Accepted Manuscript

Effect of a new specimen size on fatigue crack growth behavior in thick-walled

pressure vessels

Mahmoud Shariati, Ehsan Mohammadi, Reza Masoudi Nejad

PII: S0308-0161(15)30004-1
DOI: 10.1016/j.ijpvp.2016.12.009
Reference: IPVP 3590

To appear in: International Journal of Pressure Vessels and Piping

Received Date: 1 July 2015

Revised Date: 14 December 2016
Accepted Date: 29 December 2016

Please cite this article as: Shariati M, Mohammadi E, Masoudi Nejad R, Effect of a new specimen size
on fatigue crack growth behavior in thick-walled pressure vessels, International Journal of Pressure
Vessels and Piping (2017), doi: 10.1016/j.ijpvp.2016.12.009.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
Effect of a new specimen size on fatigue crack growth behavior in

thick-walled pressure vessels

PT
Mahmoud Shariati1,a Ehsan Mohammadib Reza Masoudi Nejada

RI
a
Faculty of Engineering, Department of Mechanical Engineering, Ferdowsi University of Mashhad, Mashhad, Iran
b
Department of Mechanical Engineering, Shahrood University of Technology, Shahrood, Iran

SC
Abstract

U
Fatigue crack growth in thick-walled pressure vessels is an important factor affecting their
AN
fracture. Predicting the path of fatigue crack growth in a pressure vessel is the main issue
discussed in fracture mechanics. The objective of this paper is to design a new geometrical
specimen in fatigue to define the behavior of semi-elliptical crack growth in thick-walled
M

pressure vessels. In the present work, the importance of the behavior of fatigue crack in test
specimen and real conditions in thick-walled pressure vessels is investigated. The results of
D

fatigue loading on the new specimen are compared with the results of fatigue loading in a
TE

cylindrical pressure vessel and a standard specimen. Numerical and experimental methods are
used to investigate the behavior of fatigue crack growth in the new specimen. For this
EP

purpose, a three-dimensional boundary element method is used for fatigue crack growth
under stress field. The modified Paris model is used to estimate fatigue crack growth rates. In
order to verify the numerical results, fatigue test is carried out on a couple of specimens with
C

a new geometry made of ck45. A comparison between experimental and numerical results
AC

has shown good agreement.

Keywords: Crack growth, Pressure vessel, Stress intensity factor, Standard specimen,
Fatigue.

1
Corresponding author. Tel.: +98 912 173 3750.
E-mail addresses: mshariati44@um.ac.ir (M. Shariati),
e.mohammadi@gmail.com (E. Mohammadi),
masoudinejad@stu.um.ac.ir (R. Masoudi Nejad).

1
 ACCEPTED MANUSCRIPT

Content
Nomenclature .................................................................................................................................. 3

1- Introduction ................................................................................................................................ 4

PT
2- Material and specimens .............................................................................................................. 5

3- Evaluation of fatigue crack growth ............................................................................................ 7

RI
3-1- Numerical Method ................................................................................................................ 7

3-2- Experimental results.............................................................................................................. 8

SC
3-3- Tension test of new specimen ............................................................................................... 9

U
4- Discussion ................................................................................................................................. 10
AN
5- Conclusions .............................................................................................................................. 12

References ..................................................................................................................................... 13
M
D
TE
C EP
AC

2
 ACCEPTED MANUSCRIPT

Nomenclature

da crack growth rate

dN
da crack extension length
dN cycle jump

PT
KΙ mode I stress intensity factor at the crack tip

C material constant

RI
n material constant
K max maximum stress intensity factor

SC
K op the crack opens on the loading portion of the cycle

∆K eff effective mixed mode stress intensity factor

Re external radius of vessel

U
AN
Ri internal radius of vessel

a initial crack length

M

c initial crack half-width

b specimen width
D

W specimen thickness
TE

T applied force

f (x ) a function of the crack length and specimen width

EP

x The ratio of the crack length to specimen width

R Load ratio
C

k the radial ratio

E Young’s modulus
AC

σ ys yield stress

σult ultimate stress

∆K th threshold stress intensity factor range

Kc the fracture toughness

3
 ACCEPTED MANUSCRIPT

1. Introduction

The analysis of a cylindrical vessel is based on constant tensile stress along the thickness

of the cylindrical wall and the real stress distribution is considered for the analysis of thick

wall vessels. A cylindrical pressure vessel is manufactured by bending smooth plates and

