11 Fatigue Analysis of Dissimilar Metal Welded …
379
enough safety margin for batch of specimens compared to 95% survival probability
[10]. Hence, it is better to choose 95% survival probability fatigue design curve with
endurance strength of 78.523 MPa which is comparatively conservative to assess
the fatigue performance of TIG welded SS316 L/Monel 400 specimens. Endurance
strength (78.523 MPa) to ultimate tensile strength (548.06 MPa) ratio is found to
be 0.143. It shows that ultimate tensile strength is 6.98 times higher than endurance
strength of the weld material.
11.6.3 Fatigue Damage
Based on the service conditions, the service life can be determined using fatigue
damage theory. It calculates the remaining service life of a structure if the fatigue
life (N f ) is known at a particular stress level. The prediction model for damage
accumulation (D) is obtained by [16]
N f =
(β + 1)
σ
β+1
max − σ
β+1
min
−1
2B(β + 2)
(11.6)
D = 1 −
1 −
n
N f
1 / (β+2)
(11.7)
where σ max and σ min = maximum and minimum stresses and n = number of fatigue
cycles.
Using regression analysis and experimental data, the material constants B and
β can be determined. The fatigue damage curve with β = 4.014 could be drawn
according to Eq. (11.7) as shown in Fig. 11.8. Initially, slope of the curve gradually
increased due to increasing of fatigue cycles. So, damage curve increased linearly
when cyclic ratio ranges between 0 and 0.5. It shows crack initiation in the specimen
with maximum damage value of 0.12. After cyclic ratio 0.5, damage curve becomes
nonlinear up to cyclic ratio 0.9. It shows crack propagation in the specimen with
maximum damage value of 0.33. From a cyclic ratio of 0.9, the cumulative damage
drastically increases; hence, the effective bearing area of specimen decreases which
results in transient fracture. Crack propagation stage is not fully developed which
accounts for the reduced portion of the fatigue life during the cyclic ratio from 0.9
to 1.0. [16] (Fig. 11.9).
11.7 SEM for Fractures Surfaces
The fracture surface look normally undistinguished to visual examination, whereas
it is also challenging to categorize the mode of failure from macroscopic features.
379
enough safety margin for batch of specimens compared to 95% survival probability
[10]. Hence, it is better to choose 95% survival probability fatigue design curve with
endurance strength of 78.523 MPa which is comparatively conservative to assess
the fatigue performance of TIG welded SS316 L/Monel 400 specimens. Endurance
strength (78.523 MPa) to ultimate tensile strength (548.06 MPa) ratio is found to
be 0.143. It shows that ultimate tensile strength is 6.98 times higher than endurance
strength of the weld material.
11.6.3 Fatigue Damage
Based on the service conditions, the service life can be determined using fatigue
damage theory. It calculates the remaining service life of a structure if the fatigue
life (N f ) is known at a particular stress level. The prediction model for damage
accumulation (D) is obtained by [16]
N f =
(β + 1)
σ
β+1
max − σ
β+1
min
−1
2B(β + 2)
(11.6)
D = 1 −
1 −
n
N f
1 / (β+2)
(11.7)
where σ max and σ min = maximum and minimum stresses and n = number of fatigue
cycles.
Using regression analysis and experimental data, the material constants B and
β can be determined. The fatigue damage curve with β = 4.014 could be drawn
according to Eq. (11.7) as shown in Fig. 11.8. Initially, slope of the curve gradually
increased due to increasing of fatigue cycles. So, damage curve increased linearly
when cyclic ratio ranges between 0 and 0.5. It shows crack initiation in the specimen
with maximum damage value of 0.12. After cyclic ratio 0.5, damage curve becomes
nonlinear up to cyclic ratio 0.9. It shows crack propagation in the specimen with
maximum damage value of 0.33. From a cyclic ratio of 0.9, the cumulative damage
drastically increases; hence, the effective bearing area of specimen decreases which
results in transient fracture. Crack propagation stage is not fully developed which
accounts for the reduced portion of the fatigue life during the cyclic ratio from 0.9
to 1.0. [16] (Fig. 11.9).
11.7 SEM for Fractures Surfaces
The fracture surface look normally undistinguished to visual examination, whereas
it is also challenging to categorize the mode of failure from macroscopic features.