API RP 2A-WSD-2007 Recommended Practice for Planning, Designing and Constructing Fixed Offshore Platforms-Working Stress Design

Подождите немного. Документ загружается.

RECOMMENDED PRACTICE FOR PLANNING, DESIGNING AND CONSTRUCTING FIXED OFFSHORE PLATFORMS—WORKING STRESS DESIGN 49

Figure 4.2-2—In-Plane Joint Detailing

See Section 3.4

D/4 or 12 in.

(300 mm) min.

Se a m

weld

or 24 in.

(600 mm) min.

or 24 in.

(600 mm) min.

or 24 in.

(600 mm) min.

/4 or 6 in.

(150 mm) min.

See Section 3.4

Gap 2 in.

(50 mm) min.

Can girth weld

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---

50 API RECOMMENDED PRACTICE 2A-WSD

Figure 4.2-3—Out-of-Plane Joint Detailing

D/4 max.

or 24 in.

(600 mm) min.

/4 or 6 in.

(150 mm) min.

12 in. (300 mm) min.

Gap 2 in. (50 mm) min.

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---

RECOMMENDED PRACTICE FOR PLANNING, DESIGNING AND CONSTRUCTING FIXED OFFSHORE PLATFORMS—WORKING STRESS DESIGN 51

The minimum nominal gap between adjacent braces,

whether in- or out-of-plane, is 2 inches (50 mm). Care should

be taken to ensure that overlap of welds at the toes of the joint

is avoided. When overlapping braces occur, the amount of

overlap should preferably be at least d/4 (where d is the diam-

eter of the through brace) or 6 inches (150 mm), whichever is

greater. This dimension is measured along the axis of the

through member.

Where overlapping of braces is necessary or preferred, and

which differ in nominal thickness by more than 10% the

brace with the larger wall thickness should be the through

brace and be fully welded to the chord. Further, where sub-

stantial overlap occurs, the larger diameter brace should be

specified as the through member. This brace may require an

end stub to ensure that the thickness is at least equal to that of

the overlapping brace.

Longitudinal seam welds and girth welds should be located

to minimize or eliminate their impact on joint performance.

The longitudinal seam weld of the chord should be separated

from incoming braces by at least 12 inches (300 mm), see

Figure 4.2-3. The longitudinal seam weld of a brace should

be located near the crown heel of the joint. Longer chord cans

may require a girth weld. This weld should be positioned at a

lightly loaded brace intersection, between saddle and crown

locations, see Figure 4.2-2.

4.3 SIMPLE JOINTS

4.3.1 Validity Range

The terminology for simple joints is defined in Figure

4.3-1.

The validity range for application of the practice defined in

4.3 is as follows:

0.2 ≤β ≤ 1.0

10 ≤γ ≤ 50

30°≤ θ ≤ 90°

≤ 72 ksi (500 MPa)

g/D > -0.6 (for K joints)

The Commentary discusses approaches that may be

adopted for joints that fall outside the above range.

4.3.2 Basic Capacity

Tubular joints without overlap of principal braces and hav-

ing no gussets, diaphragms, grout or stiffeners should be

designed using the following guidelines.

(4.3-1a)

(4.3-1b)

(plus

increase in both cases where applicable)

where:

= allowable capacity for brace axial load,

= allowable capacity for brace bending moment,

= the yield stress of the chord member at the joint (or

0.8 of the tensile strength, if less), ksi (MPa),

FS = safety factor = 1.60.

For joints with thickened cans, P

shall not exceed the

capacity limits defined in 4.3.5.

For axially loaded braces with a classification that is a mix-

ture of K, Y and X joints, take a weighted average of P

based on the portion of each in the total load.

4.3.3 Strength Factor Q

varies with the joint and load type, as given in Table

4.3-1.

Where the working points of members at a gap connection

are separated by more than D/4 along the chord centerline, or

where a connection has simultaneously loaded branch mem-

bers in more than one plane, the connection may be classified

as a general or multi-planar connection, and designed as

described in the Commentary.

4.3.4 Chord Load Factor Q

is a factor to account for the presence of nominal loads

in the chord.

(4.3-2)

The parameter A is defined as follows:

(4.3-3)

(Where

increase applicable, FS = 1.20 in 4.3-2 and 4.3-3.)