PT
welding them to each other. In many cases of surface defects in cylindrical vessels, the main

RI
cause may be the corrosion on the inner surface. These defects occur because of unfavorable

acid conditions or oxidation inside the vessel. This corrosion can also create a porous area in

SC
the surface defect zone and a kind of surface stress concentration may happen. This surface

failure may be the source of some tiny semi-elliptical crack growth. These cracks propagate

U
and then link to each other. These tiny cracks accrete due to cyclic pressure loading and
AN
continue to grow and create a greater semi-elliptical crack (Figure 1).
M

Placement of Figure 1
D

Nowadays, pressure vessels are used in a wide range of sizes according to the geometry of
TE

the vessels and the type of fluid that is under pressure in these vessels [1]. Some efforts have

been made in the past in order to describe crack behavior so that we may estimate the vessel’s
EP

lifetime in order to prevent the vessels from exploding. Nageswara and Acharia [2] presented

an empirical relationship between the failure stress and the elastic stress intensity factor upon
C

failure. This relationship was established by means of three fracture toughness parameters of
AC

34Cr Mo4 grade steel material. Also, Nageswara and Acharia [3] proposed a three parameter

fracture criterion which relates the stress and the elastic stress intensity factor at failure to

each other. These three parameters are determined by using the fracture data for surface crack

tension specimens made of M300 grade maraging steel. Amaro et al. [4] presented a fatigue

crack propagation model for pipeline steel exposed to high-pressure hydrogen accounting for

4
 ACCEPTED MANUSCRIPT
the acceleration at the fatigue crack growth rate due to hydrogen. The fatigue life of

AISI4340 and 4137 type I pressure vessels containing high-pressure hydrogen [5,6] and that

of 4130X type I pressure vessels containing high-pressure hydrogen [7,8] were examined.

Nowadays, with developing the analytical and numerical methods in solid mechanics, it is

intended to study the different parameters both quantitatively and more accurately instead of

PT
qualitative study of their effects [9-13]. The numerous studies have been accomplished to

RI
estimate the residual stresses [14-16]. Up to now, researchers working on failures in vessels

with longitudinal cracks have found new methods to anticipate stress intensity factor in

SC
different points of a semi-elliptical crack under various types of complex loads. An example

of these methods is the weight function method in case of semi-elliptical cracks [17,18].

U
Also, many research works have been performed on failure of vessels containing cracks with
AN
critical pressure and the net section stress computation [19]. More detailed investigations
M

have been carried out about the quality of crack growth and the geometric characteristics of a

semi-elliptical crack under fatigue loading [20, 21]. The study of weight functions and the
D

evaluation of the energy balance formula of Rice [22] allowed the extension of the use of
TE

weight functions by several authors such as Bortmann and Banks-Sills [23] and Oore and

Burns [24]. Gao and Rice [25] presented a study of a semi-infinite crack front during its
EP

coplanar propagation which can be used to obtain the values of stress intensity factor (SIF)
C

along the crack front. Sun and Wang [26] gave in depth interpretations of the energy release
AC

rate of the crack front.

All of these publications indicate the importance of crack growth in pressure vessels. In

spite of many research studies done about semi-elliptical cracks in cylindrical vessels,

standard specimens are usually used in experimental tests. To the best of our knowledge,

there have been no reported works in which a new specimen with more veritable behavior to

crack growth in real cylindrical vessels is presented.

5
 ACCEPTED MANUSCRIPT
2. Material and specimens

We introduce a specimen for analyzing the behavior of the fatigue crack in cylindrical

vessels as arc-shaped specimen with geometric properties shown in Figure 2 using standard

ASTM-E647 [27]. These specimens are used in experimental tests as standard specimens to

approximate the behavior of semi-elliptical cracks in cylindrical pressure vessels.

