Professional Documents
Culture Documents
Determination of Stress Intensity Factor For A Crack Emanating From A Hole In-Libre
Determination of Stress Intensity Factor For A Crack Emanating From A Hole In-Libre
Engineering
and Technology (IJMET),
ISSN 0976
INTERNATIONAL
JOURNAL
OF MECHANICAL
ENGINEERING
6340(Print), ISSN 0976 6359(Online) Volume 4, Issue 2, March - April (2013) IAEME
AND TECHNOLOGY (IJMET)
IJMET
IAEME
(2)
ABSTRACT
The machine elements with cylindrical profile such as pressure vessel, cylindrical
shells, which have been used extensively as the structural configuration in aerospace and
shipping industries needs to be leak proof. But however its not possible to fabricate 100%
leak proof pressure vessel / cylindrical shell as the industrial materials do not have uniform
composition. Thus defects or cracks are inevitable in their substructure, also during their
service life a crack may initiate on an internal/external boundary of circular cylinder which
has important influence on stress distribution in the structure. Hence the structural
assessments of hallow cylinders ranging from thick walled pressure vessel to thin walled
pipes has to be carried out, that in-turn relays on availability of Stress Intensity Factor (
S.I.F) for fracture analysis. The magnitude of the S.I.F determines the propagation of crack.
In this paper, Considerable effort has been devoted for computation of the S.I.F of crack
emanating from a hole in pressurized cylinder. The objective of this work is to determine SIF
(Plane Strain) for a crack emanating from a hole in a Pressurised cylinder using Finite
Element Method (FEM). From this study it was observed that the value of SIF rises suddenly
when the crack tip is near to the hole and it stabilises as the crack tip move far from the hole.
The SIF values evaluated for different crack length using FEM is normalised with the
analytical values obtained from theoretical equation with respect to (a/D) ratio which
provides important information for subsequent studies such as the crack growth rate
determination and prediction of residual strength with plane strain and plane stress
conditions.
373
INTRODUCTION
Pressure vessel and cylindrical shells have been used extensively as the structural
configuration in aerospace and shipping industries needs to be leak proof. However new
advancement in computers has made Finite Element Analysis (FEA) a practical tool in the
study of pressure vessels [1], especially in determining stresses in local areas such as
penetrations and service holes. The most likely places for crack initiating and development of
cracks are the service holes. Due to the high stress concentration in this area cracks may grow in
time, leading to a loss of strength and the reduction of the lifetime of the product as shown in
Figure1. If the structure is concerned with different loading, the crack behaviour must be assessed
in order to avoid catastrophic failures. For this, the knowledge of the crack size, service stress,
material properties and Stress Intensity Factor (SIF) is required [2]. Hence the structural
assessments of hallow cylinders /shell ranging from thick walled pressure vessel to thin
walled pipes has to be done, that in-turn relays on availability of stress intensity factor for
fracture and fatigue analysis. Thus it has been recognized that the stress intensity factor is an
important parameter to determine the safety of a cracked component, but the basic practical
problem a designer faces, is to make a decision to opt the method for determining stress
intensity factors. It is not easy to strike a balance between the accuracy of the method, time
required to get a solution, and cost. Numerous equations for stress intensity factors are
available in the literature [1 6]. These factors represent various geometries and loading
conditions of fundamental importance in the prediction of structural failure of cracked
cylindrical bodies. In all there are probably more than 600 formulas for calculating K values
for different crack configurations, body geometries, and loading situations.
Here the objective of the work is to determine S.I.F for a crack emanating from a hole in
a pressurized cylinder.
Fig.1: Larger crack formed by the link-up of fatigue cracks at adjacent rivets.
2. FRACTURE MECHANICS
Fracture mechanics involves a study of the presence of the cracks on overall properties
and behaviour of the engineering component. The process of fracture may be initiated at defect
locations like micro-cracks, voids, and the cavities at the grain boundaries. These defects can
lead to the formation of a crack due to the rupture and disentanglement of molecules, rupture of
atomic bonds or dislocation slip [3].
