Proceedings of the ASME 2014 33rd International Conference on Ocean, Offshore and Arctic Engineering

OMAE2014
June 8-13, 2014, San Francisco, California, USA

OMAE2014-23066

ESTIMATING LONG-ENDURANCE FATIGUE STRENGTH OF GIRTH-WELDED

PIPES USING LOCAL STRESS APPROACH
Y-H Zhang S J Maddox
TWI Ltd, Cambridge TWI Ltd, Cambridge
CB21 6AL UK CB21 6AL UK
[email protected]

ABSTRACT scatter in fatigue test data, scatter that is particularly wide in the
Nominal and hot spot stresses are conventionally used for long-endurance regime. Thus, although the above design
fatigue design of welded joints. In this paper, use of local stress approach is convenient for designers, weaknesses are that the
approaches was attempted to estimate the long endurance long-endurance fatigue life and the constant amplitude fatigue
fatigue strength of girth welded pipes. Finite element modelling limit (CAFL) are not well defined and that it is not possible for
was carried out to determine the dependence of the stress the designer to make allowance for known characteristics
concentration and through-wall stress distribution on weld root related to the factors that influence the local stress
bead profile, with hi-lo values ranging from 0.25 up to 1.0mm. concentration and hence fatigue life. To overcome these
Two local stress approaches, critical distance and reference inadequacies, there is increasing interest in the development of
radius, were used to estimate the fatigue strength of girth welds local stress approaches that include allowance for the notch-
at 5x106 and 107 cycles, which were then compared with effect of the weld itself, for the fatigue design of welded joints.
available full-scale fatigue test results. To use the critical Local stress-based fatigue assessment approaches involve
distance approach, the relevant material properties, such as use of the stress range at the point of potential fatigue crack
threshold stress intensity factor range and fatigue limit for flush initiation under consideration in conjunction with a suitable S-
ground welds, were determined experimentally. This paper N curve. This might be an S-N curve for the same material
presents the results of the fatigue strength estimates and draws without a stress-concentrating feature, typically that for smooth
conclusions about the applicability of the local stress specimens when assessing machined notches. In this context,
approaches to girth welds. one way to determine the local stress range is to multiply the
applied nominal stress range by the maximum stress
Key words: concentration factor (SCF), Kt. This is a simple task if the
Fatigue, girth welds, risers, local stress, critical distance, assessment concerns a well-defined geometry for which a
reference radius, finite element analysis. published SCF is available. However, even then the approach
can prove to be too conservative, with the actual life being
INTRODUCTION significantly higher than that determined by this method. In
Fatigue design of welded joints is generally based on S-N other words, the SCF is higher than the fatigue strength
curves expressed in terms of either the nominal or structural reduction factor, Kf [3]. A simple example is a plate with a hole
hot-spot stress range [eg, 1,2]. Neither of these stresses includes in it, where Kt = 3.0. For a very small hole (for example, in
any allowance for the notch effect of the stress concentration at m), Kf is found to be 1.0. However, for a hole with a sufficient
the fatigue crack initiation site. Instead, this is built into the S-N size (for example, in mm), Kf approaches Kt. A number of
curve by basing it on the results of fatigue tests on actual researchers have attempted to produce methods of calculating a
welded specimens. That stress concentration depends on the local stress that takes account of this behaviour by assuming
local weld geometry which in turn depends on such factors as that the apparent increase in strength for small notches is due to
the welding process, the welding position and the type and the fact that the stress concentration occurs over only a small
extent of any welding flaws. Variations in these factors lead to volume of material. The underlying assumption is that, in order

