Acceptance Levels of W e l d Defects
for Fatigue Service
The derivation and use of standards for the acceptance
of defects in welded joints undergoing fatigue loading is
considered with fitness for purpose as a major criterion
Introduction
The rapid development of nonde-
structive testing (NDT) techniques and
the improvement in operator efficien-
cy that has occurred in recent years
have resulted in more and more w e l d '
defects being located. In existing-fabri-
cation standards, the basis of accept-
ance criteria for weld defects appears
to be arbitrary. Generally, such stan-
dards demand one universal level of
quality, regardless of the service con-
ditions of the joints in question.
Research has shown that all welded
joints will contain flaws; whether or
not they are detected will depend
upon the sensitivity of the NDT
method employed. Clearly the remov-
al of such flaws may be unnecessary
and the use of a standard which
demands their removal becomes ridic-
ulous—particularly when, with the aid
of hindsight, it is known that welded
joints containing the very same flaws
have behaved perfectly satisfactorily in
service, and no one has been con-
cerned w i t h their presence as no one
had detected them. Because of this,
emphasis is now being placed on
defect acceptance based upon a "fit-
ness for purpose" approach whereby
flaws which could not give rise to
premature failure of the structure or
joint in question are left intact.
Despite the above comments it can
be argued that very few structures
w h i c h have been inspected to existing
arbitrary codes have failed. Moreover,
it w o u l d be difficult to establish that
those which had failed had done so
specifically because of the weld
acceptance standard employed. The
question is thus posed: W h y is there
any need to change?
The arguments for adopting realistic
acceptance standards are t w o - f o l d .
The first is economic. Salter and
Gethin (Ref. 1) have carried out a
survey of repairs actually carried out to
BY C. F. BOULTON
the main seams of pressure vessels by a
number of British fabricators. They
found that 87% of weld metal removed
was to eliminate three dimensional
weld defects (84% for slag inclusions,
3% for porosity). Based on a fitness for
purpose assessment, all these defects
w o u l d have been considered accept-
able. The remaining 13% of weld metal
was removed to eliminate t w o - d i m e n -
sional (planar) defects, of w h i c h some
but not all were considered unaccept-
able based on fitness for purpose. The
cost involved in making unnecessary
repairs does not need to be ex-
pounded here.
Of greater technical importance, the
repair of harmless three-dimensional
defects may introduce more harmful,
and less easily detectable, planar
defects. Repair welding is often diffi-
cult to carry out satisfactorily since the
repair welds are usually not made
under such favorable conditions as
were present when the original weld
was deposited. For example, repair
welds must often be made under
conditions of difficult access or high
restraint. The introduction of cracking
while repairing harmless defects has
been reported by Lundin (Ref. 2). O n
the other hand, failures from small,
apparently innocuous defects have
initiated failures (Refs. 3, 4). Thus there
is ample evidence that, based on "fit-
ness for purpose", acceptance stan-
dards could be relaxed, but also should
be tightened up in certain areas.
The literature on the effect of weld
defects on service performance is vast
and steadily increasing. It is a major
task to interpret and edit this informa-
Paper presented at the AWS 57th Annual"
Meeting held in St. Louis, Missouri, during
May 10-14, 1976.
C F. BOULTON is Senior Research Engi-
neer, The Welding Institute, Abington,
Cambridge, England.
W E L D I N G RESEARCH S U P P L E M E N T 113-s
tion in such a way that it can be readily
used to quantify the effects of the
presence of weld defects on structural
integrity. However, a major step in this
direction has now been taken by the
British Standards Institution (BSI),
whose specialist committee dealing
with the subject issued a draft docu-
ment for public comment entitled
"Draft Standard Rules for the Deriva-
tion of Acceptance Levels for Defects
in Fusion Welded Joints" (Ref. 5).
It is not the purpose of this paper to
review the contents of the draft docu-
ment; for this the reader is referred to a
paper by Burdekin (Ref. 6). Operation
of defect acceptance based on fitness
for purpose obviously necessitates
some means of establishing the effect
of a particular defect on the strength
or integrity of a welded joint. This
paper deals with the derivation of
acceptance levels of weld defects
operating under fatigue conditions,
which is one of the primary failure
mechanisms. The defects considered
comprise those most commonly en-
countered and can be categorized
under one of the f o l l o w i n g headings:
1. Three-dimensional (non-planar)
defects: (a) solid inclusions (slag); (b)
gas pores.