PT
Nevertheless, in this case, the purpose is to reduce the difference between lifetime of pressure

RI
vessel containing semi-elliptical cracks and standard specimen under two loading directions,

if possible. The initial depth in the semi-elliptical crack in the vessel’s wall will equal to the

SC
initial crack length in the specimen. The aim of this analysis is to find a new geometrical

shape of specimen to estimate the stress intensity factor at the locus of crack tip at the depth

U
of a semi-elliptical crack (Point A in Figure 1) with a determined aspect ratio of a/c, with the
AN
most possible accuracy. It should be mentioned that in this case, the critical stress intensity
M

factor may occur at the depth of the crack’s tip, not at the inner surface located inside the

vessel. In this study, the new specimen is characterized by k=Re / Ri =1.2 and the aspect ratio
D

of a/c=0.1.
TE

Placement of Figure 2
EP

Distribution of stress in a real cylindrical wall along the thickness is completely tensile,
C

while it is not so in the present standard specimens and the distribution of stress is tensile
AC

only near the internal wall (Figure. 4). In addition to standard specimens, the stress intensity

variance is a lot near the internal surface and it causes difficulties in forecasting the behavior

of the crack for these specimens.

In fact, it would be highly desirable to design a new specimen in which the distribution of

stress is such that the far field stress is always tensile throughout the whole thickness and has

a similar distribution like that of thick-walled pressure vessel. In addition, it would be nice if
6
 ACCEPTED MANUSCRIPT
the curve of the stress intensity factor during the growth of the crack along the specimen’s

wall has the most conformity with the curve of stress intensity factor at the tip of the crack at

the tip of the crack deep inside the cylinder’s wall. Such a specimen would be an ideal one to

be used in the study of crack growth in pressure cylinders. With the procedure that the

conformity of the curve of stress intensity factor sets as scale instead of the stress curve, after

PT
analyzing different geometric shapes, a specimen with an especial geometry as shown in Fig.

RI
3 was chosen such that it had the most conformity not only from the point of view of the

curve of stress distribution but also from the curve of the index of stress intensity factor along

SC
the wall of cylindrical pressure vessel when compared with existing standard specimens.

Placement of Figure 3

U
AN
The curves of the distribution of stress in the wall of standard specimens, the new

specimen and the cylinder are compared in Figure 4 under an equal stress in the internal
M

surface for a special substance (steel ck45) with the characteristics according to Table 1.
D

Notice that in the new specimen, in addition to the stress components which are tensile all
TE

along the wall, the intensity of stress variances has been reduced near the internal surface. In

comparison with the new specimen, no considerable differences between the two standard
EP

specimens were seen.

In the new specimen proposed in this study, the existence of a bow that is observed in area
C

A, results in the formation of gentle pressure loads in the area of the internal surface all along
AC

the thickness of the specimen and it results in a reduction of fast changes of stress like what

happens in present standard specimens in the movement toward the internal surface. On the

other hand, area B sustains the bend loads and causes the reduction of the pressure loads in

the area of the external surface.

Placement of Figure 4

7
 ACCEPTED MANUSCRIPT
3. Evaluation of fatigue crack growth

3.1. Numerical Method

The results of numerical analysis for the new specimen have been obtained by using a

software package called Franc 3D [28-31] which is a software for boundary element analysis

and it is especially developed for studies on crack growth in two-dimensional geometries.

PT
This software is supplied by the researchers at Cornell University and is distributed for public

RI
use. With this software the user can opt to use different equations of fatigue crack growth in

order to forecast the lifetime of the specimen under fatigue loads. Also, in this software, the

SC
issues that are related to destruction as a result of the net-section stress of profile in the

U
presence of the crack for the lifetime prediction of the section have been considered. The
AN
calculated crack growth path is found using the Maximum Tangential Stress criterion. The

crack tip curve can be fit after determining the direction of crack growth. The Modified Paris
M

model is used to calculate crack growth rate and fatigue life. The Modified Paris model

considers the effects of closure of the crack. The growth rate is defined by equation (1) as
D

follows [32]:
TE

da
= C ( ∆K eff ) = C (K max − K op )n
n
(1)
dN
EP

where the effective stress intensity factor ∆K eff is defined as the difference between the
C

maximum mode I stress intensity factor ( K max ) and stress level where the crack first opens.
AC

Also note that K op , C and n are material constants.