1 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 for fatigue failure to occur, the average stress must exceed the limit as the crack size approaches zero. However, in reality, it is
fatigue limit of the material without stress concentration, So, known that, when a crack is small, the fatigue limit will not be
over some critical volume surrounding the notch, rather than as sensitive to the change of crack size as predicted by Eq.1 and
the maximum stress range. Neuber [4] proposed a there will be a definite fatigue limit of So for a smooth,
microstructural support concept which considers the effect of a uncracked body.
stress gradient on fatigue behaviour. The effect can be taken To calculate the fatigue limit for components containing
into account by different hypotheses in the elastic stress small cracks, Haddad et al [11] proposed the following
analysis of a notch. For simplicity of calculation, this approach empirical equation by introducing the material constant ao:
is often reduced to a consideration of the stress at a single point
[3,5-7], the stress averaged over a given distance [4,6-8] or the K
stress averaged over an area [7,9]. The resulting local stress S  [2]
F ( a  ao )
might still be suitable for comparison with the notch-free S-N
curve for the material or, and certainly in the case of welded
joints, a better comparison might be with an S-N curve derived where S is the stress range and K is stress intensity factor
from fatigue test results obtained from notched samples that are range. When K is equal to Kth, S is equal to Soc. By noting
expressed in terms of the same local stress. that, when crack size a approaches zero, Soc= So. ao can be
Steel catenary risers (SCRs) are increasingly used for deep- estimated from:
water oil and gas developments. Pipes are joined by girth welds
commonly made from one side. The fatigue performance of the 1 K th 2
ao  ( ) [3]
welds, especially in the long endurance regime, is a major  FSo
design consideration since SCRs can experience cyclic loading
from a variety of sources, including very high numbers of low
The introduction of the parameter ao was to overcome the
stress cycles due to vortex induced vibration (VIV). Thus, a
difficulty of applying LEFM in the small crack regime. The
local approach that improves the accuracy of long-endurance
above equations have been used to calculate fatigue limits for
fatigue life or CAFL estimations could be highly relevant. A
cracked components [7-8].
recent TWI project involving fatigue tests of girth welded pipes
Taylor [7] found that ao is an important material parameter
included comprehensive post-test examination of the welds
which can also be used to estimate the fatigue limit for notched,
with particular attention to the geometry in the weld root region
but uncracked, components. He observed that the stress at a
where fatigue cracking initiated [10]. This provides a good
distance of ao/2 from the maximum stress location, the average
opportunity to examine the suitability of local stress approaches
stress over a line distance of 2ao and the average stress over an
for girth welds. Thus, two widely recognised approaches,
area within a radius of ao from the surface with the maximum
critical distance and reference radius, were used to estimate the
stress were almost identical to the fatigue limit of the uncracked
long-endurance fatigue strengths of girth welds.
material (So). He termed the three stresses point, line and area
respectively. Thus, the stress distribution approaching a notch
OUTLINE OF THE LOCAL STRESS APPROACHES
(as seen later, in Figure 6) could be used to determine the
USED IN THE ANALYSIS
fatigue limit of the notched component as the applied stress
needed to produce the material’s So value at these three
Critical Distance Approach
This approach was first proposed by Tanaka [6] and further locations. Indeed, he found that the same method produced
developed by Taylor [7] after analysing the stress distribution reasonable estimates of the long-endurance fatigue strength,
ahead of a crack and relating the average stress to a material endurances less than, but approaching, the fatigue endurance
constant ao, originally proposed by Haddad et al [11]. For a limit. In such an application, So would be the fatigue strength
component containing a crack, the fatigue limit, Soc, can be of the plain material at the endurance of interest. The cases
estimated using linear elastic fracture mechanics (LEFM) by concerned were a wide plate containing a circular hole [7], a T-
the following equation: shaped fillet welded joint and a butt weld [14].
It will be evident that the fatigue strengths estimated by the
above methods depend on ao, which in turn depends on the
K th
Soc  [1] parameters in Eq.3. The effect of changes is not clear cut. A
F a decrease in F would increase ao and hence increase the fatigue
strength estimate. However, the value of ΔSo influences both ao
where Kth is the threshold stress intensity factor range, F is a and the stress to be compared with that obtained from the stress
parameter dependent on geometry, crack size and shape which distribution curve such that the estimated fatigue strength of the
is equal to 1.12 for an edge crack [12], and a is crack length. detail under consideration may be increased or decreased. This
However, when the crack size is very small, less than about is considered later when the method is applied to girth welds.
1mm, the above relation becomes invalid, due to the so-called The apparent success of the approach for estimating the
crack size effect [13]. LEFM predicts an almost infinite fatigue fatigue strength of welded joints is surprising. A key parameter