2. Two-dimensional (planar) de-
fects: (a) cracks; (b) lack of penetra-
tion/fusion.
The recommendations presented in
the paper regarding defect severity are
based primarily upon the results of
laboratory testing and experience for
non-planar defects and on fracture
mechanics principles for planar de-
fects. It is hoped that the information
given w i l l be of assistance to fabrica-
tors, designers and licensing authori-
ties in the determination of which
flaws can be accepted w i t h o u t impair-
ing structural integrity. To this end the
information has been presented in
such a way that fatigue strength esti-
mates can be made by personnel w i t h
a m i n i m u m k n o w l e d g e of the fatigue
process itself. T h e assistance o f f a t i g u e
researchers a n d f r a c t u r e m e c h a n i c s
experts w i l l not be r e q u i r e d , p r o v i d e d
t h e s i m p l e g u i d e l i n e s l a i d d o w n are
followed.
M o s t o f t h e d a t a h e r e refers t o s t r u c -
t u r a l C - M n steels o p e r a t i n g i n a m b i e n t
t e m p e r a t u r e air e n v i r o n m e n t s ; h o w -
ever, t h e p r i n c i p l e s d e s c r i b e d s h o u l d
be applicable to any m a t e r i a l / e n v i r o n -
m e n t c o m b i n a t i o n . M e t h o d s of t a k i n g
into account other material/environ-
m e n t c o m b i n a t i o n s are d i s c u s s e d .
T h e f a t i g u e s t r e n g t h o f e a c h part of a
s t r u c t u r e w i l l , as o f t e n as n o t , b e
d e c i d e d by t h e l o w e s t f a t i g u e s t r e n g t h
j o i n t w i t h i n that m e m b e r . This b e i n g
t h e case, t h e r e is l i t t l e p o i n t
r e q u i r i n g a d e f e c t - f r e e b u t t - w e l d if it is
s i t u a t e d w i t h i n a g l o b a l stress f i e l d
w h i c h also e n c o m p a s s e s , f o r e x a m p l e ,
a load-carrying fillet w e l d w h i c h w i l l
exhibit a m u c h lower fatigue strength.
A c c e p t a n c e levels o f w e l d d e f e c t s
w i t h i n a s t r u c t u r e m u s t , t h e r e f o r e , be
r e l a t e d t o t h e f a t i g u e s t r e n g t h of other
joints w i t h i n that structure.
Table 1—Nomenclature
Crack length
Initial crack length or flaw size
• Final crack length for failure
• Constant in equation describing
fatigue crack propagation
Constant in hypothetical equa-
tion accounting for crack
propagation in an aggressive
environment.
D - Constant in equation describing
S-N data
D, • Constant in equation describing
S-N data in aggressive envi-
ronment
• Young's modulus of elasticity
- Constant representing value of
integral / * ! da/(Y\/wa)m
- Plane strain fracture toughness
- Exponent in equation describ-
ing fatigue crack propagation.
Represents the slope of the
data.
n - Exponent in equation describ-
ing S-N data. Represents the
slope of the data.
N - Fatigue life, also termed endur-
ance and cycles to failure
V, W , X Y, Z-Designated quality bands for
defect acceptance
Y - Variable depending upon
geometry of a fatigue crack
da/dN - Rate of fatigue crack propaga-
tion
HAZ - W e l d heat-affected zone
SCF - Stress concentration factor
S-N - Representation of fatigue data
in terms of stress, S, and cy-
cles to failure, N.
AK • Range of stress intensity factor
A7 Total range of strain
Aff Total range of stress
14-s I J A N U A R Y 1977
In m a n y cases, it is o b v i o u s l y u n n e c -
essary t o insist u p o n p e r f e c t i o n f o r
many welds w i t h i n a structure and
q u a l i t y s t a n d a r d s g e a r e d t o real service
requirements can s h o w considerable
e c o n o m i c return for a c o m p a n y . H o w -
ever, it m u s t b e stressed t h a t t h e a d o p -
t i o n of d e f e c t a c c e p t a n c e s t a n d a r d s
based o n f i t n e s s f o r p u r p o s e d o e s n o t
i m p l y t h a t a g e n e r a l l o w e r i n g in t h e
o r i g i n a l q u a l i t y of w e l d s is a c c e p t -
able.