Where P

and M

are the nominal axial load and bending

resultant (i.e., ) in the chord,

is the yield axial capacity of the chord,

is the plastic moment capacity of the chord, and

, C

and C

are coefficients depending on joint and

load type as given in Table 4.3-2.

The average of the chord loads and bending moments on

either side of the brace intersection should be used in Equa-

tions 4.3-2 and 4.3-3. Chord axial load is positive in tension,

chord in-plane bending moment is positive when it produces

compression on the joint footprint. The chord thickness at the

joint should be used in the above calculations.

FS θsin

------------------=

FS θsin

------------------=

1 C

FSP

------------

⎝⎠

⎛⎞

FSM

ipb

-----------------

⎝⎠

⎛⎞

– C

–+=

FSP

------------

⎝⎠

⎛⎞

FSM

--------------

⎝⎠

⎛⎞

0.5

ipb

= M

opb

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---

52 API RECOMMENDED PRACTICE 2A-WSD

Statistics are presented in the Commentary, to permit both

the estimation of mean strength and the conduct of reliability

analyses.

4.3.5 Joints with Thickened Cans

For simple, axially loaded Y and X joints where a thick-

ened joint can is specified, the joint allowable capacity may

be calculated as follows:

= [r + (1 – r) (T

/ T

)

] (P

)

(4.3-4)

where

)

= P

from Equation 4.3-1a based on chord can

geometric and material properties, including Q

calculated with respect to chord can,

= nominal chord member thickness,

= chord can thickness,

r = L

/ (2.5 D) for joints with β ≤ 0.9

=(4β - 3) L

/ (1.5 D) for joints with β >0.9,

= effective total length. Figure 4.3-2 gives exam-

ples for calculation of L

In no case shall r be taken as greater than unity.

Alternatively, an approximate closed ring analysis may be

employed, including plastic analysis with appropriate safety

factors, using an effective chord length up to 1.25D either side

of the line of action of the branch loads at the chord face, but

not more than actual distance to the end of the can. Special con-

sideration is required for more complex joints. For multiple

branches in the same plane, dominantly loaded in the same

sense, the relevant crushing load is

Sin

. Any reinforce-

ment within this dimension (e.g., diaphragms, rings, gussets or

the stiffening effect of out of plane members) may be consid-

ered in the analysis, although its effectiveness decreases with

distance from the branch footprint.

4.3.6 Strength Check

The joint interaction ratio, IR, for axial loads and/or bending

moments in the brace should be calculated using the following

expression:

(4.3-5)

4.4 OVERLAPPING JOINTS

Braces that overlap in- or out-of-plane at the chord member

form overlapping joints. Examples are shown in Figures 4.2-2

and 4.2-3.

Joints that have in-plane overlap involving two or more

braces in a single plane (e.g., K and KT joints), may be

designed using the simple joint provisions of 4.3, using nega-

tive gap in Q

, with the following exceptions and additions:

Figure 4.3-1—Terminology and Geometric Parameters, Simple Tubular Joints

Q = Brace included angle

g = Gap between braces, in. (mm)

t = Brace wall thickness at intersection, in. (mm)

T = Chord wall thickness at intersection, in. (mm)

d = Brace outside diameter, in. (mm)

D = Chord outside diameter, in. (mm)

B =

G =

T =

Brace

Crown toe

Crown heel

Saddle

Chord

-----

-------

⎝⎠

⎛⎞

ipb

-------

opb

++ 1.0≤

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---

RECOMMENDED PRACTICE FOR PLANNING, DESIGNING AND CONSTRUCTING FIXED OFFSHORE PLATFORMS—WORKING STRESS DESIGN 53

a. Shear parallel to the chord face is a potential failure mode

and should be checked.

b. Section 4.3.5 does not apply to overlapping joints with bal-

anced loads.

c. If axial forces in the overlapping and through braces have

the same sign, the combined axial force representing that in

the through brace plus a portion of the overlapping brace

forces should be used to check the through brace intersec-

tion capacity. The portion of the overlapping brace force can

be calculated as the ratio of cross sectional area of the brace

that bears onto the through brace to the full area.