2 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 in the definition of the critical distance ao is the fatigue limit for GIRTH-WELD DATA USED TO APPLY LOCAL STRESS
the un-notched material. This implies a connection between the APPROACHES
fatigue performance of a welded joint and that of the parent
metal, whereas none exists. This is particularly the case when Fatigue Performance of Full-Scale Girth Welded Pipes
high-strength materials are considered. In general, the fatigue Details of the material, testing conditions and fatigue
strength and fatigue limit of the plain material increase with performance of the girth welds are contained in [10]. In brief,
increase in tensile strength whereas those for welded joints are 406mm outside diameter by 19.1mm wall thickness seamless
independent of tensile strength. Nevertheless, in spite of these steel pipes to API 5L-X70 specification were tested. The girth
misgivings it was decided to investigate the application of the welds were made from one side using PGMAW/GMAW
approach to girth welded joints in steel pipes. processes, in the 5G position. The weld cap at each start/stop
position was ground flush with the pipe surface. All welds were
Reference Radius Approach acceptable to a typical SCR weld quality specification. Each
This is also an average stress approach that represents a weld was instrumented with strain gauges to allow
sharp notch by a reference round notch. The basic idea is that measurement of axial strain and its variation around the
the stress reduction due to averaging the stress over a certain circumference on the outside surface of the pipe. The nominal
distance from the notch tip can alternatively be achieved by strains were used in conjunction with axial misalignment
introducing a fictitious, enlarged notch radius. Neuber [15] measurements to calculate the stress range in the region of
proposed the following equation for the fictitious radius, f : fatigue crack initiation. The tests were performed in resonance
fatigue testing machines at around 30 Hz with an axial tensile
f=  + s* [4] mean stress of 150MPa, achieved by pressurising the pipes
internally with tap water.
where  is actual notch radius, s is a factor based on a stress A total of eleven girth welds were tested to establish the
multiaxiality and strength criterion and * is a dimension constant amplitude (CA) S-N curve. Fatigue cracking always
analogous to Haddad’s material constant ao that would need to initiated at the weld root and propagated through the pipe wall
be established experimentally. For welded joints, Radaj et al to the outside. Regression analysis produced a mean S-N curve
[16] assumed s to be 2.5 for plane strain conditions, which was of the familiar form Sm.N = C, where m and C are material
based on the stress analysis under different loading conditions constants, with m = 3.25 and C = 2.02 x 1013, slightly shallower
[17], and * to be 0.4 for low-strength steel. This results in an than the design S-N curves for weld toe or root failure, for
increase of the actual radius by 1mm. For the most severe stress which m = 3. However, the value of m = 3 is within 95%
concentration,  can be assumed to be zero, resulting in a confidence estimates of the slope and therefore it would be
reference radius of 1mm. Applying this method to welded legitimate, because of the small number of tests, to assume this
joints, the actual sharp notch (e.g. weld toe or root) is replaced value when fitting the S-N curve, giving S3.N = 6.174 x 1012.
by a reference radius of 1mm in the FE model and the The test results and the mean, lower (mean minus 2 standard
maximum stress on the radius taken to be the local stress. deviations of log N) and upper bound (mean plus 2 standard
Based on fatigue test results obtained from a variety of welded deviations) curves are plotted in Figure 1 where they are seen to
joints expressed in terms of this local stress range, the IIW be generally better than the BS 7608 [1] Class D mean S-N
FAT-225 S-N curve (the fatigue strength at N = 2 x 106 cycles curve. The lowest stress range that produced failure, with an
being 225 MPa) has been proposed for design [18]. This endurance of 4.4x107 cycles, was 55.4MPa. Thus, it appears
approach has also been included in a DNV code [2]. Although that the CAFL for these girth welds may correspond to an
less widely used, a reference radius of 0.05mm has also been endurance between 5x107 and 108 cycles which, considering
proposed [19,20], for assessing very thin (sheet) specimens, also the two unbroken welds that survived for 108 cycles, would
related to the much higher FAT-630 design curve. The current be between 45 and 55MPa. It is apparent that the Class D mean
IIW recommendation is that it should only be needed for fatigue limit of 74MPa, corresponding to 107 cycles, over-
thicknesses up to 5mm. Although the 0.05mm radius would not estimates the value suggested by the present test results.
be considered suitable for assessing the girth welds, this Extensive post-test examination of failure locations and
approach was explored here for comparison. macro-sectioning of unbroken welds were carried out. The
To evaluate the accuracy of the present fatigue strength fatigue performance of the girth welds strongly depended on
estimates, the corresponding mean S-N curves should be used. the profile of the weld root bead toe where the fatigue cracks
According to Sonsino [21], the mean curves for both FAT 225 always initiated. Poor weld root profile, which was indicated by
and FAT 630 can be estimated by multiply the FAT number by a large step between the surfaces of the weld root bead and the
1.37. This results in fatigue strengths at 2 x 106 cycles of adjacent pipe, created a high stress concentration.
308MP for the 1mm reference radius and 863MPa for the In the present girth welds the step was equivalent to the hi-
0.05mm. The estimated fatigue strength was equivalent to the lo, the step between the pipe surfaces either side of the weld,
nominal stress when the local stress was equal to the fatigue because of the very flat weld root bead, as indicated in Figure
strength corresponding to the designated mean S-N curve. 2. For each girth weld, the majority of the weld along the pipe