In g e n e r a l , it costs l i t t l e m o r e to
make welds which are relatively
d e f e c t - f r e e t h a n it d o e s t o m a k e d e f e c -
tive welds. Thus, defect acceptance
s t a n d a r d s are i n t e n d e d t o be u s e d t o
d e c i d e w h e t h e r or n o t t o r e p a i r or
in reject a particular s t r u c t u r e w h e n a
p a r t i c u l a r d e f e c t h a p p e n s t o be p r e s -
e n t as a r e s u l t , f o r e x a m p l e , o f h u m a n
error. If a s t r u c t u r e c o n t a i n s m a n y
d e f e c t s , it is clear t h a t s o m e t h i n g
radical may be w r o n g w i t h the fabrica-
tion procedure w h i c h the fabricator,
himself, s h o u l d investigate.
P h i l o s o p h y of Flaw Assessment
The c o n v e n t i o n a l m e t h o d of repre-
s e n t i n g t h e f a t i g u e p r o p e r t i e s of m a n y
w e l d e d j o i n t s lies in t h e use o f t h e S-N
diagram w h e r e , for a given material
a n d j o i n t t y p e , t h e a p p l i e d stress is
p l o t t e d a g a i n s t e n d u r a n c e . I m p l i c i t in
t h e use o f S-N c u r v e s f o r d a t a r e p r e -
s e n t a t i o n is t h e fact t h a t all t h e r e l e -
vant v a r i a b l e s f o r a p a r t i c u l a r t y p e o f
j o i n t in a p a r t i c u l a r m a t e r i a l r e m a i n
a p p r o x i m a t e l y constant. Given these
c o n d i t i o n s , s c a t t e r in f a t i g u e d a t a m a y
be t a k e n i n t o a c c o u n t b y statistical
analysis, w h e r e b y a l o w e r c o n f i d e n c e
l i m i t o f d a t a c a n b e s p e c i f i e d . By s u c h
means a design curve to avoid fatigue
f a i l u r e can b e e s t a b l i s h e d .
The philosophy b e h i n d the deriva-
t i o n o f a c c e p t a n c e levels f o r n o n -
p l a n a r w e l d d e f e c t s (slag, p o r o s i t y ) lies
in t h e c o n c e p t o f t h e S-N d i a g r a m .
,05 2 3
Design life , cycles
Fig. 1-Quality bands for defect acceptance
B o t h t y p e s of d e f e c t c a n be c a t e g o -
r i z e d in t e r m s of s e v e r i t y by s i m p l e
g e o m e t r i c m e a s u r e m e n t s — i n t h e case
o f slag i n c l u s i o n s , in t e r m s o f d e p t h
a n d l e n g t h a n d for p o r o s i t y in t e r m s o f
volume. Thus, comparisons between
the e x p e c t e d fatigue life of the defec-
t i v e j o i n t , t h e f a t i g u e lives of o t h e r
w e l d e d d e t a i l s p r e s e n t a n d t h e re-
q u i r e d s e r v i c e l i f e c a n be made
easily.
A c o n s i d e r a b l e v o l u m e o f research
has b e e n c a r r i e d o u t w o r l d w i d e i n t o
the effect of non-planar w e l d defects
o n f a t i g u e b e h a v i o r . T h e first a t t e m p t
t o survey a n d c o d i f y t h e results w a s
m a d e b y H a r r i s o n (Ref. 7, 8). H e
d i v i d e d t h e S-N d i a g r a m i n t o a n u m b e r
o f a r b i t r a r y areas, t e r m e d " Q u a l i t y
Bands", w h e r e b y each band was taken
to represent a certain q u a l i t y require-
ment. In fact five q u a l i t y bands
labelled V-Z were chosen, although
more or fewer c o u l d h a v e b e e n u s e d .
In t h i s a r r a n g e m e n t c a t e g o r y V p e r m i t s
n o f l a w s , w h i l e i n c r e a s i n g f l a w size is
p e r m i t t e d t h r o u g h c a t e g o r i e s W t o Z.
T h e q u a l i t y b a n d s are s e p a r a t e d b y
lines d r a w n at a s l o p e o f — 'A, w h i c h
fitted the fatigue crack propagation
l a w p r o p o s e d at t h e t i m e a n d c o n -
f o r m e d w i t h a n e a r l i e r t h e o r y (Ref. 9)
d e v e l o p e d to quantify the presence of
lack o f p e n e t r a t i o n d e f e c t s in t e r m s o f
fatigue strength.
