A Finite Element Based Study On Stress Intensification Factors - SIF - For Reinforced Tees

NAFEMS 2011
REPRODUCTION AND DISTRIBUTION PROHIBITED
www.nafems.org
A FINITE ELEMENT BASED STUDY ON STRESS

INTENSIFICATION FACTORS (SIF) FOR REINFORCED
FABRICATED TEES.

FABRICATED TEES.
Anindya Bhattacharya,
Senior EngineerI
THEME
Stress Intensification factors used in ASME Piping codes B31 relate to fatigue
failure. In this paper an attempt has been made to compute these factors for
Reinforced Fabricated Tees using Finite Element Method and Elastic Stress
Categorization route of ASME Boiler and Pressure Vessel code SEC VIII Div
2.
KEY WORDS
Stress Intensification Factor, Peak Stress , Markl, Shell elements.
SUMMARY
Stress Intensification Factors (SIF), as used in American Power (B31.1) and
Process Piping (B31.3) codes, correlate the fatigue strength of piping
components with respect to girth butt welds in straight pipes subjected to
bending moments. These codes provide empirical formulas for this factor.
These formulas are based on experimental findings by A.R.C. Markl and his
team in the 1950s. However, the applicability of such factors is restricted to a
diameter over thickness (D/T) ratio of 100. In this paper an attempt will be
made to compute SIF for Reinforced Fabricated Tee (normal intersection)
using Finite Element Analysis (FEA) for both D/T < 100 and D/T>=100. Shell
based analysis results will be compared to continuum /shell solid submodelling techniques to evaluate the SIFs. The objective of the paper will be
two-fold: to check the FEA computed values with respect to the code specified
ones, as well as to check the applicability of code formulas for D/T>100.
I Senior Engineer (Stress analysis) C.Eng, Eur.Ing; [email protected]
CB&I, London W2 6LG

Phone: +44 (0) 20 7053 5668
Fax: +44 (0) 20 7053 3737
Presented at the NAFEMS World Congress 2011
May 23-26 2011 | Boston, USA
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Three FEA codes, FE-Pipe (Version 5.0), NOZZLEPRO Version 7.5 and
ABAQUS version 6.9 have been used for the above study.
NOMENCLATURE:
D= outside diameter of header pipe.
d= outside diameter of branch pipe.
T= thickness of header pipe.
t= thickness of branch pipe.
R= mean radius of header pipe.
r = mean radius of branch pipe.
1:
Elastic stress classification route
The concept of stress intensification factor, as used in [1] is based on linear

elastic behavior. The American Piping Codes B31 do not explicitly use the
terms primary stress, secondary stress, peak stress, etc. (These are outlined in
the ASME Boiler and Pressure Vessel Codes Section VIII, Division 2 and
Section III [3] [4]) although the concepts are inherent in the specification of
different allowable stresses for load and displacement driven stresses. The
concepts are important to develop the methods to be used in computing such
factors using FEA. To define these terms in a nutshell: primary stresses are
load driven and do not reduce due to redistribution; secondary stresses develop
to maintain displacement compatibility and are self limiting; and peak stresses
are significant only from the fatigue-failure standpoint. [3][4]
The individual stress categories have separate failure modes associated with
them. Primary stresses result in gross plastic deformation type failure. Primary
plus secondary stresses result in ratcheting (progressive plastic deformation or
PD) and peak stresses result in fatigue-failure. Henceforth, in line with ASME
Boiler and Pressure Vessel Code terminology [3][4], local primary membrane
stresses will be termed as PL, primary bending stresses as Pb, Q as secondary
stresses and F as peak stresses.
The stress intensification term as used in [1] is for peak stress only under
flexural loading. ASME Boiler and Pressure Vessel Code Sec III [4] addresses
stress indices (a term not exactly equivalent to stress intensification factor) for
other types of stresses as well. B31.3 factors are applicable for both in- and
out-plane bending moments with the corresponding stress intensification
factors termed as in-plane SIF and out-plane SIF. Flexibility factors can also
have similar terminology, although ASME B31.3 expresses single flexibility
factor for both types of loading.
2
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
2:
The origin of the stress intensification and flexibility factors in the