d. For either in-plane or out-of-plane bending moments, the

combined moment of the overlapping and through braces

Table 4.3-1—Values for Q

Joint

Classification

Brace Load

Axial

Tension

Axial

Compression

In-Plane

Bending Out-of-Plane Bending

(16 + 1.2γ) β

1.2

but ≤ 40 β

1.2

(5 + 0.7γ)β

1.2

2.5 + (4.5 + 0.2γ)β

2.6

T/Y 30β

2.8 + (20 + 0.8γ)β

1.6

but ≤ 2.8 + 36 β

1.6

23β for β ≤ 0.9

20.7 + (β – 0.9)(17γ – 220) for β > 0.9

[2.8 + (12 + 0.1γ)β]Q

Notes:

(a) Q

is a geometric factor defined by:

=for β >0.6

= 1.0 for β ≤ 0.6

(b) Q

is the gap factor defined by:

= 1 + 0.2 [1 – 2.8 g/D]

for g/D ≥ 0.05

but ≥ 1.0

= 0.13 + 0.65 φ γ

0.5

for g/D ≤ -0.05

where φ = t F

/(TF

)

The overlap should preferably not be less than 0.25βD. Linear interpolation between the limiting values of the above two Q

expressions may be used for –0.05 < g/D < 0.05 when this is otherwise permissible or unavoidable. See Commentary C4.3.3.

= yield stress of brace or brace stub if present (or 0.8 times the tensile strength if less), ksi (MPa)

term for tension loading is based on limiting the capacity to first crack. The Q

associated with full ultimate capacity of

tension loaded Y and X joints is given in the Commentary.

(d) The X joint, axial tension, Q

term for β > 0.9 applies to coaxial braces (i.e., e/D ≤ 0.2 where e is the eccentricity of the two

braces). If the braces are not coaxial (e/D > 0.2) then 23β should be used over the full range of β.

0.3

β 10.833β–()

----------------------------------

Table 4.3-2—Values for C

, C

Joint Type C

K joints under brace axial loading 0.2 0.2 0.3

T/Y joints under brace axial loading 0.3 0 0.8

X joints under brace axial loading*

β ≤ 0.9

β = 1.0

0.2

-0.2

0.5

0.2

All joints under brace moment loading 0.2 0 0.4

*Linearly interpolated values between β = 0.9 and β = 1.0 for X

joints under brace axial loading.

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---

54 API RECOMMENDED PRACTICE 2A-WSD

should be used to check the through brace intersection

capacity. This combined moment should account for the

sign of the moments. Where combined nominal axial and

bending stresses in the overlapping brace peak in the over-

lap region, the overlapping brace should also be checked on

the basis of its chord being the through brace, using Q

1.0. That is, through brace capacity should be checked for

combined axial and moment loading in the overlapping

brace. In this instance the Q

associated with the through

brace should be used.

Joints having out-of-plane overlap may be assessed on the

same general basis as in-plane overlapping joints, with the

exception that axial load capacity may be calculated as for

multi-planar joints in Commentary C4.3.3.

4.5 GROUTED JOINTS

Two varieties of grouted joints commonly occur in prac-

tice. The first relates to a fully grouted chord. The second is

the double-skin type, where grout is placed in the annulus

between a chord member and an internal member. In both

cases, the grout is unreinforced and, as far as joint behavior is

concerned, benefit for shear keys that may be present is not

permitted.

For grouted joints that are otherwise simple in configuration,

the simple joint provisions defined in Section 4.3 may be used

with the following modifications and limitations:

a. For fully grouted and double-skin joints, the Q

values in

Table 4.3-1 may be replaced with the values pertinent to

grouted joints given in Table 4.5-1. Classification and joint

can derating may be disregarded. The adopted Q

values

should not be less than those for simple joints.

b. For double-skin joints, failure may also occur by chord

ovalization. The ovalization capacity can be estimated by

substituting the following effective thickness into the simple

joint equations:

(4.5-1)

Table 4.5-1—Q

for Grouted Joints

Brace Load Q

Axial tension

2.5 β γ K

where

Bending 1.5 β γ

Note that no term is provided for axial compression since most

grouted joints cannot fail under compression; compression capacity

is limited by that of the brace.

