Crack Paths 2006
-2
-2
10
10
λ=∞
λ=3
(a)
Periodic Overload
101
102
103
T
104
105
106
107
108
e
Torsion
(b)
.
n
s
S
T
ile
S
h
e
e
h
n
a
s
e
ε
il
r /
.
a
xy
e
r /
A
A
r e
r e
a
/ E
a
/ E
.
n
e
n
rg
e
r g
Energy
y
0.00166
y
Discontinuity
Energy Disco tinuity
0.00109
-3
10
-3
Periodic Overload
.
Predictions ol=overload Shear/Calibration TAreeaEnnesrgiyle Equ valent Cycles to Failure (Nf) γolmax) ean= 0) Axial OL, γscmean= 0)
sc=small cycle
Torsional (γscmax=
ε
100-4
.
xx
PredictionsShear
.
.
Tensile
ε
Area
.
xy
Energy
10-4
.
101
102
103
104
105
106
107
108
Equivalent Cycles to Failure (Nf)
102
103
108
10-432 101
104
105
106
107
λ=3/2
Energy Disco tinuity
Predictions Shear
0.003
(c)
Shear
Periodic OverlEoquaidvalent Cycles to Failure (Nf)
Tensile/Area
+Energy
10
(d)
10 -4
3 0-2 101
102
103
104
105
106
. . . . . . 107
108
Δεxxy/22
Tensile
Area
Energy
Periodic OverlEoqauidvalent Cycles to Failure (Nf) λ=3/4 .
λ=0
Uniaxial
Tensile Cal.
Area/Energy
Shear
Predictions STensilheTenseilear Calibration Ar
10-4 10 -3 2 (e)
Tensile/Area +SEhneargy
Tensile
.
Energy
102
105
107
101
103
104
106
108
109
ε xx xy
.
Predictions STAEehrnnesarlrgey
Periodic Overload
.
εxx
.
.
Equivalent Cycles to Failure (Nf)
Figure 5. Life predictions madewith the Area, Energy, Shear and Tensile models.
rates [12]. Because the modeI and II crack growth rates are almost equal, the differences
in fatigue lives predicted by the models using the area and energy fatigue criteria is small.
The experiments also support a hypothesis that the driving forces for shear and tensile
crack growth are almost the same since the cracks in a few of the specimens tested at each
strain ratio repeatedly alternated between growth on shear planes and growth on tensile
planes.
Both the area and energy life gave predictions close to the observed fatigue lives.
However, the fatigue life predictions become more conservative at short lives (higher
Made with FlippingBook Digital Publishing Software