3 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 c)
FIGURE 2 Macro-sections of two girth welds:
a) Low magnification view of weld W2 in pipe S13 which was
FIGURE 1 Fatigue test results obtained from full-scale girth- 6
un-broken after 4.0x10 cycles at a stress range of
welded pipes. 116MPa;
b) Higher magnification view of weld W2 in pipe S13. The
circumference exhibited a relatively good weld root profile, crack and the hi-lo were respectively 0.32mm and 0.5mm at
with hi-lo values less than 0.2mm. However, regions with poor the section.
weld root bead profile were always found along the weld and c) Macro-section of weld W2 in pipe S10 which was unbroken
7
crack initiation often occurred at a location with the poorest after 2.97x10 cycles at a stress range of 72MPa. The crack
weld root bead profile. Figure 2 shows two examples. In the and the hi-lo were respectively 0.24mm and 0.5mm at the
section.
first (Fig.2(a) and (b)), a small crack, 0.32mm deep, had
developed at a location with a hi-lo of 0.5mm after the weld
(Fig.2(c)), where the hi-lo was also 0.5mm, a 0.24mm deep had
had endured 4.0x106 cycles at 116MPa. In the second example
developed after 2.97x107 cycles at 72MPa. Both welds were
declared to be run-outs since failure occurred in other welds in
the specimens.
The worst hi-lo values from a total of 33 girth welds were
measured (the project involved further tests under variable
amplitude loading). They represented either the largest present
in an unbroken weld or the value at the failure location. The
average hi-lo value was 0.47mm. The measured hi-lo values of
these girth welds are shown in Figure 3. Statistical analysis
showed that 80% of them fall between 0.25 and 0.68mm.

Threshold Stress Intensity Factor Range

In order to calculate the material constant ao using Eq. 3,
the threshold stress intensity factor range, Kth, is required.
a) This was determined in the girth weld project using single-edge
notch bend specimens, notched in the welds which were at the
centre. The tests were carried out under decreasing K in
accordance with BS 12108 [22] at a stress ratio R=0.7 and a
loading frequency of 10Hz. Kth corresponding to a crack
growth rate of 10-7mm/cycle, in accordance with ASTM E647,
was determined to be 134N/mm3/2.

Fatigue limit of flush ground welds

In order to consider a welded joint, So in Eq.3 should be the
fatigue limit for flush ground girth welds. Relevant
Hi-lo experimental data from flush ground girth welds in strip
specimens cut from pipes [23] are presented in Figure 4.
b)

4 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 ESTIMATION OF LONG-CYCLE FATIGUE STRENGTH
OF GIRTH WELDS

Finite Element Analysis

A 2D axi-symmetric FE model, incorporating a girth weld
in a full-scale pipe, was generated using Abaqus software.
Eight-noded quadratic elements were used in a static, elastic
analysis. A nominal axial stress of 50MPa was applied to the
pipe. In all models, the outer pipe wall surfaces were assumed
to be aligned (no hi-lo at weld cap) and one side of the weld
root bead was flush with the inner pipe surface, while a hi-lo
was included on the other side of the bead, as found in most of
the girth welds.

FIGURE 3 The worst hi-lo values measured from a total of 33

girth welds.

loading

a)

FIGURE 4 Experimental results of flush ground girth welds [22]

