Fatigue TNS

27/11/2006
An Introduction to Fatigue
Analysis
………………Rupesh Purwar
Fatigue Definition
Fatigue is the progressive, localized, and permanent
structural damage that occurs when a material is
subjected to cyclic or fluctuating stresses.
The resulting stress may be below the ultimate tensile
stress, or even the yield stress of the material.
Still cause catastrophic failure.
FATIGUE-Rupesh 2
Loads in Fatigue Analysis
Only cyclic or fluctuating loads to be cosidered

Wind
Ocean waves
Mechanical vibrations (Operating machines)
FATIGUE-Rupesh 3
Methods of analysis
Simplified Fatigue Analysis

Deterministic Fatigue Analysis
Spectral Fatigue Analysis
Spectral Fatigue Analysis is detailed method of analysis
is recommended for Category L-1(Manned-
nonevacuated) & Category L-2 (Manned-evacuated).
For Category L-3 (Unmanned) deterministic analysis
method can be used.
FATIGUE-Rupesh 4
Applicability of Simplified/Deterministic Fatigue Analysis:
As per API-RP 2A (Suppliment-2 Oct.2005) this method is

adopted for platforms that:
Are in water depth < 122m.
Are constructed in ductile steels.
Have a redundant, inspectable structural framing.
Have natural periods less than 3 seconds.
API-RP 2A (Suppliment-2 Oct.2005)

FATIGUE-Rupesh 5
Fatigue Analysis:
Input Data:
Wave data in each direction
•Wave height
•Period
•No. of Occurrences data
•Kinematics factor
•Cd, Cm coefficients
FATIGUE-Rupesh 6
1 wave
No. of actual cycles (n)

No. of cycles for particular wave ht.
Hot Spot Stress Range (S)
=(Nominal Stress X SCF) for each
1 Stress range (Max. & Min.) type of loading (Calculated
at 8 HS points)
Calculate N from Standard SN curve Basic API-RP 2A curve
We get N
Repeat for each wave height and wave direction
Damage factor, D= n/N
Life=1/D
SACS ALGORITHM
Calculation of Life
FATIGUE-Rupesh 7
Hot Spot Stress (HSS):
HSS is the stress in the immediate vicinity of a structural

discontinuity.
More specifically (mathematically)--- ”it is defined as the
linear trend of shell bending and membrane stress,
extrapolated to the actual weld toe”
HSS = Nominal Stress X SCF

SCF = Stress Concentration Factor
Nominal Stress: is the stress determined from member section properties and
the resultant forces and moments from global stress analysis at the member
end.

FATIGUE-Rupesh 8
Stress Concentration Factor (SCF):
The welds at tubular joints are among the most fatigue

sensitive areas in offshore platforms because of the high
local stress concentrations.
SCF = HSS / Nominal stress
The basis of this concept is to capture a stress (or strain) in

the proximity of weld toe which characterize the fatigue life
of joint.

FATIGUE-Rupesh 9
Shell Bending Stress
Membrane Stress
FATIGUE-Rupesh 10
FATIGUE-Rupesh 11
Factors affecting SCF :
Details of Geometry
Joint type
Loading
•Axial
•Inplane Bending
•Out of plane Bending
Also SCF should be calculated at 8 Hot Spot points.
FATIGUE-Rupesh 12
FATIGUE-Rupesh 13
Methods for calculation of SCF :
Empirical equations are given in API-RP 2A (pg. 208-211)
for -Different Loadings
- Different
locations (Chord/ Brace & Crown/Saddle)
Also some parametric equations are available such as:
•Wordsworth-Smedley
•Efthymiou
•Kuang
•Marshall
•Kellog
The method to be used depends on joint type.
HSE Offshore Technology Report-OTH 92 390
FATIGUE-Rupesh 14
SCF method for X-joints:
(Experimental Results)

FATIGUE-Rupesh 15
SCF method for K-joints:

FATIGUE-Rupesh 16
SCF method for TY-joints:

FATIGUE-Rupesh 17
Assessment of database & Equations
The equations were assessed by recording the ratio of the

predicted SCF to the recorded SCF (P/R) for each joint.
If,
P / R < 1.0 ……Under prediction
P / R > 1.5 ……Over prediction (Conservative)
1.5 < P / R < 1.0 ……Satisfactory/Recommended
From these studies it was concluded that Efthymiou and

Lloyd’s equations are most consistent.

FATIGUE-Rupesh 18
Basic standard SN curve:

FATIGUE-Rupesh 19
Derivation of Basic Design SN curve:
The best linear fit to a log10 (S) Vs log10 (N) relationship was
calculated using least mean square method and the equation
of the form:

FATIGUE-Rupesh 20
Why slope m=3 and m=5 ?
A preliminary assessment of test data showed that the slope of mean log10 (S) Vs
log10 (N) line had a value very close to 1/3 (i.e. m=3).
And it has 2.3% probability of failure.
A constant amplitude fatigue endurance limit could not exist in high cycle region
beyond 10^7 cycles. Hence to account for variable amplitude loading, the equation
takes a form:
FATIGUE-Rupesh 21
Data excluded in derivation of Basic SN curve:
1. Specimens with chord thickness < 16mm.