8
 ACCEPTED MANUSCRIPT
3.2. Experimental results

The experimental results of the CT specimen fatigue test for the tested material (a kind of

steel with the trade name of ck45) can specify its mechanical characteristics as shown in

Table 1.

Placement of Table 1

PT
The test equipment is Servo hydraulics INSTRON 8802 [33] that is controlled with a

RI
computer and is able to make any fatigue test up to a frequency of 20 Hz for finding out the

SC
crack growth according to ASTM E647 standard. The dimensions of a CT specimen have

been shown in Figure 5 for which the stress intensity factor can be found by equation 2 based

on the ASTM E24 standard [34].

U
AN
Ta
KI = f  (2)
t W W 
M

f ( x) = (2 + x) (1 − x) −3 / 2 × (0.886 + 4.64 x − 13.32 x 2 + 14.72 x 3 − 5.64 x 4 )

D

After drawing the graph of da/dN in terms of ∆K a on a logarithmic scale, the gradient of
TE

the linear zone is found to be m and its distance from the origin is found to be Log(C).

Figure 6 shows the fatigue lifetime of the CT specimen versus crack length that is compared
EP

with the graph made by Franc 3D software and the specimen’s lifetime according to the
C

material specifications in Table 1. In this Figure, there is a good conformity between the
AC

results of testing and this shows the accuracy of finding the mechanical characteristics of the

tested material as input for Franc 3D software. In addition, the da/dN versus ∆K is

compared for the CT specimen (experimental) and the new specimen (FEM) on a logarithmic

scale as shown in Figure 7. This Figure shows that there is a good conformity between the

results.

9
 ACCEPTED MANUSCRIPT
Placement of Figure 5

Placement of Figure 6

Placement of Figure 7

PT
3.3. Tension test of the new specimen

Three parts were produced according to the new specimen dimensions with CNC wire cut

RI
device with 0.01 mm accuracy. They were used for fatigue lifetime tests as it can be seen in

SC
Figure 8. The dimensions of the specimen and the geometry of the initial crack are shown in

Table 2. Moreover, Figure 9 shows the image of the CT and the new specimens after

U
finishing the tests.
AN
Placement of Table 2

Placement of Figure 8
M

Placement of Figure 9
D

The specimens are alternatively loaded in a sinusoidal form with the amplitude of 20 kN
TE

and the load ratio of R ≅ 0 and finally the experimental results for all three specimens are
EP

compared with the results of boundary element analysis as shown in Figure 10. There is good

agreement between the experimental results and the results obtained from boundary element
C

analysis using the Franc 3D software. According to these results obtained from the use of the
AC

proposed new specimen, numerical analysis is valid for predicting the behavior of crack

growth. Figure 11 shows the form of meshing used for boundary element analysis of the new

specimen.

Placement of Figure 10

Placement of Figure 11

10
 ACCEPTED MANUSCRIPT
4. Discussion

In the discussion of design related to growth of the fatigue crack in high pressure thick-

wall vessels, it should be considered that it is not simple to find correct information about the

geometry of the resulting crack. Thus, critical conditions should be studied to design thick-

wall pressure vessels with a suitable safety factor and the growth of the crack in this

PT
condition should be considered. According to Table 3, upon increasing a/c from 0.1 to 0.2 a

RI
collapse happens inside the vessel at point B. Also, the lifetime of fatigue crack increases by

1.5 times. Thus, a/c=0.1 is a good ratio for a safe design and it is suitable for design

SC
calculation. With this value of aspect ratio, the collapse will occur at point A and the number

of fatigue cycles will be minimum.