and the S-N curve derived from these data. Stress distribution
Regression analysis of the data produces a mean S-N curve
with a slope of m = 4.2, compared with the value of 4 for the
relevant BS 7608 design curve (Class B). Re-analysis of the
data assuming m = 4 gives the mean curve shown in Figure 4, b)
FIGURE 5 The FE model and the stress contour calculated for the
S4N=5.88 x1015. It will be evident that the fatigue limit is not weld root bead profile with a hi-lo of 1.0mm:
obvious from the available data. Therefore, for the benefit of a) FE model;
the analysis of the girth welds values corresponding to the b) Meshing and the stress contour predicted.
fatigue strengths at 5x106 and 107 cycles were used to define
So. From the fitted mean S-N curve these are 185.2 and A fine mesh was adopted in the vicinity of the notch
155.7MPa, respectively. With the above Kth value, the created by the corner between the weld root bead and the pipe
material constant ao was calculated to be 0.133mm for the inner surface.
fatigue strength at 5x106 cycles and 0.188mm for the fatigue
strength at 107 cycles. Estimation by the Critical Distance Approach
It was noted that most of the above fatigue data for flush Four FE models were produced, with hi-lo values of 0.25,
ground girth welds were obtained at R = 0.1, lower than those 0.5, 0.75 and 1.0mm. Figure 5 shows that for a hi-lo of 1mm.
used in the tests on as-welded girth welded pipes where a Through-thickness stress distributions, starting from the weld
constant mean stress of 150MPa was applied. The impact of root on the inside, were determined at locations where the stress
this on the fatigue strength estimates is considered later. along the pipe axis was at a maximum. Stress/distance curves

5 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 were recorded along lines perpendicular to the local maximum
principal stress, which corresponds to the direction normal to
the surface. Two examples of the stress distributions are shown
in Figure 6. It can be seen that the local stress, and hence SCF,
increases with increase in hi-lo and decreases sharply within
0.1mm from the notch.
Only the first two of the three stress calculation methods
described before were used. The third, the area method, was not
used because of the difficulty of obtaining an average stress at
the hi-lo. To obtain an average stress over a certain line distance
for the line method, either a fourth-order polynomial equation
or a power function was first derived for each model to
characterise the stress distribution as a function of the distance
from the inside surface. In all cases, the correlation coefficient
(R2) for each curve was better than 0.95. By integrating the
equation from zero to 2ao, the average stress corresponding to
the line method was obtained. For the point method, the stress FIGURE 6 Examples of the stress distributions determined by
at the distance ao/2 from the inside surface was read from the FEA (the applied nominal stress was 50MPa).
stress distribution for each model. The estimated fatigue
strength was then equivalent to the nominal stress when the
stress at ao/2 was equal to the fatigue strength of the flush
ground welds for the particular number of cycles (So).
Similarly, for the line method, the estimated fatigue strength
was equivalent to the applied nominal stress when the average
stress over the line distance of 2ao from the notch surface was
equal to So.
Attention was focused on the long-endurance regime, in
particular the estimated fatigue strengths at endurances of 5x
106 and 107 cycles. The former was selected because it is the
assumed fatigue endurance limit (N corresponding to the
CAFL) in some design codes [e.g. 24] while the latter is the
more commonly accepted endurance limit, e.g. BS 7608 [1] and
IIW [25]. The estimated results for different hi-lo values are FIGURE 7 Dependence on hi-lo of the fatigue strength of the
given in Table 1 and plotted in Figure 7. It will be seen that the girth welds calculated by the critical distance approaches.
calculated fatigue strengths decreased with increasing hi-lo. For
example, for a fatigue endurance of 107, the estimated fatigue gave similar results, the line method often gave slightly higher
strength using the point method was about 85MPa for a girth strengths than those obtained by the point method. A good
weld with a hi-lo of 1mm. Although the point and line methods estimate was made of the fatigue strength of the full-scale pipes
at 107 cycles, but only by assuming a hi-lo value of 1.0mm. The
Table 1 Fatigue strengths for different weld root bead profiles estimate based on the actual average hi-lo was reasonable only
calculated using critical distance approaches at two different at the lower endurance of 5x106 cycles
endurances (107 and 5x106 cycles).
Estimation by the Reference Radius Approach
Hi-lo, For endurance of 107 cycles (ao=0.236mm)
mm So, full- Point method Line method Since the wall thickness of the girth welded pipes
scale Calculated Difference Calculated Difference considered in the present analysis was greater than 5mm, the
girth weld, So, MPa from So, MPa from relevant reference radius to be used when calculating the local
MPa experiment experiment
stress is 1mm. However, for comparison the value normally
0.25 119.8 41% 121.7 43%
0.50 101.9 20% 105.9 24% confined to assessments of welded joints in plates ≤ 5mm was
85.2 also considered. Two FE models were established for the 1mm
0.75 91.0 7% 95.4 12%
1.0 84.8 0% 87.7 3% reference radius, one with a hi-lo of 0.5mm and the other 1mm.
For endurance of 5x106 cycles (ao=0.167mm) In addition, four FE models were established for the 0.05mm
0.25 131.2 22% 137.2 28% reference radius , with a hi-lo values of 0.25, 0.5, 0.75 and
0.50 109.6 2% 117.4 9% 1mm. Figure 8 shows two FE models with the same hi-lo of
107.3
0.75 97.0 -10% 104.0 -3%
1.0 90.4 -16% 95.6 -11%
1mm, one with a radius of 1mm and the other 0.05mm.

