the best solution for

the following reasons:

• Installation of the ring is not trivial, offering one more possibility for an error in

assembly of the string.

• The tolerance on ring dimensions vis-à-vis the dimensions of the groove,

particularly for the API design, admits the possibility of either inducing

unnecessary stresses with an oversized ring or failing to achieve a seal with

an undersized ring.

• Teflon

has a coefficient of thermal expansion roughly ten times that of steel.

Use of Teflon

in a high-temperature environment could unduly stress a

connection.

• Above a temperature of roughly 120°C (250°F) Teflon

loses its memory and

will not return to its previous configuration during cooling.

Despite these limitations, and particularly in instances where a pressure seal is

critical but the pressure is not extreme, Teflon

represents a positive alternative

to thread lubricant as a primary seal.

5.3.2 Metal-to-Metal Seals

The most popular and most satisfactory seal in a threaded connection is to

include a segment of the inter-thread region, usually at the end of the pin, where

positive interference between metal surfaces exists for the entire circumference

of the connection.

The intricacies of metal-to-metal seal design are beyond the scope of this text.

However, certain basic principles should be followed in rating a metal-to-metal

sealed connection.

5-4 Casing/Tubing Design Manual

October 2005

A radial seal, that is a sealing surface where the direction of contact stress is

radial, are preferred over an axial seal. A torque shoulder used to limit the

relative axial displacement of the pin and box during assembly is not to be

considered a seal. Such surfaces usually lose some or all of their contact stress

when loaded axially, and are particularly susceptible to deep environments.

5.4 Internal Stresses

A well-recognized shortcoming of the round thread design is that the only aspect

of the design opposing excess assembly turns is the thread taper. According to

the relative wall thicknesses in the pin and box, excess assembly can result in

yielding at the end of the pin or, conversely, excess tension in the box. This latter

situation is of particular concern in sour service. Here, with the middle of the box

exposed to wellbore fluids, sulfide stress cracking is a documented failure mode.

To combat the possibility of excess assembly stress, the following two

alternatives are typically applied.

5.4.1 Cylindrical Threads

Designing a thread with zero taper will, of course, eliminate stresses associated

with a tapered geometry. An entire line of highly respected threads, originally

developed by Hydril

, employ this concept.

The one concern with cylindrical threads, inadvertent disassembly in service, has

not proved to be a problem. Cylindrical thread designs typically possess a torque

shoulder. The friction force associated with this shoulder is usually sufficient to

prevent connection rotation in the wellbore.

Because of the low interference in the inter-thread region, cylindrical threads

have proven to be particularly advantageous in applications where galling is a

concern, such as in the use of “13 Chrome” metallurgies.

5.4.2 Torque Shoulder

A number of thread designs maintain thread taper, but limit the amount of

interference between the pin and box by means of a positive stop, a torque

shoulder. Provided it is not mistaken for a metal-to-metal pressure seal, the

torque shoulder is a simple, effective solution to the problem of excess assembly

stress.

5.5 Other Considerations

Miscellaneous additional items worth considering in the selection of a connection

for a specific application include the following:

Casing/Tubing Design Manual 5-5

October 2005