T h i s s l o p e is n o w k n o w n t o b e
i n c o r r e c t b u t , s i n c e it d o e s fit t h e
results of tests c a r r i e d o u t o n t h r e e -
dimensional d e f e c t s , it has been
retained in the Draft British Standard
(Ref. 5). For n o n - p l a n a r d e f e c t s t h e
p r o c e d u r e a d o p t e d t o arrive at t h e
p e r m i s s i b l e d e f e c t size f o r e a c h o f t h e
q u a l i t y areas w a s t o p l o t results t a k e n
from the w o r l d w i d e literature for
d e f e c t s o f g r a d u a l l y i n c r e a s i n g size o n
t o t h e S-N d i a g r a m c o n t a i n i n g t h e
quality bands. The permissible defect
size f o r e a c h area w a s t a k e n as t h a t f o r
w h i c h o n e or m o r e results first fall i n t o
4 5
t h e area.
T h e q u a l i t y b a n d s are d e p i c t e d o n
t h e S-N d i a g r a m in Fig. 1 , a n d t h e
stresses c o r r e s p o n d i n g t o f a t i g u e l i f e at
IO3 a n d 2 x IO3 c y c l e s are g i v e n in
T a b l e 2. It s h o u l d b e m e n t i o n e d t h a t
t h e s e v a l u e s are as o r i g i n a l l y p r o p o s e d
by H a r r i s o n et al (Ref. 10). T o i n t r o -
duce further conservatism into accept-
a n c e levels f o r use i n t h e D r a f t B r i t i s h
S t a n d a r d , t h e stresses g i v e n in T a b l e 2
h a v e b e e n f a c t o r e d b y 0.88. H o w e v e r ,
in this paper reference w i l l be m a d e t o
t h e q u a l i t y b a n d s as o r i g i n a l l y p r o -
p o s e d , a n d it w i l l be s h o w n t h a t t h e s e
are c o n s e r v a t i v e i n t h e m s e l v e s . As c a n
be seen in Fig. 1, q u a l i t y V has n o
u p p e r l i m i t ; t h i s is w h y it is necessary
t o s t i p u l a t e t h a t n o d e f e c t s can b e
tolerated.
Three Dimensional Defects
T h e results o f f a t i g u e tests o n j o i n t s
c o n t a i n i n g p o r o s i t y c l a s s i f i e d t h e se-
v e r i t y o f s u c h d e f e c t s in t e r m s o f
p e r c e n t a g e b y v o l u m e , w h i c h can b e
taken to be approximately equal to
p e r c e n t a g e o n a cross s e c t i o n . T h e
p e r m i s s i b l e levels d e t a i l e d in T a b l e
3(a) w e r e t h e n d e r i v e d b y p l o t t i n g t h e
results as d e s c r i b e d t o d e t e r m i n e t h e
c r i t i c a l d e f e c t size for e a c h q u a l i t y
band.
A t t h i s p o i n t , it w i l l be a p p r e c i a t e d
t h a t p o r o s i t y is an i n n o c u o u s d e f e c t
f r o m the point of view of fatigue and
w i l l o n l y h a v e a s i g n i f i c a n t e f f e c t if
p r e s e n t in v e r y large a m o u n t s . H o w -
ever, t h e r e is a d a n g e r t h a t p o r o s i t y i n
s u c h h i g h levels c o u l d m a s k the
presence of other, more harmful,
d e f e c t s ; f o r t h i s r e a s o n , it w a s d e c i d e d
to l i m i t p o r o s i t y to l o w e r levels t o
m a k e i n s p e c t i o n p o s s i b l e . It w a s , in
f a c t , d e c i d e d t o w o r k o n a basis o f t h e
p e r c e n t a g e o f p r o j e c t e d area as s h o w n
on a radiograph, and the proposed
levels are g i v e n in T a b l e 3 ( b ) . T h e y are
considerably m o r e lenient than those
a l l o w e d in existing codes.