6 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 7 6
Table 2 Fatigue strengths at 10 and 5x10 cycles calculated The estimated fatigue strengths at 107 and 5x106 cycles for
using the reference radius approach [9,25] the two reference radii and different hi-lo values are compared
with the experimental values in Table 2. As the reference S-N
For endurance of 107 cycles curves have the same slope as that for the full-scale girth welds
So, full-scale Calculated (m = 3), the differences between the actual and calculated
Difference
Radius Hi-lo, girth weld, fatigue
strength, MPa
from experiment fatigue strengths were the same for both endurances. It can be
mm MPa
0.5 99.6 17% seen that, for both reference radii, the calculated fatigue
1mm strengths at both long endurances agreed well with the actual
1.0 83.5 -2%
0.25
85.2
146.4 72% fatigue strengths for a hi-lo value of 1mm, with the differences
0.05mm
0.5 114.8 35% being less than 5%. However, for a hi-lo of 0.5mm, which was
0.75 98.7 16%
about the average of the worst hi-lo values of the 33 girth welds
1.0 88.3 4%
For endurance of 5x106 cycles examined, both methods over-estimated the actual fatigue
0.5 125.5 17% strength, by 17% for the 1mm reference radius and 35% for the
1mm
1.0 105.1 -2% 0.05mm reference radius.
0.25 184.4 72%
107.3
0.5 144.7 35%
0.05mm
0.75 124.4 16%
1.0 111.3 4%

FIGURE 9 Dependence on hi-lo of the fatigue strength of the

girth welds calculated by the critical distance approaches based
on flush-ground weld CAFL corrected for mean stress.

a) DISCUSSION
The 1mm reference radius approach is considered first
since, in view of its widespread use in some industries and even
acceptance by some design standards, it is the least
controversial of the two fatigue strength calculation methods
investigated here. Referring to Table 2, good estimates of the
actual fatigue strengths at 5x106 and 107 cycles, within 5%,
were achieved. However, this was the largest hi-lo seen among
the 33 welds examined. Assuming a hi-lo of 0.5mm, the case
closest to the average actual hi-lo of 0.47mm, over-estimated
the fatigue strengths at the two endurances considered, by 17%.
This might reflect a deficiency in the analysis method, but it
might also be an indication that the IIW FAT 225 classification
is too high. Although it was derived from experimental data
[18] subsequent check fatigue tests on fillet-welded joints have
produced significantly lower lives than expected for this S-N
b) curve [26]. The specimens concerned were produced using
FIGURE 8 The FE model and the calculated stress contour for industrial production welding conditions whereas the database
the weld root bead profile with a hi-lo of 1.0mm (in this case, hi- used to derive the FAT 225 curve was produced using
lo=weld root bead height, 18.5/19.5mm wall thickness, applied
specimens made in laboratory conditions. Their lower fatigue
nominal stress=50MPa):
a) Corner radius = 1mm strengths were attributed to poorer weld profiles and the
b) Corner radius = 0.05mm. presence of undercut or cold laps, although the welds met
typical weld quality acceptance limits. It is certainly possible