2. Specimens tested in seawater without cathodic protection
(i.e. Free corrosion).
3. Specimens subjected to weld improvement techniques.
4. Ring stiffened joints.
5. Grouted joints.
Hence,
Basic SN curve needs to be corrected for above exclusions
FATIGUE-Rupesh 22
Thickness effect in tubular joints:
Why 16mm is considered?
RPFG suggested that one approach should be used as base
line.
16 & 32mm thick tubular has consistency in test results.
16 thk covers widest range of joint geometries.
HSS needs to modify if t > 16mm

No thickness correction if t < 16 mm (Conservative)
RPFG = Review Panel for Fatigue Guidance
FATIGUE-Rupesh 23
Thickness correction in tubular joint

FATIGUE-Rupesh 24
Influence of Environment:
The basic curve is given for Joints in air and in water with CP
but not for free corrosion.
Free corrosion was more damaging to fatigue performance.
This is because:
Corrosion leads to pitting which acted as source of crack initiation
It accelerates crack growth
No increase in life should be considered for weld

improvements for free corrosion.
FATIGUE-Rupesh 25
Ring stiffened Joints:
Fatigue life of joint can be increased by welding rings to inside of the chord
under footprint of incoming braces.
Data of fatigue performance of tubular joints with internal ring stiffeners

was very limited.
Tests on ring stiffened joints showed that crack development was more
complex than in conventional unstiffened joints.
Reduction in chord ovalization & hence reduction in stresses increases the

fatigue life.
It is recommended not to use ring stiffeners at initial stage of analysis.

FATIGUE-Rupesh 26
Grouted Joints:
Effective thickness can be calculated by any of

following method (SACS)
SACS MANUAL
FATIGUE-Rupesh 27
SACS MANUAL
FATIGUE-Rupesh 28
SACS MANUAL
FATIGUE-Rupesh 29
Comparison of API-RP 2A
supplement 1 & supplement 2:
 SN CURVE  SN CURVE
FATIGUE-Rupesh 30
Derivation of SN curve:
FATIGUE-Rupesh 31
Life Safety Factors :
FATIGUE-Rupesh 32
 The Basic SN curve is  SCF is modified for
modified for weld weld improvements
improvements.
 Thickness correction  Thickness correction
curve is generated for curve is depend on tau
25 mm thk. pipe
FATIGUE-Rupesh 33
Points to be noted
1. Minimum SCF of 1.5 should be used for any joint type.
2. Ring stiffeners should not be used at initial stage of
analysis (if used min. SCF = 6.0).
3. Avoid joints in Splash zone.
4. Adequate cathodic protection should be provided to
the structure in sea water.
5. Weld profiling is more effective when there is brace
side failure.
6. External ring stiffener are not recommended.
7. Min. gap between members should be maintained.
FATIGUE-Rupesh 34
8. Favor X bracing instead of K bracing (Increases
Ductility)
9. To increase life of joint
- Reduce stress concentrations
- Remove weld toe defects.
10. If possible avoid cantilevers in framing (conductor
framing).
FATIGUE-Rupesh 35
THANK YOU
SPECIAL THANKS TO:

B D Ravikumar
Milind Salunke
FATIGUE-Rupesh 36

Fatigue TNS

Uploaded by

Copyright:

Available Formats

Fatigue TNS

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fatigue TNS

Uploaded by

Copyright:

Available Formats

27/11/2006

Only cyclic or fluctuating loads to be cosidered

Simplified Fatigue Analysis

As per API-RP 2A (Suppliment-2 Oct.2005) this method is

API-RP 2A (Suppliment-2 Oct.2005)

No. of actual cycles (n)

Calculate N from Standard SN curve Basic API-RP 2A curve

Damage factor, D= n/N

HSS is the stress in the immediate vicinity of a structural

HSS = Nominal Stress X SCF

API-RP 2A (Suppliment-2 Oct.2005)

The welds at tubular joints are among the most fatigue

SCF = HSS / Nominal stress

The basis of this concept is to capture a stress (or strain) in

API-RP 2A (Suppliment-2 Oct.2005)

Also SCF should be calculated at 8 Hot Spot points.

HSE Offshore Technology Report-OTH 92 390

HSE Offshore Technology Report-OTH 92 390

HSE Offshore Technology Report-OTH 92 390

The equations were assessed by recording the ratio of the

From these studies it was concluded that Efthymiou and

HSE Offshore Technology Report-OTH 92 390

API-RP 2A (Suppliment-2 Oct.2005)

API-RP 2A (Suppliment-2 Oct.2005)

1. Specimens with chord thickness < 16mm.

HSS needs to modify if t > 16mm

RPFG = Review Panel for Fatigue Guidance

API-RP 2A (Suppliment-2 Oct.2005)

No increase in life should be considered for weld

Data of fatigue performance of tubular joints with internal ring stiffeners

Reduction in chord ovalization & hence reduction in stresses increases the

It is recommended not to use ring stiffeners at initial stage of analysis.

Effective thickness can be calculated by any of

SPECIAL THANKS TO:

You might also like