---

⎝⎠

⎛⎞

11 θsin⁄+()=

a b

Chord can

Nominal

chord

Brace 3

Brace 2

Brace 1

Brace

Length, L

2a + d

2b + d

/sin Q

2c + d

Figure 4.3-2—Examples of Chord Length, L

+()

0.5

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---

RECOMMENDED PRACTICE FOR PLANNING, DESIGNING AND CONSTRUCTING FIXED OFFSHORE PLATFORMS—WORKING STRESS DESIGN 55

where

= effective thickness, in. (mm),

T = wall thickness of chord, in. (mm),

= wall thickness of inner member, in. (mm).

should be used in place of T in the simple joint equations,

including the γ term.

c. The Q

calculation for both fully grouted and double-

skinned joints should be based on T; it is presumed that cal-

culation of Q

has already accounted for load sharing

between the chord and inner member, such that further con-

sideration of the effect of grout on that term is unnecessary.

However, for fully grouted joints, Q

may normally be set

to unity, except in the instance of high β (≥ 0.9) X joints

with brace tension/OPB and chord compression/OPB.

d. The minimum capacity requirements of 4.2.3 should still

be observed.

4.6 INTERNALLY RING-STIFFENED JOINTS

Primary joints along launch trusses of steel jacket structures

are often strengthened by internal ring stiffening. Internal stiff-

ening is also used in some structures to address fatigue require-

ments or to avoid very thick chord cans.

The Commentary outlines the salient features of several

common approaches to the design of internally ring-stiffened

joints.

4.7 CAST JOINTS

Cast joints are defined as joints formed using a casting pro-

cess. They can be of any geometry and of variable wall thick-

ness.

The design of a cast joint requires calibrated finite element

analyses. An acceptable design approach for strength is to limit

stresses at all locations in the joint due to nominal loads to

below yielding of the material using appropriate yield criteria

with a 1.6 safety factor. Such an approach can be quite conser-

vative when compared to welded joints, which are designed on

the basis of overall ultimate behavior.

Often, the manufacturer of the cast joint carries out the

design process.

4.8 OTHER CIRCULAR JOINT TYPES

Joints not covered by 4.3 to 4.7 may be designed on the

basis of appropriate experimental, numerical or in-service

evidence. Strength-of-materials approaches may be

employed although extreme care is needed in identifying

all elements that are expected to participate in resisting

incoming brace loads, and in establishing the acting load

envelopes prior to conducting strength checks. Often,

strength-of-materials checks are complemented with cali-

brated FE analyses to establish the magnitude and location

of acting stresses.

4.9 DAMAGED JOINTS

Joints in existing installations could be damaged as a result

of fatigue loading, corrosion or overload (environmental or

accidental). In such cases, the reduced joint capacity can be

estimated either by simple models (e.g., reduced area or

reduced section modulus approaches), calibrated numerical

(FE) models, or experimental evidence.

4.10 NON-CIRCULAR JOINTS

Connections with non-circular chord and/or brace sections

are typically used on topside structures. Common types

include wide flange (I beam, column, plate girder) sections

and rectangular/square sections. For some arrangements,

detailed land-based design practice is available. For arrange-

ments for which little or no practice is available, the provi-

sions noted in Section 4.8 apply.

5 Fatigue

5.1 FATIGUE DESIGN

In the design of tubular connections, due consideration

should be given to fatigue action as related to local cyclic

stresses.

A detailed fatigue analysis should be performed for all

structures, except as provided below. It is recommended that

a spectral analysis technique be used. Other rational methods

may be used provided adequate representation of the forces

and member responses can be demonstrated.

In lieu of detailed fatigue analysis, simplified fatigue anal-

yses, which have been calibrated for the design wave climate,

may be applied to tubular joints in Category L-3 template

type platforms as defined in Section 1.7 that:

1. Are constructed of notch-tough ductile steels.

2. Have redundant, inspectable structural framing.

3. Have natural periods less than 3 seconds.

Such simplified methods are particularly useful for prelim-

inary design of all structure categories and types, in water

depths up to 400 feet (122 m). These are described in the

Commentary. Caissons, monopods, and similar non-jacket

structures deserve detailed analysis, with consideration of

vortex shedding where applicable.

5.2 FATIGUE ANALYSIS

A detailed analysis of cumulative fatigue damage, when

required. should be performed as follows:

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---

56 API RECOMMENDED PRACTICE 2A-WSD

5.2.1 The wave climate should be derived as the aggre-

gate of all sea states to be expected over the long term. This

may be condensed for purposes of structural analysis into

representative sea states characterized by wave energy spec-

tra and physical parameters together with a probability of

occurrence.

