4.1 Basic S-N curves
4.1.1 Capacity
The capacity of welded steel joints and steel base material with respect to
fatigue strength is defined by S-N curves which provide the relationship between the
stress range applied to the detail and the number of constant amplitude load cycles to
failure.
4.1.2 Design S-N curves
The fatigue assessment is based on use of S-N curves which are obtained from
fatigue tests. The design S-N curves are established at two standard deviations below
the mean S-N curves corresponding to 50% of probability of survival for relevant
experimental data. Design S-N curves given in Table 2 and Table 3 correspond to a
probability of survival of 97.7%.
Table 1 : Welded joints: thickness
exponents
|
No
|
Joint category description
|
Geometry
|
Condition
|
n
|
|
1
|
Cruciform joints, transverse T-joints, plates with
transverse attachments
|
|
As-welded
|
0.25
|
|
Weld toetreatedby post-weld improvement method
|
0.2
|
|
2
|
Transverse butt welds
|
|
As-welded
|
0.2
|
| Ground flush or weld toe treated by post-weld
improvement method
|
0.1
|
|
3
|
Longitudinal welds or attachments to plate edges
|
|
Any
|
0.1
|
|
Weld toe treated by post-weld improvement method
|
0.1
|
|
4
|
Longitudinal attachments on the flat bar or bulb
profile
|
|
Any
|
0
|
| Weld toe treated by post-weld improvement method
(1)
|
0
|
|
5
|
Longitudinal attachments and doubling plates
|
|
As-welded
|
0.2
|
|
Weld toetreated by post-weld improvement method
|
0.1
|
|
6
|
Longitudinal attachments and doubling plates supported
longitudinally
|
|
As-welded
|
0.1
|
|
Weld toetreated by post-weld improvement method
(1)
|
0
|
| (1) No benefit applicable
for post-weld treatment of longitudinal end connections.
|
4.1.3 S-N curve scope of application
The S-N curves are applicable to normal and high strength steels up to a
specified minimum yield stress equal to 390 N/mm2.
4.1.4 In-air environment
The basic design curves in-air environment shown in Figure 3 are represented
by linear relationships between log (Δσ) and log (N) as follows:
log (N) = log (K2) – m ⋅ log (Δσ)
where:
l log(K2) = log (K1) − 2 ⋅
log(δ).
K1 : Constant related to mean S-N curve, as given in Table
2.
K2 : Constant related to design S-N curve, as given in
Table 2.
δ : Standard deviation of log (N), as given in Table 2.
Δσq : Stress range at N = 107 cycles
related to design S-N curve, in N/mm2, as given in Table 2.
Table 2 : Basic S-N curve data,
in-air environment
|
Class
|
K1
|
m
|
Standard deviation
δ
|
K2
|
Design stress range at
107cycles
|
Design stress range at
2×106cycles
|
| K1
|
log10K1
|
log10δ
|
K2
|
ΔσqN/mm2
|
N/mm2
|
| B
|
2.343E15
|
15.3697
|
4.0
|
0.1821
|
1.01E15
|
100.2
|
149.9
|
| C
|
1.082E14
|
14.0342
|
3.5
|
0.2041
|
4.23E13
|
78.2
|
123.9
|
| D
|
3.988E12
|
12.6007
|
3.0
|
0.2095
|
1.52E12
|
53.4
|
91.3
|
Figure 3 : Basic design S-N curves, in-air environment
4.1.5 Corrosive environment
The basic design curves for corrosive environment shown in Figure 4 are
represented by linear relationships between log(Δσ) and log(N) as follows:
log (N) = log (K2) – m ⋅ log (Δσ)
N : Predicted number of cycles to failure under stress range Δσ.
K2 : Constant related to design S-N curve as given in Table
3.
Table 3 : Basic S-N curve data,
corrosive environment
| Class
|
K2
|
m
|
Design stress range at 2×106 cycles,
N/mm2
|
| Bcorr
|
5.05 x 1014
|
4.0
|
126.1
|
| Ccorr
|
2.12 x 1013
|
3.5
|
101.6
|
| Dcorr
|
7.60 x 1011
|
3.0
|
72.4
|
Figure 4 : Basic design S-N curves, corrosive environment
4.2 Selection of S-N curves
4.2.1 Welded joints
For fatigue assessment of welded joints exposed to in-air environment, S-N
curve D as defined in Table 2 is to be used. For corrosive environment, S-N curve
Dcorr as defined in Table 3 is to be used.
4.2.2 Base material free edge
For fatigue assessment of base material at free edge exposed to in-air
environment, S-N curves B or C as defined in Table 2 are to be used. For corrosive
environment, S-N curves Bcorr or Ccorr as defined in
Table 3 are to be used.
4.2.3 Surface finishing factor
The S-N curve C is applicable to most of non-welded locations taking
into account the likelihood of some notching from corrosion, wear and tear in service
with surface finishing factor as given in Table 4.
Higher surface finishing quality may be applied in using S-N curve B as given
in Table 4, provided adequate protective measures are taken against wear, tear and
corrosion and finite element analysis according to Ch 9, Sec 5, [2] is carried out.
Table 4 : Non-welded joints:
thickness exponent and surface finishing factor
| Joint configuration,
fatigue crack location and stress direction
|
Edge cutting process
|
Edge treatment
|
Surface finishing
|
n
|
Ksf
|
S-N
curve
|
|
1
|
Rolled or extruded plates and sections
as well as seamless pipes, no surface or rolling defects
|
N/A
|
N/A
|
No surface nor roll defect (1) (2)
|
0
|
0.94
|
B
|
|
2
|
Cut edges
|
Machine- cutting e.g. by a thermal process or sheared
edge cutting
|
Cutting edges chamfered or rounded by means of smooth
grinding, groove direction parallel to the loading direction
|
Smooth surface free of cracks and notches (1)
(2)
|
0.1
|
1.00
|
B
|
|
Cutting edges broken or rounded
|
Smooth surface free of cracks and notches (1)
(2)
|
0.1
|
1.07
|
B
|
|
No edge treatment
|
Surface free of cracks and severe notches (inspection
procedure) (1) (2)
|
0.1
|
1.0
|
C
|
| Manually thermally cut e.g. by flame cutting
|
No edge treatment
|
Surface free of cracks and severe notches (inspection
procedure) (1) (2)
|
0.1
|
1.24
|
C
|
|
(1) Stress increase due to geometry of cut-outs to
be considered.
(2) Fine mesh FE analysis according to Ch 9, Sec
5, [2].
|