U
Figure 12 shows the comparison of fatigue lifetime in terms of crack length for a standard
AN
specimen, the new specimen and cylindrical vessels. According to this Figure, new
M

specimens show good adaptation with cylindrical vessels regarding the length of critical

crack and the lifetime of fatigue crack. Figure 13 shows a comparison of stress intensity
D

factors in terms of crack length for the standard specimen, the new specimen and cylindrical
TE

vessels that have been obtained with the numerical method. In this condition, there is well

adaptation between new specimens and cylindrical vessels. This point is valid if the length of
EP

crack is under %65 of wall thickness since there is no notable difference between stress
C

intensity factors until this situation occurs.

AC

Placement of Table 3

Placement of Figure 12

Placement of Figure 13

The maximum lifetime of the vessel, the new specimen and the standard specimen after

reaching the threshold of the stress intensity factor are tabulated in Table 4 for various values
11
 ACCEPTED MANUSCRIPT
of the initial crack. This means that by increasing the length of the initial crack, the volume of

the load will decrease to some extent such that the stress intensity factor at the top of the

crack will be equal to the threshold value. These changes are compared in the diagram of

Figure 14. As it is indicated in this Figure, there is a very good agreement between the results

for the new specimen and the cylindrical vessel.

PT
Placement of Table 4

RI
Placement of Figure 14

SC
In this study we also tried to investigate the effect of the type of material. Thus, according

to Table 5, two materials with different mechanical specifications were considered. The

U
fatigue lifetime of crack for a standard specimen, the new specimen and cylindrical vessel
AN
were compared in Table 6 by applying the data presented in Table 5. This comparison was

done by changing the type of specimen and the way of loading for two kinds of steel. Thus
M

the superiority of the new specimen in the estimation of fatigue lifetime of the crack was
D

observed by comparison of percentage errors.

TE

Placement of Table 5

Placement of Table 6
EP

Another point which is considered is the radial ratio of the vessels for which we can use
C

the ratio k= Re/Ri of the new specimen. In Figure 15, the curve of fatigue lifetime of one
AC

vessel with steel of type A with specifications shown in Tables 5 and 6 with different radial

ratios in the scope of 1.1≤ k ≤2 are reported. The investigation of this curve shows that the

maximum value of fatigue lifetime of the vessel is about 58000 cycles at the radial ratio of

k=1.4 and the minimum value of fatigue lifetime of the vessel is about 53000 cycles in the

considered scope at the radial ratio of k=2. Comparing with the new specimen whose lifetime

12
 ACCEPTED MANUSCRIPT
is 56100 cycles, the estimated error of fatigue lifetime for the new specimen is in the range of

-6% < error < +2% for each radial ratio of k. Compared with the standard specimen in which

the estimated error is +37% < error < +39%, this is very precise. Therefore, we may conclude

that the new specimen is the proper one for investigation of growth of cracks in thick-walled

pressure vessels.

PT
Placement of Figure 15

RI
5. Conclusions

SC
Considering the results of this study, a new specimen was introduced for fatigue test in

U
order to investigate crack propagation in thick-wall pressure cylindrical vessels in this
AN
research study. The results indicate closer agreement regarding fatigue crack growth behavior

and distribution of stress as well as fatigue life with real thick wall pressure vessels than
M

existing standard specimens. In this research, a new specimen with radial ratio of k=1.2 and

initial crack length of a=6mm in a semi-elliptical crack (a/c=0.1) was investigated. Moreover,
D

the effect of k (radial ratio) on the lifetime of crack was investigated. It was noted that this
TE

effect is negligible. So, it is suggested to use this dimensionally fixed new specimen for

future research works in future investigations on the growth of cracks in thick-wall pressure
EP

vessels.
C
AC

References

[1] Shigley JE, Mischke CR, Brown TH , “Standard handbook of machine design”,

McGRAW-HILL publication , 2004; 39.2.

13
 ACCEPTED MANUSCRIPT
[2] Nageswara B, Acharia A.R, “Failure assessment on 34Cr Mo4 grade steel cylindrical

pressure vessels with an axial surface crack”, International Journal of Pressure Vessels and

Piping, 1997; 72:157–163.