American piping codes
In the late 1950s, A.R.C. Markl and his team [2] conducted a series of
experiments using displacement controlled fatigue tests to evaluate stress
intensification factors. Details of the experimental set up and method can be
found in [2] [5] [7] [8] [9]. Markls original work, based on which stress
intensification factors were derived, was based on the following equation (in
psi).
i.Sf = 490000N(-0.2)
eq.1 [2]
where i=stress intensification factor, Sf= stress range to failure, N=no. of cycles
to failure
2.1: Markls test for establishing SIF for headers with branch connections is
shown in Fig (1) below. The boundary condition used was fixing the branch
end in all six degrees of freedom and applying displacement input (cyclically
varying) at one end of the header. Fig (1) is for in-plane SIF. Fig (2) shows
boundary conditions (BCs) for header SIF, branch SIF and Markls test set up.
Fig (3) shows the schematic arrangement for branch SIF BC (in plane) and Fig
(4) shows schematic boundary condition for out-of-plane SIF as per Markls
original test set up.
Markl assumed that his set-up would produce the same results as if the loads
were applied to the branch and one end of the header was fixed. This according
to the principles of static mechanics is that the reaction at the base must be the
same as the applied force. However, when the out-plane load is applied to the
end of the header and the branch is fixed, the branch is exposed to torsion plus
bending (instead of pure bending) [6]. It does not make much difference but it
is to be noted. ASME B31.3 also does not specify separate SIFs for header and
branch and Markls boundary condition is essentially for a branch SIF. Using
Markls boundary conditions, as per ASME B31.3, Table 1 shows in-plane and
out-of-plane SIFs for different header-branch combinations. It is seen clearly
that out-of-plane SIF > in-plane SIF, regardless of D/T ratio.
3:
Finite Element models description
References [14][15][16][17] discuss the issues of cylinder-cylinder

intersections.
Finite element analysis (FEA) was done using both continuum and shell
elements (including both linear and quadratic variations) using ABAQUS,
NOZZLEPRO and FE-Pipe codes. Final results shown (shell elements) in
Table 1 are based on 8 noded reduced integration isoparametric shell element (
ABAQUS element S8R) ,which is also a Reissner-Mindlin element [12]. Six
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Degrees of freedom per node (three translations and three rotations) has been
defined for this element. The continuum element shown in the final result
(Table 1) is the 20-node brick element (ABAQUS element C3D20R). This
element has 3 translational degrees of freedom per node. The length from the
centerline of the branch pipe to the header end has been taken as 5D and the
length from the centreline of header pipe to branch end has been taken as 4D to
remove the end effects [14][15]. On the header pipe, element size has been
kept as less than 0.3 RT at and close to intersection with aspect ratio less
than 5. [14][15] On the branch size, element size has been kept less than
0.3 rt at and close to intersection with aspect ratio less than 6.Welds were
included in some models (for sensitivity check) based on [11][13]. Stresses in
welds were not computed using hot spot [3] or linearization methods [3], but
were computed using fatigue strength reduction factor (FSRF) method as
outlined in [3]. Linear variation of S8R, ABAQUS element S4 (4-node full
integration shell element) has also been used for comparison along with full
integration linear continuum element (ABAQUS element C38D). Linear
elements both in shell and continuum version with full integration have shown
lower value of SIF.
Results in Table 1 are maximum values based on the three FE codes used and
for minimum width of pad and for quadratic elements only.
The following models were used:
Header 12 inch NPS (wall thickness 9.52 mm) and Branch 6 inch NPS
(wall thickness 7.5 mm). D/T= 34, d/D=0.5
(wall thickness 7.5mm). D/T= 72, d/D=0.18
(wall thickness 7.5mm). D/T= 96, d/D=0.13
4
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Fig (1) Markls test set up (in-plane SIF) [6]
Fig (2). Header SIF, Branch SIF and Markls

Boundary conditions [6]
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Fig (3) Schematic arrangement for Branch SIF Boundary condition [6]
Fig (4) Schematic arrangement for Markls out-of-Plane BC [6]