5.2.2 A space frame analysis should be performed to

obtain the structural response in terms of nominal member

stress for given wave forces applied to the structure. In gen-

eral, wave force calculations should follow the procedures

described in Section 2.3.1. However, current may be

neglected and. therefore, considerations for apparent wave

period and current blockage are not required. In addition,

wave kinematics factor equal to 1.0 and conductor shielding

factor equal to 1.0 should be applied for fatigue waves. The

drag and inertia coefficients depend on the sea state level, as

parameterized by the Keulegan-Carpenter Number K (see

Commentary C2.3.lb7). For small waves (1.0 < K < 6.0 for

platform legs at mean water level), values of C

= 2.0, C

0.8 for rough members and C

= 0.5 for smooth members

should be used. Guidelines for considering directionality,

spreading, tides and marine growth are provided in the com-

mentary for this section.

A spectral analysis technique should be used to determine

the stress response for each sea state. Dynamic effects should

be considered for sea states having significant energy near a

platform's natural period.

5.2.3 Local stresses that occur within tubular connections

should be considered in terms of hot spot stresses located

immediately adjacent to the joint intersection using suitable

stress concentration factors. The micro scale effects occurring

at the toe of the weld are reflected in the appropriate choice of

the S-N curve.

5.2.4 For each location around each member intersection

of interest in the structure, the stress response for each sea

state should be computed, giving adequate consideration to

both global and local stress effects.

The stress responses should be combined into the long

term stress distribution, which should then be used to calcu-

late the cumulative fatigue damage ratio, D, where

D = ∑ (n/N) (5.2.4-1)

where

n = number of cycles applied at a given stress

range,

N = number of cycles for which the given stress

range would be allowed by the appropriate S-N

curve.

Alternatively, the damage ratio may be computed for each

sea state and combined to obtain the cumulative damage

ratio.

5.2.5 In general the design fatigue life of each joint and

member should not be less than the intended service life of

the structure multiplied by a Safety Factor. For the design

fatigue life, D, should not exceed unity.

For in-situ conditions, the safety factor for fatigue of steel

components should depend on the failure consequence (i.e.

criticality) and in-service inspectability. Critical elements are

those whose sole failure could be catastrophic. In lieu of a

more detailed safety assessment of Category L-1 structures, a

safety factor of 2.0 is recommended for inspectable, non-fail-

ure critical, connections. For failure-critical and/or non-

inspectable connections, increased safety factors are recom-

mended, as shown in Table 5.2.5-1. A reduced safety factor is

recommended for Category L-2 and L-3 conventional steel

jacket structures on the basis of in-service performance data:

SF=1.0 for redundant diver or ROV inspectable framing, with

safety factors for other cases being half those in the table.

When fatigue damage can occur due to other cyclic load-

ings, such as transportation, the following equation should be

satisfied:

< 1.0 (5.2.5-1)

where

= the fatigue damage ratio for each type of loading,

= the associated safety factor.

For transportation where long-term wave distributions are

used to predict short-term damage a larger safety factor

should be considered.

5.3 STRESS CONCENTRATION FACTORS

5.3.1 General

The welds at tubular joints are among the most fatigue sen-

sitive areas in offshore platforms because of the high local

stress concentrations. Fatigue lives at these locations should

be estimated by evaluating the Hot Spot Stress Range

(HSSR) and using it as input into the appropriate S-N curve

from Section 5.5.

Table 5.2.5-1—Fatigue Life Safety Factors

Failure critical Inspectable Not Inspectable

No 2 5

Yes 5 10

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

No reproduction or networking permitted without license from IHS

RECOMMENDED PRACTICE FOR PLANNING, DESIGNING AND CONSTRUCTING FIXED OFFSHORE PLATFORMS—WORKING STRESS DESIGN 57

For each tubular joint configuration and each type of brace

loading, the SCF is defined as:

SCF = HSSR / Nominal Brace Stress Range (5.3.1-1)

The Nominal Brace Stress Range should be based on the

section properties of the brace-end under consideration, tak-

ing due account of the brace-stub, or a flared member end, if

present. Likewise, the Stress Concentration Factor (SCF)

evaluation shall be based on the same section dimensions.