[3] Nageswara B, Acharia A.R, “Failure assessment on M300 grade maraging steel

cylindrical pressure vessels with an internal surface crack”, International Journal of

PT
Pressure Vessels and Piping, 1998; 75:537–543.

RI
[4] Amaro RL, Rustagi N, Findley KO, Drexler ES, Slifka AJ. Modeling the fatigue crack

growth of X100 pipeline steel in gaseous hydrogen. Int J Fatigue 2014;59:262e71.

SC
[5] Takasawa K, Wada Y, Ishigaki R, Tanaka Y, Iwadate T, Ohnishi K. Internal pressure

fatigue test of Cr-Mo steel in 45 MPa hydrogen environment. In: Proc. ASME PVP2007,

San Antonio, TX; 2007. PVP2007e26508.

U
AN
[6] Wada Y. Hydrogen gas embrittlement and safety of stationary hydrogen storage vessel
M

made by low alloy steel for hydrogen station. Petrotech 2009;32:459e64.

[7] Nibur KA, San Marchi C, Somerday BP. Fracture and fatigue tolerant steel pressure
D

vessels for gaseous hydrogen. In: Proc. ASME PVP2010, Bellevue, WA; 2010.
TE

PVP2010e25827.

[8] San Marchi C, Dedrick DE, Van Blarigan P, Somerday BP, Nibur KA. Pressure cycling
EP

of type 1 pressure vessels with gaseous hydrogen. In: Proc. 4th International Conference
C

on Hydrogen Safety (ICHS4), San Francisco, CA; 2011.

AC

[9] Masoudi Nejad R, Salehi SM, and Farrahi GH, Simulation of railroad crack growth life

under the influence of combination mechanical contact and thermal loads, In 3rd

International conference on recent advances in railway engineering, Tehran, Iran, 2013.

[10] Salehi, SM, Farrahi GH, Sohrabpoor S, and Masoudi Nejad R, "Life Estimation in the

Railway Wheels Under the Influence of Residual Stress Field." International Journal of

Railway Research 2014;1(1):53-60.

14
 ACCEPTED MANUSCRIPT
[11] Masoudi Nejad R, Shariati M, and Farhangdoost K, "Effect of wear on rolling contact

fatigue crack growth in rails", Tribology International 2016;94:118-125.

[12] Ghahremani Moghadam D, Farhangdoost K, and Masoudi Nejad R, "Microstructure and

Residual Stress Distributions Under the Influence of Welding Speed in Friction Stir

Welded 2024 Aluminum Alloy", Metallurgical and Materials Transactions B

PT
2016;47(3):2048-2062.

RI
[13] Masoudi Nejad R, Farhangdoost K, and Shariati M, "Numerical study on fatigue crack

growth in railway wheels under the influence of residual stresses", Engineering Failure

SC
Analysis 2015;52:75-89.

[14] Hadipour M, Alambeigi F, Hosseini R, and Masoudinejad R, "A study on the vibrational

U
effects of adding an auxiliary chassis to a 6-ton truck", Journal of American Science
AN
2011;7(6):1219–26.
M

[15] Masoudi Nejad R, "Using three-dimensional finite element analysis for simulation of

residual stresses in railway wheels", Engineering Failure Analysis 2014;45:449-455.

D

[16] Masoudi Nejad R, Shariati M, and Farhangdoost K, "3D finite element simulation of
TE

residual stresses in UIC60 rails during the quenching process", Thermal Science, 2016,

just-accepted.
EP

[17] Mettu SR, Raju IS, Forman RG, “Stress intensity factors for part-through surface cracks
C

in hollow cylinders”, NASA Technical Report No. JSC 25685, LESC 30124, 1992.
AC

[18] Zheng X. J, KICIAK A , Glinka G., “Weight functions and stress intensity factors for

internal surface semi-elliptical crack in thick-walled cylinder”, University of Waterloo,

Engineering Fracture Mechanics, 1997; 58:207-221.