Cracked body can be subjected to three modes of loads as shown in Figure 2. In some
cases, body may experience combination of the three modes:
374
1. Opening mode: The principal load is applied normal to the crack surfaces, which tends
to open the crack. This is also referred as Mode I loading (Figure 2a).
2. In-plane shear mode: This mode corresponds to in-plane shear loading which tends to
slide
One crack surface with respect to the other. This is also referred as Mode II loading
(Figure2b).
3. Out-of-plane shear mode: This is the tearing and anti-plane shear mode where the crack
surfaces move relative to one another and parallel to the leading edge of the crack (Figure 2c).
(a)
(b)
(c)
v=+
K1
2G
w =+
r
(1 + k )
2
2 K 111
G
(2)
r
2
(3)
375
Where:
u, v, w = displacements in a local Cartesian coordinate system as shown in figure 3
r, = coordinates in a local cylindrical coordinate system as shown in figure 3.
G = shear modulus
In Plane Stress
(4)
K=
1+ v
K= 3 - 4 In Plane strain..(5)
= Poisson's ratio
For Mode-1, SIF at crack tip is expressed as
G V
(6)
K1 = 2
1+ k r
Where v, are the motions of one crack face with respect to the other.
Then A and B are determined so that
V
= A + Br
(7)
r
At points J and K.
Next, let r approach 0
V
lim r 0
=A
(8)
r
Fig. 3: Nodes Used for the Approximate Crack-Tip Displacements for Full crack Model
Thus, Equation 5 becomes:
2GA N
mm
K 1 = 2
1 + k mm 2
(9)
A shell with a longitudinal crack was meshed using three different mesh densities.
Mainly, the area around the imperfection was modelled with a finer mesh. Further the crack
tip singular elements were created using KSCON command. For this model there are 36
singular elements around the crack tip and the radius of the first row elements is a (Where a =
a/100).The model is then solved (Static Analysis) by subjecting it to an internal pressure of 1MPa
load with appropriate boundary conditions. Then the S.I.F is evaluated in general postprocessor
by using KCALC command.
The geometry of the meshed test model with crack tip singular elements in ANSYS 12 is
as shown in the Figure 5.The material considered is 304 steel (ASME). The material is assumed
to be linear elastic with youngs modulus of 2.5GPa and poisons ratio 0.3
Where,
D= Diameter of the hole (20mm),
a= Half Crack length
=applied hoop stress (Pr/t)
P= Internal pressure 1MPa
t=Thickness of the cylindrical shell
10mm
(a)
(b)
(c)
Fig. 5: (a) Finite Element Model meshed with Boundary Condition (b) Zoomed View of
elements near crack tip (c) Zoomed View of Crack Tip Singular Elements
377
a
( Rt )
The half crack length was varied from 20mm to maximum half crack length of 439.53mm .The
maximum crack length in a given dimensions of cylindrical shell was determined using curvature
parameter [9]
a 4
12(1 2 ) .(11)
Rt
if =8 for longitudinal cracks , thickness of the cylindrical shell is 10mm and radius of the
cylindrical shell is 1000mm then the maximum crack length for given set of cylindrical shell
dimension is 439.53mm.
The values of S.I.F obtained by the theoretical and FEA is given in table 1.
378
Table 1: Mode-1 S.I.F (KI) Using FEA and theoretical KI(Theo) for different Half Crack
lengths (a)
Half Crack
Length (a)
mm
20
40
60
80
100
120
140
180
220
260
300
340
380
400
439.532
Mode -I SIF by
FEA
MPa
mm
Mode -I SIF
by Analytical
MPa
821.49
1282.6
1749.6
2265.4
2831.4
3448.9
4114.8
5570.8
7168.4
8886.5
10711
12634
14608
15685
17798
% error
mm
851.322
1330.62
1828.751
2352.63
2931.63
3555.117
4221.655
5669.381
7247.292
8933.523
10710.546
12561.166
14471.544
15446.58
17398.911
3.50
3.61
4.33
3.71
3.42
2.99
3.42
1.74
1.09
0.53
0.00
-0.58
-0.94
-1.54
-2.29
Fig.7: variation of theoretical and FEA values of S.I.F Vs crack length for longitudinal
crack in Pressurized cylinder
From the Fig 7 it is indicated that the results which were obtained by using the finite
element method are in good agreement with theoretical equation for a longitudinal through crack
emanating in internally pressurized cylindrical shell with an average percentage of error 1.53%
which is negligible. Thus the proposed methodology to determine the Mode-1 S.I.F for
longitudinal cracks in pressurised cylindrical shell is validated against a standard Procedure.