Nominal cyclic stress in the chord may also influence the

HSSR and should be considered; see Commentary.

The SCF should include all stress raising effects associated

with the joint geometry and type of loading, except the local

(microscopic) weld notch effect, which is included in the S-N

curve. SCFs may be derived from Finite Element analyses,

model tests or empirical equations based on such methods. In

general. the SCFs depend on the type of brace cyclic loading

(i.e. brace axial load, in-plane bending, out-of plane bending),

the joint type, and details of the geometry. The SCF varies

around the joint, even for a single type of brace loading.

When combining the contributions from the various loading

modes, phase differences between them should be accounted

for, with the design HSSR at each location being the range of

hotspot stress resulting from the point-in-time contribution of

all loading components.

For all welded tubular joints under all three types of load-

ing, a minimum SCF of 1.5 should be used.

5.3.2 SCFs in Unstiffened Tubular Joints

For unstiffened welded tubular joints, SCFs should be

evaluated using the Efthymiou equations; see Commentary.

The linearly extrapolated hot spot stress from Efthymiou

may be adjusted to account for the actual weld toe position,

where this systematically differs from the assumed AWS

basic profiles; see Commentary.

For the purpose of computing SCF, the tubular joints are

typically classified into types T/Y, X, K, and KT depending

on the joint configuration, the brace under consideration and

the loading pattern. As a generalization of the classification

approach, the Influence Function algorithm discussed in the

Commentary may be used to evaluate the hot spot stress

ranges. This algorithm can handle generalized loads on the

braces. Moreover the Influence Function algorithm can han-

dle multi-planar joints for the important case of axial loading.

The Commentary contains a discussion on tubular joints

welded from one side.

5.3.3 SCFs in Internally Ring-Stiffened Tubular

Joints

The SCF concept also applies to internally ring stiffened

joints, including the stresses in the stiffeners and the stiffener

to-chord weld. Ring-stiffened joints may have stress peaks at

the brace-ring intersection points. Special consideration

should be given to these locations. SCFs for internally ring-

stiffened joints can be determined by applying the Lloyds

reduction factors to the SCFs for the equivalent unstiffened

joint, see Commentary. For ring-stiffened joints analyzed by

such means, the minimum SCF for the brace side under axial

or OPB loading should be taken as 2.0.

Ring stiffeners without flanges on the internal rings should

consider high stress that may occur at the inner edge of the

ring.

5.3.4 SCFs in Grouted Joints

Grouting tends to reduce the SCF of the joint since the

grout reduces the chord deformations. In general, the larger

the ungrouted SCF, the greater the reduction in SCF with

grouting. Hence, the reductions are typically greater for X

and T joints than for Y and K joints. The Commentary dis-

cusses approaches for calculating SCFs for grouted joints.

5.3.5 SCFs in Cast Nodes

For cast joints, the SCF is derived from the maximum

principal stress at any point on the surface of the casting

(including the inside surface) divided by the nominal brace

stress outside the casting. The SCFs for castings are not

extrapolated values, but are based on directly measured or

calculated values at any given point, using an analysis that is

sufficiently detailed to pick up the local notch effects of fillet

radii, etc. Consideration should also be given to the brace-to

casting girth weld, which can be the most critical location for

fatigue.

5.4 S-N CURVES FOR ALL MEMBERS AND

CONNECTIONS, EXCEPT TUBULAR

CONNECTIONS

Non-tubular members and connections in deck structures,

appurtenances and equipment; and tubular members and

attachments to them, including ring stiffeners, may be subject

to variations of stress due to environmental loads or opera-

tio

nal loads. Operational loads would include those associ-

ated with machine vibration, crane usage and filling and

emptying of tanks. Where variations of stress are applied to

conventional weld details, identified in ANSI/AWS D1.1-

2002 Table 2.4, the associated S-N curves provided in AWS

Figure 2.11 should be used, dependent on degree of redun-

dancy. Where such variations of stress are applied to tubular

nominal stress situations identified in ANSI/AWS D1.1-2002

Table 2.6, the associated S-N curves provided in AWS Figure

2.13 should be used. Stress Categories DT, ET, FT, Kl, and

K2, refer to tubular connections where the SCF is not known.

