Casing/Tubing design manual october 2005 Chevron

Подождите немного. Документ загружается.

Casing/Tubing Design Manual 4-7

April 2007

Figure 4-2. Tube Collapse Pressure Measurement from the DEA-130 Project

Figure 4-3 presents the statistical results of the measured tube collapse pressure

distribution data from a JIP (DEA-130). Tube collapse performance may vary

among different tube manufacturers.

0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8

Normalized Collapse (Actual/API-min)

Erlanger-Newport(1.293; 0.077; 21)

Tamsa(1.48; 0.137; 9)

Grant-Prideco(1.378; 0.175; 9)

Kawasaki(1.224; 0.055; 9)

Maverick(1.139; 0.059; 12)

NKK(1.53; 0.129; 9)

Lone Star(1.394; 0.113; 11)

North Star(1.365; 0.121; 9)

Sumitomo(1.349; 0.049; 10)

US Steel(1.317; 0.066; 9)

Siderca(1.29; 0.042; 9)

V&M(1.502; 0.166; 17)

Figure 4-3. Tube Collapse Performance Statistical Results from the DEA-130 Project

4.3 High Collapse Casing

As an exception to the API collapse resistance calculation procedure, high-

collapse casing with higher tube collapse resistance may be ordered from some

4-8 Casing/Tubing Design Manual

April 2007

manufacturers. Chevron recognizes some manufacturer's high-collapse ratings

on the following grades:

Table 4-5. Acceptable High-Collapse Vendors and Products

Company 80 Grade 95 Grade 110 Grade

Mannesmann MW95HC MW110HC

Nippon Steel

NT95HS NT110HS

NKK NKT95 NKT110

Sumitomo SM95T SM110T

Lone Star Steel HCL80 S95

TCA L80HC*

USS FSS95**

Tenaris (TAMSA) TAC95***

* For D/t >14 and with 800-psi collapse-rating reduction

** With 900-psi collapse-rating reduction (FSS95 is a high-collapse 80-grade from USS)

*** For diameter 9-5/8 in. only

Ratings are based on review of manufacturing specification and statistical analysis on tube-

collapse pressure test data.

The adjustment of the high collapse casing rating for tension load uses the

biaxial tension correction factor rather than the API procedure outlined above.

Given the following input variables:

• Axial stress,

• API or specified minimum yield stress of tube material,

API

• Manufacturer's high collapse rating in the absence of tension, p

The collapse adjustment for tension is calculated by the following procedure:

1. Compute the high-collapse resistance adjusted for the presence of the axial

tension according to the formula:

Collapse

API

p=−

⎛

⎝

⎜

⎞

⎠

⎟

−

⎛

⎝

⎜

⎞

⎠

⎟

. (4-14)

2. Adjustment of the high collapse casing rating for compression is ignored.

3. Adjustment of collapse resistance for internal pressure follows the API

procedure outlined above.

4.3.1 Example

Given a tube of 244.48 mm (9-5/8 in.), 69.94 kg/m (47 lb/ft), S-95 casing (high-

collapse rating 48.95 MPa (7,100 psi)) in the presence of 1,334 kN (300,000 lbs)

tension and an internal pressure of 6.895 x 10

-3

MPa (1,000 psi), and compute

the collapse resistance.

Casing/Tubing Design Manual 4-9

April 2007

• The input variables are:

ppsi

1000

()

psi=

−

300000

07854 9 625 8681

22104

.. .

Din

9625.

tin

0.472

API

psi

95000

• The collapse resistance in the absence of internal pressure is:

Collapse

psi=−

⎛

⎝

⎜

⎞

⎠

⎟

−

⎛

⎝

⎜

⎞

⎠

⎟

22104

95000

22104

95000

7100 6130

• With 1,000 psi internal pressure the external pressure necessary to

collapse the tube is:

ppsi

=+−

⎛

⎝

⎜

⎞

⎠

⎟

=6130 1

2039

1000 7030

4.4 Collapse Due to Non-Uniform

The collapse performance property of casing as defined by the API assumes a

tube to be loaded everywhere by external fluid pressure. Unfortunately, the

stresses imposed by a flowing formation are not necessarily directed radially,

and, therefore, API collapse ratings for a tube are no longer applicable.