[19] Perl M, Levinsohn JA, “Measurement of the actual fracture toughness of a maraging-300

pressurized cylinder using the vessel as the test specimen”, Engineering Fracture

Mechanics, 1989; 34:525–530.

15
 ACCEPTED MANUSCRIPT
[20] Shlyannikov V.N, “Fatigue shape analysis for internal surface flaw in a pressurized

hollow cylinder”, International Journal of Pressure Vessels and Piping, 2000; 77:227–234.

[21] El Khalil Hashi B, Rechak S, Belkacemi Y, Maurice G “Modeling of elliptical cracks in

an infinite body and in a pressurized cylinder by a hybrid weight function approach”,

International Journal of Pressure Vessels and Piping, 2005; 82:917–924.

PT
[22] Rice JR. Some remarks on elastic crack-tip stress fields. Int J Solids Struct 1972;8:751–

RI
8.

[23] Bortmann Y, Banks-Sills L. An extended weight function method for 2D and 3D-

SC
problems. J Appl Mech 1983;50:907–9.

[24] Oore M, Burns DJ. Estimation of stress intensity factors for embedded irregular cracks

U
subjected to arbitrary normal stress fields. Trans ASME, J Press Vessel Technol
AN
1980;102(2):202–11.
M

[25] Gao H, Rice JR. Shear stress intensity factors for planar crack with slightly curved front.

J Appl Mech 1986;53:774–8.

D

[26] Sun CT, Wang CY. A new look at energy release rate in fracture mechanics. Int J Fract
TE

2002;113:295–307.

[27] ASTM E647-00. "Standard Test Method for Measurement of Fatigue Crack Growth
EP

Rates." ASTM annual standards, 2011; 03.01.

C

[28] Masoudi Nejad R, Farhangdoost K, Shariati M, and Moavenian M. “Stress intensity

AC

factors evaluation for rolling contact fatigue cracks in rails”, Tribology Transactions,

2016, just-accepted.

[29] Masoudi Nejad R, Farhangdoost K, and Shariati M. “Three-dimensional simulation of

rolling contact fatigue crack growth in UIC60 rails”, Tribology Transactions, 2016;

59(6):1059-1069

16
 ACCEPTED MANUSCRIPT
[30] Masoudi Nejad R. “Rolling contact fatigue analysis under influence of residual stresses",

MS Thesis, School of Mechanical Engineering, Sharif University of Technology, Tehran,

Iran, 2013.

[31] Masoudi Nejad R., Salehi SM, Farrahi GH, and Chamani M. “Simulation of crack

propagation of fatigue in Iran rail road wheels and Effect of residual stresses”, In

PT
Proceedings of the 21 st International Conference on Mechanical Engineering, Tehran,

RI
Iran, 2013.

[32] Shariati M, and Masoudi Nejad R. “Fatigue strength and fatigue fracture mechanism for

SC
spot welds in U-shape specimens”, Latin American Journal of Solids and Structures, 2016;

13:2487-2501.

U
[33] Instron Model 8802,8803,8804,8805 and 8806, Reference Manual-Pre-Installation.
AN
[34] ASTM E24. "Draft ASTM Test Standard for Fracture Toughness Testing of Weldment."
M

ASTM annual standards, 1991; 06.05.

D
TE
C EP
AC

17
 ACCEPTED MANUSCRIPT

Figure captions

Fig. 1. Schematic of a semi-elliptical a) in a wall section of a cylindrical vessel b) in Franc

3D software

Fig. 2. Standard specimens for analyzing the crack growth in vessels and pipes: (a) First

PT
specimen (b) Second specimen

RI
Fig. 3. Geometry of new specimen

Fig. 4. Stress distribution along the wall for different specimens

SC
Fig. 5. The geometry of CT specimen

Fig. 6. The crack lifetime of CT specimen versus crack length

U
Fig. 7. The da / dN versus ∆K for different specimens
AN
Fig. 8. Experimental test setup
M

Fig. 9. Experimental specimens after tension and fatigue tests: (a) CT specimen (b) New

specimen
D

Fig. 10. Fatigue lifetime due to fatigue test of new specimen and results of boundary element
TE

analysis (Franc 3D) versus crack length

Fig. 11. A new specimen importing in Franc 3D software: (a) specimen without any crack (b)
EP

specimen with crack

C

Fig. 12. Fatigue lifetime in terms of length of crack in a standard specimen, new specimen
AC

and cylindrical vessel.