379
N
mm 2
mm (12)
The variation of normalized Stress Intensity Factor (KI/KO) (By Plane Strain Method)
with respect to a/D ratio [actual crack length (a) to the Diameter of the hole (D)] is as shown in
Figure 8. The normalized SIF (KI/KO) is used to obtain the characteristic curve of SIF which
depends only on the geometrical factor and its variation within the given domain (a/D).
Its observed that as the crack is near to the hole the stress concentration around holes has
a strong influence on the SIF value. For a/D ratio 0.5 there is a steep rise in SIF KI, this is due to
crack is small and the crack tip is near to stress concentration at the hole from which crack in
emanating. As the crack grows further (for a/D ranging from 0.1 to 25) the crack tip moves far
from the stressed areas hence the value of SIF increases the system will fail with increase in crack
length.
Table 2: The normalized Stress Intensity Factor (KI/KO) with respect to a/D ratio
Half
Crack
Length (a)
Hole
dia(D)
mm
mm
10
20
40
60
80
140
180
200
220
260
300
320
380
400
20
20
20
20
20
20
20
20
20
20
20
20
20
20
a/D
0.50
1.00
2.00
3.00
4.00
7.00
9.00
10.00
11.00
13.00
15.00
16.00
19.00
20.00
Mode -I SIF by
FEA(KI)
aeff
Mode -I SIF by
Theoritical(Ko)
MPa mm
mm
MPa mm
743.96
853.01
1085.8
1310.4
1540.1
2302.5
2881.6
3191.3
3514.8
4200.3
4932.1
5313.5
6519.8
6938.3
15
20
30
40
50
80
100
110
120
140
160
170
200
210
686.51
792.72
970.88
1121.07
1253.40
1585.43
1772.57
1859.09
1941.75
2097.33
2242.14
2311.15
2506.79
2568.70
380
KI / Ko
1.08368
1.076059
1.118372
1.168882
1.228742
1.452284
1.625663
1.716596
1.810118
2.002688
2.199727
2.299075
2.600855
2.701098
Fig.8: Variation of normalized S.I.F for a crack emanating from circular hole in
pressurized cylinder Vs a/D
The Deformed Geometries for crack length (a) of 20 mm is as shown in Figures 9, The
maximum Von-Misses stress is found to be at crack tip.
Fig 9: (a) VonMises Stress Distribution for Pressurised cylinder containing hole dia
D=20mm and crack length a=20mm
6. CONCLUSION
The problem of determining stress intensity factors for a crack emanating from a hole in a
pressurized cylinder is of prime importance in damage tolerance analysis. In the present study
ANSYS12, unified FEA software is chosen. It has the required pre-processing capabilities
for finite element modeling and analysis of cracked shell structures as demonstrated in this
paper.
381
The variation of normalized S.I.F (Ki/Ko) with respect to a/D ratio is used to obtain the
characteristic curve of SIF which depends only on the geometrical factor and its variation within
the given domain (a/D).
The fracture mechanics analysis on effects of pressure bulging near the crack on the
stress intensity factor is described in this paper that provides an explanation for the lower
strength of cracked cylinders.
The presented stress intensity factors in this paper are essential to predict
(1) Mixed mode fracture under static, dynamic and sustained loads
(2) Residual strength
(3) Crack growth life under cyclic loading conditions.
However there is a clear need to verify the predictions using experimental
investigations, but the method used in this paper can be utilized for calculating the stress
intensity factor for many other loading cases and many values of the crack length. This provides
important information for subsequent studies, especially for fatigue loads, where stress intensity
factor is necessary for the crack growth rate determination.
REFERENCES
[1]
[2]
382