Nester, et al. [1956] considered the problem of non-uniform cross-sectional

loading in two simple distributions (see Figure 4-4) that should be sufficient to

emphasize the relative importance of key variables. Using first yield as a failure

criterion, the authors found that for uni-directional loading, the distributed load,

to cause yield is:

≅

⎛

⎝

⎜

⎞

⎠

⎟

(4-15)

For the case of opposed line loads, the intensity,

, to cause first yield is:

≅

−

⎛

⎝

⎜

⎞

⎠

⎟

−

⎛

⎝

⎜

⎞

⎠

⎟

1096 032..

(4-16)

In either of the above cases, the integrity of the cross section is proportional to

the square of the thickness to outside diameter ratio of the tube. By contrast, for

conventional collapse of a cross section because of hydrostatic fluid pressure,

4-10 Casing/Tubing Design Manual

April 2007

the collapse resistance is linearly proportional to t/D for all tubes except those

thin enough to be governed by elastic collapse

. See Figure 4-4 for an example

of non-uniform loads.

a. Uni-directional Load

b. Opposed Line Loads

Figure 4-4. Non-Uniform Loads Considered by Nester et al. [1956]

The implications of equations (4-12) and (4-13) are important. When required to

design for a flowing formation that can be expected to impose non-uniform cross-

sectional loading, it may be more prudent to gain structural integrity by increasing

wall thickness than by increasing the yield strength of the tube material. To

double non-uniform load resistance, it is necessary to double yield strength,

whereas a wall thickness increase of approximately forty percent can achieve the

same effect.

Increasing wall thickness to take advantage of the behavior described above

provides the impetus for considering concentric configurations, a term referring to

the positioning of dual casing strings, structurally connected by a cement sheath,

opposite the offending rock. The intent is that the casing/cement/casing

composite will act as a unit to increase the effective wall thickness opposing the

non-uniform loading.

Further thought also suggests the following observations on the concentric

configuration [Pattillo et al., 1995]:

• Although it is not necessary that the cement element of the cross section

be particularly strong, this element must be sufficiently competent to

transmit loads between the casing elements. Otherwise, the cross

section does not behave as a unit, and the advantages implied by the

single cross section analysis of Equations.(4-12) and (4-13) are no

longer available. As a corollary, it is worth noting that an annular

configuration with no cement is of practically no benefit. It should be

Nester, et al., considered the case of plane strain.

Casing/Tubing Design Manual 4-11

April 2007

expected that with increasing load, the outer member will first fail as

described above, and then, with continued deformation, impose opposed

line loading on the inner casing. In fact, Pattillo et al. [1995] have

demonstrated that a partially cemented concentric configuration can

actually be weaker than the inner string by itself.

• Failure/yield of the cement element of the cross section does not imply

disaster, and, in some cases, may actually prove beneficial. Yielding of

the cement will alter the non-uniform load pattern to a more benign

distribution, and thus decrease the intensity of the load on the inner

casing element.

• Even if the inner and outer casing elements are not concentric, the

effective wall thickness of the cross section is increased everywhere. It

should, therefore, be expected that even decentralized concentric

configurations should prove beneficial in a non-uniform loading

environment.

• It is not necessary that the individual tubular elements of a concentric

configuration be of particularly high yield strength. The emphasis in non-

uniform loading is on wall thickness. One exception to this involves

formations that move during the installation of the concentric

configuration, a typical example being mobile salts. If the concentric

configuration is being formed by a liner overlap, the outer string will have

to be strong enough to resist salt movement alone until the next hole

section is drilled and the overlapping liner run and cemented.

The industry has realized enhanced casing integrity opposite deforming

formations by applying concentric casing. As examples, both in the massive

South Gharib salt of the Gulf of Suez [Pattillo and Rankin, 1981] and operations-

induced flow of the oil sands in Athabasca [Smith and Pattillo, 1980] the

implementation of concentric casing has proved a viable means of providing

completion integrity in the presence of weak rock.

Despite the successes noted above, concentric completions are not a panacea.

For example, although the concentric configuration did prevent cross-sectional

collapse in Athabasca, the completion string continued to bend with the induced

formation flow. It is important, therefore, to ascertain the type of deformation

mechanism active in a given locale, and then decide if concentric casing can

mitigate that mechanism.

4.5 Operational Considerations

An important consideration associated with the application of concentric casing is

the reduction in inside diameter at the overlap. Notice should be taken of the

possible detrimental influence this remedial step may have on subsequent

operations, such as the installation of artificial lift equipment.