4:
A finite element based approach on computation of stress

intensification factors
Stress intensification factor (peak stress) can be expressed in a simplified

manner as the ratio between peak stresses in a component to that of nominal
stress in the same component. Nominal stress in a part can be taken as M/Z for
the applied bending moment. Stress intensification factors can also be
computed for primary and secondary stresses (as required in the ASME Section
III code). For generation of peak-SIF in Reinforced Fabricated Tee connection
using an FE model, the following procedure was used - ASME B31 piping
codes (here, for the purpose of this paper, we refer to ASME B31.3 only) use
SIF based on a ratio of actual stress due to application of bending moment to
that of the nominal stress in a girth (circumferential) butt weld due to the same
bending moment. Hence, B31-SIF = Actual stress in part due to bending
moment, M upon stress in girth butt weld due to M. Girth butt welds have
6
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
stress intensification factors between 1.7 and 2.0 [9][10] and are material
dependent. Thus, conservatively, the true peak stress in a girth butt weld due to
a moment, M can be expressed as:
Peak stress in a girth butt weld (due to M) = 2(M/Z).
M is the moment in the pipe with the butt weld, and Z is the section modulus of
the pipe with the butt weld. Therefore in terms of the nominal stress in a
straight pipe without a girth butt weld, B31-SIF can be expressed as [9],[10]
Actual (Peak Stress) due to moment, M
Stress in Girth Butt Weld due to moment, M
B31.3 SIF =
or,
B31.3 SIF =
Actual (Peak Stress) due to moment, M

2 X (Moment, M)/ (Section Modulus, Z)
In terms of ASME Section VIII, Div. 2, App-5 and FEA work, the following
equation could be used interchangeably with the previous equations:
SIF =
Range of Peak Stress due to M

2 X (Moment, M)/(Section Modulus, Z)
2 X (Pl + Pb + Q + F)
2 X (M)/ (Z)
or,
SIF =
Alternating Peak Stress due to M

(Moment, M)/(Section Modulus, Z)
(Pl + Pb + Q + F)
(M)/ (Z)
The peak alternating stress, (Pl+Pb+Q+F) is usually determined from finite

element analysis. Normally, the peak stress is the product of the secondary
stress and a fatigue strength reduction factor (FSRF) [3]. For instance,
Pl+Pb+Q+F = FSRF(PL+Pb+Q)/ 2
eq.2 [9]
To implement this concept in FEA, these steps were followed:
FE discretization of the piping model.

Applying a bending moment (or force) depending on what kind of BC we
are trying to impose.
Compute peak stress in the part.
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Compute the nominal stress in the attached piping.

Insert the peak stress and the nominal stress in the above equation to get
the B31-SIF.
Computation of Pl+Pb+Q involves finding the membrane + bending stress

intensity (twice maximum shear) or Von Mises scalar depending on how it is
defined in the code of reference. [3] requires use of Von Mises scalar. For shell
elements stresses at the inner and outer surface will be membrane + bending
and for continuum elements linearization [3] is needed to extract membrane +
bending stresses. Stresses at nodes along intersection curve (shell) has not been
considered [14][15]. FSRF used in finite element models is 1.5.
5:
Applicability of ASME B31.3 SIF and flexibility factor formulas
According to B31.3 [1] the validity of stress intensification and flexibility

factors has been demonstrated for D/T<= 100. The code also uses a word of
caution that the out-plane SIF as specified in the code can be unconservative
for 0.5<d/D<1.0 . This code also uses a word of caution on proximity of other
branches on SIF values.
6:
The finite element models
Fig (5). 12x 6 RFT out-of-plane model (20 Noded reduced integration
Brick) Branch SIF BC
8
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Fig (6). 12x 6 In-plane model (20 noded reduced integration Brick)
Branch SIF BC
Fig (7). 72x 48 Out-of-plane model (20 noded reduced integration Brick)
Branch SIF BC
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Fig (8). 72x 48 Out-of-plane SIF (20 noded reduced integration Brick)
Branch SIF BC both ends of header fixed
Fig (9). 72x 48 Out-of-plane SIF (8 noded reduced integration

isoparametric shell) Branch SIF BC both ends fixed
10
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Fig (10). 72x48 In-plane SIF (8 noded reduced integration isoparametric

shell) Branch SIF BC both ends fixed.
Fig (11). 72x48 In-plane SIF (8 noded reduced integration isoparametric

shell) Header SIF BC
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Fig (12). 72x48 Out-of-plane SIF (8 noded reduced integration

isoparametric shell) Header SIF BC
Fig (13). 72x48 In-plane SIF (20 noded reduced integration brick,
exploded view), Header SIF BC
12
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Fig (14). 72x48 Out-of-Plane SIF (8 noded full integration Brick),