A s i m i l a r a p p r o a c h w a s a d o p t e d to
c o d i f y t h e results of tests o n j o i n t s
c o n t a i n i n g slag i n c l u s i o n s . For this
type of d e f e c t t h e severity parameter
selected was defect length. A l t h o u g h
the choice of such a parameter may
s e e m i r r a t i o n a l o n a t h e o r e t i c a l basis in
thai fracture mechanics analyses
Table 2—Values of Stress and Cycles for
the Quality Band System of Defect
Acceptance
Stress ( N / m m 3 ) at
Line dividing 10' 2 x IO"
quality bands
cycles cycles
v-w 309 142
w-x 208 96
X-Y 139 65
Y-Z 93 43
w o u l d specify the d e p t h t h r o u g h the
t h i c k n e s s as t h e r e l e v a n t s e v e r i t y c h a r -
a c t e r i z i n g p a r a m e t e r , t h e c h o i c e is
justified o n empirical grounds since
t h e d e p t h w i l l n o t v a r y w i d e l y in p r a c -
t i c e so t h a t t h e f u l l r a n g e of l i k e l y
d e p t h s w i l l h a v e b e e n c o v e r e d in t h e
tests.
C a t e g o r i z a t i o n o f f a t i g u e test results
o n t h i s basis has p r o v e d s a t i s f a c t o r y
(Ref. 8) a n d d e f e c t l e n g t h is t h e m o s t
r e a d i l y m e a s u r e d d i m e n s i o n . Results
f r o m t h e literature i n d i c a t e that fa-
t i g u e s t r e n g t h falls q u i t e r a p i d l y w i t h
i n c r e a s i n g slag l e n g t h , b u t o n c e t h e
d e f e c t e x c e e d s 10—20 m m t h e r e is l i t t l e
c h a n g e . It has also b e e n f o u n d t h a t
stress r e l i e f improves the fatigue
strength of slag-bearing w e l d s , mainly
b e c a u s e it r e m o v e s h y d r o g e n w h i c h is
deleterious to strength, and therefore
a s - w e l d e d a n d stress-relieved joints
may be separately categorized. The
p r o p o s e d a c c e p t a n c e levels f o r slag
i n c l u s i o n s are d e t a i l e d i n T a b l e 3 ( c ) .
As i n d i c a t e d p r e v i o u s l y , a h i g h
p r o p o r t i o n o f all w e l d d e f e c t s d e -
t e c t e d are n o t cracks (Ref. 1). T h u s ,
relaxation of acceptance requirements
for three-dimensional defects should
yield considerable e c o n o m i c returns.
The safety of the quality levels
p r o p o s e d f o r t h e s e d e f e c t s t y p e s has
b e e n p a r t i a l l y c o n f i r m e d b y Soete a n d
Sys (Ref. 11) i n tests o n large m o d e l
p r e s s u r e vessels a n d r e c e n t l y r e c e i v e d
f u r t h e r c o n f i r m a t i o n (Ref. 12).
Planar Defects
A n a p p r o a c h similar to that for
t h r e e - d i m e n s i o n a l flaws cannot be
u s e d t o c a t e g o r i z e p l a n a r f l a w s . Planar
defects can i n c l u d e the f o l l o w i n g :
c r a c k s , lack o f p e n e t r a t i o n , lack of
f u s i o n , lack of i n t e r r u n f u s i o n , lack o f
r o o t f u s i o n , lack o f s i d e w a l l f u s i o n a n d
o x i d e i n c l u s i o n s in a l u m i n u m w e l d s .
U n d e r c u t a n d r o o t c o n c a v i t y c a n also
be t r e a t e d as d e f e c t s b r e a k i n g t h e
surface, since they w i l l certainly c o n -
t a i n s m a l l c r a c k - l i k e d e f e c t s at t h e i r
roots.
Table 3—Permissible Levels of Defects for the Quality Band System of Defect Acceptance
Quality
(a) Porosity as proposed by Harrison
et al (Ref. 10)
Permissible % of porosity by volume
(b) Porosity as proposed in the Draft
Standard (Ref. 5)
Permissible % of projected area on
radiograph
(c) Slag inclusions as proposed by
Harrison et al (Ref. 10) and
the Draft Standard (Ref. 5)
Inclusion length, m m , (Max) as-welded
Inclusion length, m m , (Max) stress-
relieved
WELDING
In t h e past, it has b e e n c o n v e n t i o n a l
to reject, w i t h a f e w exceptions, welds
c o n t a i n i n g p l a n a r d e f e c t s a n d s u c h an
approach appears reasonable w h e n
s u c h d e f e c t s are k n o w n t o b e t h e m o s t
d e l e t e r i o u s . H o w e v e r , it is k n o w n t h a t
v i r t u a l l y all w e l d e d j o i n t s contain
p l a n a r d e f e c t s of o n e k i n d or a n o t h e r ,
but m o s t are s m a l l e n o u g h to e s c a p e
d e t e c t i o n . Structures c o n t a i n i n g such
defects have given q u i t e satisfactory
service.