Fig. 13. Stress intensity factor in terms of crack length for standard specimen, new specimen

and cylindrical vessel that have been obtained with boundary element method

Fig. 14. Fatigue lifetime of the crack in terms of the crack length for standard specimen, new

specimen and cylindrical vessel

18
 ACCEPTED MANUSCRIPT
Fig. 15. Indication of the fatigue lifetime changes of the semi-elliptical crack by the changes

of radial ratio of k

PT
RI
U SC
AN
M
D
TE
C EP
AC

19
 ACCEPTED MANUSCRIPT

Table captions

Table 1. The features of the tested material

Table 2. Characteristics of crack geometry in new specimen

Table 3. Effect of a/c on maximum possible lifetime of semi elliptic crack in a cylindrical

PT
vessel

RI
Table 4. The maximum fatigue lifetime for vessel and new specimen after the stress intensity

factor reaches to the threshold

SC
Table 5. Mechanical specifications of the tested materials

Table 6. Fatigue lifetime comparison of two different materials

U
AN
M
D
TE
C EP
AC

20
 ACCEPTED MANUSCRIPT
Table 1.

E  ys ult C n K th R 0
Kc t 10 mm
(MPa) (MPa)
(MPa) ( Mpa mm ) ( Mpa mm )

      

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 2.
Ri W b a
(mm) (mm) (mm) (mm)
125 25 10 6

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
Table 3.

Collapse a/c Fatigue Internal

Point Lifetime Pressure
(Figure 1) (Cycle) (Mpa)
   
   

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Table 4.
a Cylinder New Standard New Standard
(mm) Fatigue Specimen Specimen #1 Specimen Specimen #1
Lifetime Fatigue Fatigue Fatigue Fatigue
(Cycle) Lifetime Lifetime Lifetime Lifetime
(Cycle) (Cycle) Error % Error %
6 55300 56100 35000 1.43 -36.7
8 57800 58900 33900 1.91 -41.35

PT
10 58700 58300 30800 -0.69 -47.53
12 58200 57600 26850 -1.03 -53.87

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 5.

Material C n K th R 0
K max a 6 mm
Kc t 10 mm

( Mpa mm ) ( Mpa mm )
A 2.16E-15 3.69 694 695 1597

B 2.73E-10 2 118 1110 2731

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 6.

Material Cylinder New Standard New Standard

(Cycle) Specimen Specimen #1 Specimen Specimen #1
Fatigue Fatigue Fatigue Fatigue
Lifetime Lifetime Lifetime Lifetime
(Cycle) (Cycle) Error % Error %
A 55300 56100 35000 1.43 -36.7

PT
B 20500 18900 11200 -7.81 -45.4

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
a)
M
D
TE
EP

b)

Figure 1.
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
a)
M
D
TE
EP
C
AC

b)

Figure 2.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D

Figure 3.
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 4.
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 5.
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 6.
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D

Figure 7.
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE

Figure 8.
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
a)

U
AN
M
D
TE
EP
C
AC

b)

Figure 9.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 10.
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN (a)
M
D
TE
EP
C
AC

(b)
Figure 11.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
Figure 12.
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
Figure 13.
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
Figure 14.
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
SC
Figure 15.

U
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

Highlights

• This paper provides a new specimen to define the behavior of fatigue crack growth

PT
• We estimate the behavior of fatigue crack growth in specimen and pressure vessel

• A 3D finite element model has been applied to estimate the fatigue life

RI
• We compare the results of fatigue loading for cylindrical vessel and specimens

• Comparison between experimental and numerical results has shown a good agreement

U SC
AN
M
D
TE
C EP
AC