Branch SIF BC both ends fixed.
7:
Results and discussions
The trend of Markl test results could not be simulated using FEA; i.e., Out-ofplane SIF> In-plane SIF. This trend could however be simulated if Branch SIF
BC is used and this is independent of D/T and d/D. For header SIF BC in-plane
SIF >Out-of-plane SIF and this is independent of D/T and d/D. No significant
effect of modelling (or simulating) the weld was observed. The trend of
behaviour was similar for element types (linear vs. quadratic and shell vs.
continuum). Use of continuum elements showed (in general) lower value of
SIF. The pattern of behaviour is similar using three FEA codes. A stress
analyst who wants to apply FEA to compute SIF has to be careful about the BC
to be used; i.e., different BC is needed for header and branch SIF. ASME
B31.3 values for SIF are indeed underestimated for out-of-plane SIF for
0.5<d/D<1.0 but this can be observed only from using Branch SIF BC. There
was no major change in results when force-BC was used instead of Moment.
The effect of pad width was checked for the two parameters: Pad width =
2.5 RT and half OD of branch pipe. When d/D is less than 0.5, (e.g., for the
36x6 Connection, Table 1, 2.5 RT is 190 mm and Half OD is 84 mm) the
results do not show significant change. However, for d/D>=0.5 the change in
result was a 10-20% reduction in both in-plane and out-of -plane SIF (Branch
BC) values for higher pad width. As an example, the value of 2.5 RT is
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
significantly different from that of half OD of branch for 72x48-connection.
For 2.5 RT it was 200 mm and for half OD of branches 609.6 mm. It is
recommended that minimum pad width should be greater of 2.5 RT and half
OD of branch pipe. Use of full integration 4 noded shell element showed lower
value of SIF compared to 8 noded reduced integration shell element. In some
cases the SIF value (out-of-plane) was lower by nearly 50% by the linear full
integration shell element (typical variance in out-of-plane SIF by using linear
full integration element is 20% for the cases considered). Use of linear
continuum elements (full integration) also showed similar behaviour. The
lower SIFs in linear full integration elements can be understood to be because
of the tendency to shear lock which these elements typically exhibit.
Table 1
Description
12X6
36X6
48X6
72X48
72X56
D/T
34
72
96
262
262
d/D
Header SIF
Shell In-plane
Header SIF
Solid In-plane
Header SIF
Shell Out-ofplane
Header SIF solid
out-of-plane
Branch SIF
Shell In-plane
Branch SIF
Solid In-plane
Branch SIF
Shell Out-ofplane
Branch SIF solid
out-of-plane
ASME B31.3
SIF In-plane
ASME B31.3
SIF Out-ofplane
0.5
0.18
0.13
0.66
0.77
1.18
1.28
1.35
3.59
4.19
1.61
1.82
1.96
3.98
4.83
1.03
1.06
3.5
2.89
3.1
13
8.97
2.1
2.03
2.53
10.2
7.92
6.8
4.19
4.26
58.6
47
3.72
2.94
3.16
51.7
43.6
1.95
2.34
2.81
9.07
9.07
2.34
2.92
3.55
11.76
11.76
14
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
8:
Conclusions
ASME B31.3 SIF trend (i.e., out-of -plane SIF>In-plane SIF) which is
based on Markls tests could not be replicated using FEA.
ASME B31.3 SIF trend however can be replicated using Branch SIF
conditions (one or both ends of the header fixed and load applied at one end
of the header).This is independent of d/D and D/T.
Modelling of welds [11][13] did not show any significant effect on the peak
stress computation at the intersection of pad with header or branch.
However, in this study, stresses in welds were not checked using FEA.
ASME B31.3 should specify different SIF for header and Branch.
For the header SIF, in-plane SIF is greater than out-of-plane SIF.
ASME B31.3 method of specifying branch SIF for both header and branch
can be over-conservative for small d/D.
ASME B31.3 method uses a word of caution for lack of conservatism in
out-of-plane SIF for 0.5<d/D<1.0. This has been found justified with factor
of lack of conservatism as high as 5.0.
ASME B31.3 SIF values which are valid for D/T <=100 should not be used
for higher values of D/T as lack of conservatism in out-of-plane SIF can be
significantly high, a trend which increases with d/D ratio.
Shell Quadratic elements have shown very good results when compared
with solid elements(quadratic).
Linear full integration elements (both in solid and shell versions) have
shown lesser value of SIF with respect to their respective quadratic reduced
integration counterparts. Results of linear reduced integration elements
show better results with respect to full integration elements clearly showing
the effects of shear locking. In general, quadratic and reduced integration
elements are recommended both in shell and continuum versions.
The effect of pad width has been shown to reduce SIF with significant
reduction (10-20%) when 2.5 RT and half OD of branch Pipe are
significantly different.
ASME B31.3 equations for SIF of reinforced fabricated tees (RFT) should
include pad width as a parameter.
Further parametric study is required for wider range of D/T, d/D and t/T to
arrive at an empirical formula to compute SIF for D/T >100 and
0.5<d/D<1.0
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
Further study is required to investigate the effect of proximity of other
branch connections on SIF.
Further study is required to investigate the effect of pressure on stress
intensification factors.
Users of beam based finite element codes who want to use SIF computed
using shell/continuum based FEA should use separate SIF for header and
branch.
ACKNOWLEDGEMENT
The author wishes to thank Mr. Chris Hinnant of Paulin Research group,
Houston Texas; Dr. S. Saha of Reliance Industries Ltd, Jamnagar, India; and
Mr. Bruno Notarberardino of Simulia UK for interesting discussions and
valuable suggestions.
9:
References
1. Published by American Society of Mechanical Engineers, 2008 Edition,
ASME B31.3 -Process Piping.
2. A.R.C Markl, ASME Transactions 1955,Piping Flexibility Analysis,
ASME Boiler and Pressure Vessel code SEC VIII, Div-2, Part-5.
ASME Boiler and Pressure Vessel code SEC III Division 1- Rules for
Construction of Nuclear Facility Components.
5. Program Manual FE/Pipe Version 5.0, NOZZLEPRO Version 7.5( a
product of Paulin Research Group, Houston Texas)
6. Chris Hinnant ( Paulin Research Group, Houston Texas)-Private
communication
7. E.C. Rodabaugh. August 1988,Welding research council Bulletin 335A Review of Area Replacement Rules for Pipe Connections in
Pressure Vessels and Piping,
8. E.C.Rodabaugh - June 1983,NUREG CR-3243 - Comparisons of
ASME Code fatigue evaluation methods for Nuclear Class 1 piping
with Class 2 or 3 Piping
16
NAFEMS 2011
www.nafems.org