Concentric casing configurations are often run in the overburden to combat

mobile salt formations, such as salt creeps, exhibiting near fluid behavior. As a

result, it may be reasonably assumed that the magnitude of the horizontal stress

imposed by a mobile salt is equal to the local value of the overburden stress. At

depth, the gradient of overburden stress is approximately 22.6 KPa/m (1 psi/ft).

4-12 Casing/Tubing Design Manual

April 2007

The horizontal stress imposed by a mobile salt may, therefore, be approximated

by the product of the overburden gradient and the true vertical depth of the salt.

The above reasoning has lead to a rule-of-thumb of designing opposite salt by

comparing the API collapse resistance to the local overburden stress. However,

this design rule is not recommended. Although the magnitude of the horizontal

stress is equal to the overburden stress, the character of the stress will more

closely resemble the depictions of Figure 4-1 rather than the radial loading upon

which the API formulas are based. These non-uniform loading patterns are much

more detrimental to the integrity of the cross section than uniform loading. As a

general rule, when designing for non-uniform loading, whether or not the

offending formation is a mobile salt, the (combined) wall thickness used should

be the maximum practical within other constraints.

4.6 The Thick Wall Alternative

In instances where cementing may be difficult, thus compromising the integrity of

a concentric configuration, a single, thick-walled string of casing may provide a

suitable alternative. Thick-walled casing, however, is not without its own

disadvantages [Pattillo et al., 1995]:

• Thick-walled casing is a non-standard inventory item and may not,

therefore, be readily available.

• In horizontal wellbores, where difficulty in providing complete

circumferential cement coverage points to thick-walled casing, it may be

difficult to force the heavier tube the entire length of the horizontal

extension.

• The bending stiffness of thick-walled casing may cause running

difficulties in short radius build sections.

• Use of such a high weight product may exceed the load bearing capacity

of the drilling rig

4.7 Effect of Wear on Collapse

Resistance

The experimental data available on the effect of wear on collapse resistance is

scarce, but [Kuriyama et al., 1992] indicates that reduction in collapse resistance

is directly proportional to reduction in wall thickness. That is, if tool joint wear

reduces the wall thickness of a section of casing by twenty percent, then the

collapse resistance of the casing is also reduced by twenty percent.

This concern may not be as crucial for offshore operations where the added expense of

the thick-walled tubular may be offset by elimination of rig time necessary to run and

cement the inner member of the concentric configuration.. Further, weight is not an over-

riding concern if, as is often the case, the heavy string is to be run as a liner. Finally, in

some instances buoying the heavier string during running may be used to counter the

disadvantage of its weight.

Casing/Tubing Design Manual 4-13

April 2007

4.8 Burst Strength

Tube burst (Internal pressure) resistance measures the structural resistance of

the cross section to an internal pressure differential. Internal pressure resistance

of tubes is to be based on the following principles:

• The internal pressure resistance of the connection may or may not

exceed that of the tube body. Most proprietary connection vendors rate

their connection as having an internal pressure resistance that meets or

exceeds that of the tube body.

• For API connections, particularly special clearance buttress connections,

it is possible for the internal pressure rating of the tube to exceed that of

the connection.

• In actual designs, internal pressure resistance, as defined by the API is a

relatively incomplete measure of the integrity of the tube body. A more

accurate, (and recommended) approach is to recognize that internal

pressure resistance is automatically accounted for in the general check

of multi-dimensional body yield with the von Mises yield criterion.

Tube burst pressure performance is calculated using the procedure

recommended by the API latest edition. Given the following input variables:

• Tube outside diameter,

• Tube wall thickness, t

• API or specified minimum yield stress of tube material,

API

The tube burst pressure resistance is calculated by the following procedure:

Compute the internal pressure necessary to yield the tube body according to the

following formula:

API

⎛

⎝

⎜

⎞

⎠

⎟

0 875

(4-17)

4.8.1 Example Problem

Given a tube of 244.5 mm (9-5/8 in.), 69.94 kg/m (47 lb/ft), N-80 casing, compute

the internal pressure resistance.

• The input variables are:

Dmmin

2445 9 625...

tmmin

1199 0..472.

API

MPa psi

552 80000

• The internal pressure resistance is:

MPa

⎛

⎝

⎜

⎞

⎠

⎟

=0875

2 552 1199

2445

47.

psi

⎛

⎝

⎜

⎞

⎠

⎟

=0875

2 80000 0472

9625

6870.