7 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 that the quality of the girth welds considered here was more for R=0.1 and the other for R=0.47 (or 0.56), the item ΔSR=-1
comparable with the industrial specimens than the laboratory can be deleted and the ΔSo, with the same stress ratios as those
ones. The above fatigue strength estimate for 0.5mm hi-lo for the full-scale pipes, at 5 x106 and 107 cycles, were
would suggest that FAT 192 is a better choice than FAT 225 for calculated to be 159.1 and 133.8MPa, respectively. Working as
industrial welds, very close to the FAT 194 classification before, these values were used to calculate the fatigue strengths
suggested on the basis of the results from the industrial fillet at the two endurances for a series of hi-lo values. The results
welds [26]. are presented in Figure 9. Comparing these with those in Figure
Turning to the results of the analysis assuming a 0.05mm 7, it can be seen that the lower ΔSo values have improved the
reference radius, rather similar fatigue strengths were obtained correlation between estimated and actual girth weld fatigue
as for 1mm reference radius for the 1mm hi-lo case (see Table strengths by reducing the estimated values, but only slightly.
2). However, the error for 0.5mm hi-lo was even greater than The same was found by Taylor et al [7] in fatigue strength
for the 1mm reference radius. In view of this and bearing in estimates for some butt and fillet welds. In the case of the
mind the need for more detailed finite element modelling, it parameter F, it would be justifiable to assume a lower value
will be evident that there is no advantage to be gained from the than 1.12 on the basis that fatigue cracks are generally semi-
use of the 0.05mm reference radius compared with the elliptical in shape or for finite plate. However, this will simply
recommended 1mm radius for wall thicknesses greater than increase the estimated fatigue strengths for the girth welds
5mm. making those estimates even more inaccurate. Assuming the
The results obtained using the critical distance approach lowest possible value of F = 1 over-estimates the fatigue
are summarised in Table 1 and Figure 7. As with the reference strengths for 0.5mm hi-lo by 28% (point method) to 29% (line
radius method the best estimates of the fatigue strength at 107 method) at 107 cycles and 9-14% at 5 x 106 cycles. Thus, overall
cycles were obtained by both the point and line method for a the deficiencies noted earlier still stand and this method does
1mm hi-lo. For the actual average hi-lo of around 0.5mm the not seem to be well suited for application to welded joints.
fatigue strength was over-estimated by 20 to 24%. At 5x106
cycles, reasonable estimates were obtained for 0.5mm hi-lo CONCLUSIONS
while those for 1mm hi-lo were underestimated by 11 to 16%.
These variations are consistent with the fact that the slope of FE modelling was carried out to determine the through-
the S-N curve upon which the fatigue strength estimates were wall stress distribution and stress concentration at the weld root
based, that for flush-ground butt welds, was shallower (higher bead toe. Two local stress approaches, critical distance and
m value) than that obtained from the fatigue tests of the girth reference radius, were used to estimate the long-endurance
welds. This will generally be the case when the method is fatigue strengths of girth welds. The following conclusions
applied to details incorporating relatively severe stress were drawn:
concentrations, like most welded joints. A further deficiency of
the approach is that it proved to be less accurate at 107 cycles 1 FEA confirmed that the hi-lo in a girth-welded joint
than 5x106 cycles for the relevant hi-lo, when the opposite influenced the local stress, the SCF increasing with
would be expected since the method is primarily directed at increase in hi-lo.
estimates of fatigue limits. However, recalling the comments 2 Although both critical distance methods provided
made earlier about the parameters that influence ao, further reasonable estimates of the fatigue strength of the girth
analysis was carried out to assess the effect of making different welds at 5 x 106 cycles assuming the average measured hi-
assumptions. In particular, it was noted that the experimental lo, it was over-estimated considerably (20 to 24%) at 107
data used to determine ΔSo (Figure 4), and hence ao, were cycles, the endurance regime where the method was
obtained at R = 0.1whereas values of 0.47 and 0.56 were used expected to be best suited. It was necessary to assume the
for the girth weld tests [10] at the two endurances investigated maximum measured hi-lo to achieve reasonable fatigue
here. The fatigue strength of the flush-ground welds would be strength estimates.
expected to be lower at such high stress ratios, resulting in 3 Legitimate modification of the input parameters used to
higher values of ao. To investigate this, ao was re-calculated define the critical distance had little effect on the fatigue
using a value of the CAFL for the flush-ground welds modified strength estimates.
by applying the Goodman mean stress correction [27]: 4 In general the local distance approach does not seem to be
well suited for application to severely-notched components
Sm like welded joints.
S R  S R  1( 1  ) [5] 5 The reference radius method also proved to be inaccurate
SUTS
when related to the average hi-lo but this could be
explained on the basis of the assumed reference S-N curve
where SUTS is the tensile strength of material (601MPa), ΔSR=-1 rather than any deficiency in the principle of the method.
is the fatigue limit at stress ratio R = -1 and ΔSR is the fatigue As has been found in the application of the method to other
limit to be estimated for constant amplitude cyclic loading with industrial welds, the currently accepted IIW reference S-N
a mean stress of Sm. With two equations of the above form, one