FABRICATED TEES.
9. Markl, SIF's and ASME VIII -2 Fatigue Design available at
www.paulin.com
10. A.Bhattacharya, Daniel Long, NAFEMS UK conference, June 2010,
Oxford, UK. A Finite Element based investigation on stress
intensification factors and flexibility factors for pipe bends within and
outside the limitations of ASME B31 piping codes
11. ABAQUS Version 6.9-1-A product of HKS Inc RI, now marketed
under the SIMULIA brand of Dassault Systems S.A.
12. T.Hellen NAFEMS publication How to use Beam, Plate and Shell
Elements
13. Richard H. Macneal, Marcel Dekker Inc.1993 Finite Elements: Their
Design and Performance.
14. J.Wood- NAFEMS Publication Procedural benchmarks for common
fabrication details in Plate/Shell Structures.
15. W. Koves, K. Mokhtarian, E. Rodabaugh and G.E.O.Widera, Dec
2004,Large diameter ratio shell intersections- Welding Research
council bulletin No. 497.
16. L. Xue, G.E.O.Widera, Z. Sang, Journal of Pressure Vessel
Technology Feb. 2006 Vol.128. Flexibility factors for branch pipe
connections subjected to in-plane and out-of-plane moments
17. E.Rodabaugh , Dec 1987,Accuracy of stress intensification factor for
branch connections-Welding Research council Bulletin 329
18. T.Fujimoto and T.Soh, PVP Conference and Exhibition, New Orleans,
Louisiana, June 23-26, 1985 of the ASME. Flexibility factor and Stress
Indices for piping components with D/T>=100 subjected to In-Plane
and out-of-plane bending moment

A Finite Element Based Study On Stress Intensification Factors - SIF - For Reinforced Tees

Uploaded by

Copyright:

Available Formats

A Finite Element Based Study On Stress Intensification Factors - SIF - For Reinforced Tees

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Finite Element Based Study On Stress Intensification Factors - SIF - For Reinforced Tees

Uploaded by

Copyright:

Available Formats

NAFEMS 2011

REPRODUCTION AND DISTRIBUTION PROHIBITED