4-14 Casing/Tubing Design Manual

April 2007

Note the following:

The API calculation procedure actually is quite specific concerning both

intermediate and final rounding of the results of calculations. Ignoring these

round-off procedures will only slightly affect the final answer.

4.8.2 Effect of Wear on Burst Strength

Casing wear by drillstring results in a thinner portion of casing wall and a

reduction on casing burst strength (the ability to hold internal pressure). How to

estimate the reduced casing burst strength on such a “crescent-worn” casing has

been an important issue in the oil and gas industry because it is directly related

to how to safely design casing strings. A common approach is to estimate the

reduction of casing burst strength of such worn casings (from an API burst

strength equation) with a linear reduction by the remaining wall thickness or the

wear percentage [Bradley, 1975a, 1975b]. This is equivalent to a “uniform-worn”

casing model, despite a question on whether such a linear reduction of casing

burst strength is over-conservative and may result in a higher casing cost.

Recent analytic and Finite Element Analysis (FEA) modeling indicated that a

local bending at the worn section would increase the maximum hoop stress on

worn casing (Figure 4-5), which would affect the estimation of worn casing burst

strength.

Figure 4-5. Worn Casing Deformation and Local Bending Under Internal Pressure

Worn casing burst strength can be estimated by a linear reduction starting from

casing full-yield (equation 4-19) or rupture burst strength by the remaining wall-

thickness for sweet service condition, which is supported by the recent worn

casing burst test data by Shell (Figure 4-6).

Casing/Tubing Design Manual 4-15

April 2007

9 7/8" Casing Burst Strength (0.619" wall, 134,880 psi yield

strength, 155,185 psi tensile strength)

5,000

8,000

11,000

14,000

17,000

20,000

23,000

0 102030405060

Casing wear, %

Casing burst strength, psi

Initial yield (100% wall)

Full yield (100% wall)

Rupture (100% wall)

Shell burst test data (100% wall)

Figure 4-6. Worn Casing Burst Strength Prediction and Comparison

For a sour service condition, the increased maximum hoop stress by local

bending at the wear would cause a reduction of the worn casing burst strength. A

linear reduction starting from API burst strength by the remaining wall-thickness

may be used to estimate the worn casing burst strength for a sour service

condition.

4.8.3 Effect of Temperature on Burst Strength

As tube material yield stress reduces under high temperature as discussed in

Chapter 3 Tube Specification, the tube burst resistance will reduce accordingly.

Therefore, it is necessary to design the production casing/liner and tubing taking

into account the temperature effect on tube burst resistance. This can occur in

high production temperatures.

4.8.4 Burst Strength Based on von Mises

Initial Yield

A slightly higher and more accurate tube burst (internal pressure) resistance than

the API rating is also available, thus requiring more sophisticated calculations

involving triaxial stresses analysis. This alternative tube burst performance is

also called the “tube VME burst strength” and is based on tube body initial yield

under multi-dimensional stresses [Wu, 2002]. The tube body initial yield is

referred to the tube body yield starting at the inner diameter. This tube VME burst

strength is formulated as:

875.0

⎟

⎠

⎞

⎜

⎝

⎛

−

⎟

⎠

⎞

⎜

⎝

⎛

−=

BAPI

(4-18)

ooiiAB

APAPTT

⋅

⋅−=

4-16 Casing/Tubing Design Manual

April 2007

(

)

iopY

AAYT −=

Where:

• Ai: internal diameter area, in.

• Ao: external diameter area, in.

• D: pipe OD, in.

• t: pipe wall thickness, in.

• Pi: internal pressure, psi

• Po: external pressure, psi

• T

: effective tension load, lb.

• T

: Actual tension load, lb.

• Y

: Tube material yield strength, psi

The factor 2/

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

()

Dt−1

results from the tube body initial-yield condition. For most tube, it gives a higher

burst rating than API burst rating. For extremely thick wall tubing, it may give a

slightly lower burst rating than the API burst rating.

The following graph shows the comparison between casing VME (Triaxial) burst

pressure and API burst pressure, for two particular sizes (weight) of tubes. The

tube can start the body yield under much lower internal pressure when the tube

is under axial compression load, while for most tube burst design (tube is under

tension load) conditions, the tube may resist slightly higher internal pressure

before the tube body yields.