8 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 curve FAT 225 should be reduced to around FAT 190 for mixed-mode loading', Eng. Fract. Mech., 44 (1993) 5, 691-
general design. 704.
6 No advantage can be gained from the use of the 0.05mm 18 Hobbacher A, 2008: 'Database for the effective notch stress
reference radius currently recommended for assessing method at steel', IIW Joint Working Group Doc. JWG-
welded joints in thin material (<5mm) in the assessment of XIII-XV-196-08, International Institute of Welding.
thicker sections when the 1mm radius is sufficient. 19 Sonsino C M, 2009: 'A consideration of allowable
equivalent stresses for fatigue design of welded joints
REFERENCES according to the notch stress concept with the reference
radii rref=1.0 and 0.05mm', IIW-1950-08, Welding in the
1 BS 7608, 1993: 'Fatigue design and assessment of steel World, 52 (2009),No.3/4, R64.
structures', British Standards Institution, London. 20 Fricke W, 2008: 'Guideline for the fatigue assessment by
2 Det Norske Veritas (DNV), 2010: 'Fatigue design of notch stress analysis for welded structures', IIW-Doc. XIII-
offshore steel structures', DNV-RP-C203, Norway. 2240-08/XV-1289-08.
3 Peterson R E, 1959: 'Notch sensitivity', in Metal Fatigue, 21 Sonsino C M, 2012, private communication.
edited by Sines G and Waisman J L, New York: McGrow, 22 BS ISO 12108, 2002: 'Metallic materials - Fatigue testing -
293-306. Fatigue crack growth method', British Standards
4 Neuber H, 1958: 'Theory of notch stresses', Spinger, Institution, London.
Berlin. 23 Razmjoo G R, Maddox S J and Hayes B, 1998: 'Fatigue
5 Lawrence F V, Mattos R J, Higashida Y and Burk J D, performance of flush-ground TLP tendon girth welds', TWI
1978: 'Estimating the fatigue crack initiation life of welds', Report No. 5680/13/98, July.
in Proc. of Fatigue Testing of Weldments, ASTM STP 648, 24 Eurocode 3, 2009: 'Design of steel structures – Part 1-9:
edited by Hoeppner D W, pp.134-158. Fatigue' (incorporating corrigenda), European Standard,
6 Tanaka K, 1983: 'Engineering formulae for fatigue strength April, 2009
reduction due to crack-like notches', International Journal 25 Hobbacher A, 2008: 'Recommendations for fatigue design
of Fracture, 22 (1983) R33-R45. of welded joints and components', IIW document IIW-
7 Taylor D, 1999: 'Geometrical effects in fatigue: a unifying 1823-07, December 2008.
theoretical model', International Journal of Fatigue, 21 26 Barsoum Z and Jonsson B, 2007: 'Fatigue strength and
(1999) 413-420. weld defect assessment of cruciform joints fabricated with
8 Livieri P and Tovo R, 2004: 'Fatigue limit evaluation of different welding processes', IIW Document XIII-2175-07.
notches, small cracks and defects: an engineering 27 Goodman J, 1919: 'Mechanics applied to Engineering',
approach', Fatigue Fract. Engng. Mater. Struct., 27 (2004) Longmans, Green and Co, London, pp.631-636.
1037-1049.
9 Radaj D, 1990: 'Design and analysis of fatigue resistant
welded structures', Abington Publ., Cambridge.
10 Zhang Y H and Maddox S J, 2012: 'Fatigue testing of full-
scale girth welded pipes under variable amplitude loading',
Proc. of the 31st Int. Conf. on OMAE, Paper No.
OMAE2012-83054.
11 El Haddad M H, Dowling N F, Topper T H and Smith K
N, 1980: 'J Integral applications for short fatigue cracks at
notches', Int. J Fracture, 16 (1980) 15-24.
12 Murakami Y, 1992: 'Stress intensity factors handbook',
Pergamon Publishing.
13 Surech S and Ritchie R O, 'Propagation of short fatigue
cracks', Int. Metals Review, 29 (1984) 445-476.
14 Taylor D, Barrett N and Lucano G, 2002: 'Some new
recent methods for predicting fatigue in welded joints',
International Journal of Fatigue, 24 (2002) 509-518.
15 Neuber H, 1968: 'Consideration of stress concentration in
strength assessment' (English translation), Konstruktion 20
(1968) 7, 245-251.
16 Radaj D and Sonsino C M and Fricke W, 2006: 'Fatigue
assessment of welded joints by local approaches',
Woodhead Publishing, Cambridge.
17 Radaj D and Zhang S, 1993: 'On the relation between
notch stress and crack stress intensity in plane shear and

9 Copyright © 2014 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/07/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use