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---

58 API RECOMMENDED PRACTICE 2A-WSD

Where the hot spot stress concentration factor can be deter-

mined, Sections 5.3 and 5.5 of this Recommended Practice

take precedence

For service conditions where details may be exposed to

random variable loads, seawater corrosion, or submerged ser-

vice with effective cathodic protection, see Commentary.

The referenced S-N curves in ANSI/AWS D1.1.-2002 Fig-

ure 2.11 are Class curves. For such curves, the nominal stress

range in the vicinity of the detail should be used. Due to load

attraction, shell bending, etc., not present in the class type test

specimens, the appropriate stress may be larger than the nom-

inal stress in the gross member. Geometrical stress concentra-

tion and notch effects associated with the detail itself are

included in the curves.

For single-sided butt welds, see Commentary.

Reference may alternatively be made to the S-N criteria

similar to the OJ curves contained within ISO DIS

19902:2004 Clause 16.11. The ISO code proposal uses a

weld detail classification system whereby the OJ curves

include an allowance for notch stress and modest geometrical

stress concentration.

5.5 S-N CURVES FOR TUBULAR CONNECTIONS

5.5.1 Basic S-N curves

Design S-N curves are given below for welded tubular and

cast joints. The basic design S-N curve is of the form:

Log

(N) = Log

) – m Log

(S) (5.4.1-1)

where

N = the predicted number of cycles to failure under

stress range S,

=a constant,

m = the inverse slope of the S-N curve.

Table 5.5.1-1 presents the basic WJ and CJ curves. These

S-N curves are based on steels with yield strength less than

72 ksi (500 MPa).

For welded tubular joints exposed to random variations of

stress due to environmental or operational loads, the WJ

curve should be used. The brace-to-chord tubular intersection

for ring-stiffened joints should be designed using the WJ

curve. For cast joints the CJ curve should be used. For other

details, including plated joints and, for ring-stiffened joints,

the ring stiffener-to-chord connection and the ring inner edge,

see 5.4.

The basic allowable cyclic stress should be corrected

empirically for seawater effects, the apparent thickness effect

(per 5.5.2, with exponent depending on profile), and the weld

improvement factor on S per 5.5.3. An example of S-N curve

construction is given in Figure 5.5-1.

The basic design S-N curves given in Table 5.5.1-1 are

applicable for joints in air and submerged coated joints. For

Welded Joints in seawater with adequate cathodic protection,

the m = 3 branch of the S-N curve should be reduced by a

factor of 2.0 on life, with the m = 5 branch remaining

unchanged and the position of the slope change adjusted

accordingly. Plots of the WJ curves versus data, and informa-

tion concerning S-N curves for joints in seawater without

adequate corrosion protection is given in the Commentary.

Fabrication of welded joints should be in accordance with

Section 11. The curve for cast joints is only applicable to

castings having an adequate fabrication inspection plan; see

Commentary.

5.5.2 Thickness effect

The WJ curve is based on

-in. (16 mm) reference thick-

ness. For material thickness above the reference thickness,

the following thickness effect should be applied for as-

welded joints:

S = S

ref

/t)

0.25

(5.5.2-1)

where

ref

= the reference thickness,

-inch (16 mm), and

S = allowable stress range,

= the allowable stress range from the S-N curve,

t = member thickness for which the fatigue life is

predicted.

If the weld has profile control as defined in 11.1.3d, the

exponent in the above equation may be taken as 0.20. If the

weld toe has been ground or peened, the exponent in the

above equation may be taken as 0.15.

The material thickness effect for castings is given by:

S = S

ref

/t)

0.15

(5.5.2-2)

where the reference thickness t

ref

is 1.5 in (38 mm).

Table 5.5.1-1—Basic Design S-N Curves

Curve

log

)

S in ksi

log

)

S in MPa m

Welded Joints (WJ) 9.95

11.92

12.48

16.13

3 for N < 10

5 for N > 10

Cast Joints (CJ) 11.80

13.00

15.17

17.21

4 for N < 10

5 for N > 10

Provided by IHS under license with API

Licensee=Indonesia location/5940240008

Not for Resale, 10/22/2008 00:07:12 MDT

--`,,```,,,`,,,,,,,,,,,,,,`,``,`-`-`,,`,,`,`,,`---