W i t h i m p r o v e m e n t s in N D T t e c h -
n i q u e s a n d t h e i n c r e a s i n g use of
destructive examination of sample
w e l d s , it b e c o m e s clear t h a t a s t a n d a r d
w h i c h seeks t o r e j e c t all p l a n a r f l a w s is
no longer practical. A decision must,
t h e r e f o r e , b e t a k e n as t o w h e t h e r t h e
d e f e c t is large e n o u g h t o i m p a i r t h e
s e r v i c e a b i l i t y of t h e s t r u c t u r e . T h e
severity of planar flaws d e p e n d s not
o n l y o n size b u t also o n s h a p e a n d
l o c a t i o n , all o f w h i c h m a y vary f r o m
o n e w e l d t o t h e n e x t . S i n c e it w o u l d
o b v i o u s l y b e i m p o s s i b l e t o test all
c o m b i n a t i o n s o f j o i n t t y p e , d e f e c t size
a n d p o s i t i o n t o o b t a i n n o r m a l S-N
data, fracture mechanics analysis
t e c h n i q u e s must be used to q u a n t i f y
t h e e f f e c t s of p l a n a r d e f e c t s .
H a r r i s o n (Ref. 9) s h o w e d h o w it w a s
p o s s i b l e to a n a l y z e t h e results of
f a t i g u e tests o n b u t t w e l d e d s p e c -
imens containing t w o dimensional
p l a n a r f l a w s by i n t e g r a t i n g t h e f a c t u r e
m e c h a n i c s b a s e d f a t i g u e crack p r o p a -
g a t i o n l a w (see T a b l e 1 f o r n o m e n c l a -
ture):
d a / d N = C (AK)"' (I)
H a r r i s o n t o o k t h e v a l u e o f m as 4
a n d f o u n d that this f i t t e d t h e results. In
t h e D r a f t British S t a n d a r d (Ref. 5), a
simplified version of Harrison's m e t h -
o d o f analysis is g i v e n f o r g e n e r a l use.
T h i s a p p r o a c h has b e e n published
p r e v i o u s l y (Ref. 13) a n d is w e l l t r i e d
a n d p r o v e n against l a b o r a t o r y d a t a f o r
f a t i g u e tests o n w e l d s containing
p l a n a r d e f e c t s . T h e m e t h o d is b a s e d
u p o n t h e use of e q u a t i o n ( I ) , w i t h a
fixed value of m = 4 and the value of
W
20 20
1.5 10 No No
maximum maximum
6.0 No No No
maximum maximum maximum
R E S E A R C H S U P P L E M E N T 115-s

Thickness, mm
25 25
2 3 4 5
1 1 1 1
A 7-
o
r\j
:
i*
4 f
S 007
:
0 007 till
0-1 2 3 4 5
TO
Thickness
Fig. 2-Allowable sizes of two dimensional defect plotted
against thicknesses for the various fatigue qua lity levels
the constant C chosen so that the
analysis gives conservative results
against all known laboratory test
results.
The crack propagation equation is
integrated, and the results are pre-
sented in graphical form. This is done
such that, for each of the quality levels
V-Z, the maximum allowable flaw size
can be determined as a function of
thickness. This presentation is shown
in Fig. 2. The curves apply to the total
depths of long buried flaws, but
methods are given to assess and re-
categorize planar defects if they occur
at the surface and it they occur as
flaws in the stress concentration
region of a fillet w e l d , i.e., the weld
toe.
It flaw length is less than 10X the
depth, then the depth may be re-
categorized to take account of the less
severe defect aspect ratio by allowing
an increase in depth. This recategori-
zation is shown in Fig. 3. The fatigue
life is calculated to the stage where the
defect becomes a through thickness
crack (for pressure vessel applications,
this is the "leak before break" situa-
tion). If the fracture toughness and
stress conditions are such that final
fracture w o u l d occur from a fatigue
crack smaller than through thickness,
this can be allowed for by calculating
the life from this crack size to the full-
thickness, and then deducting this
number of cycles from the total life to
the lull thickness crack. This correc-
tion is usually small.
Analyses are also given for flaws at
fillet weld toes for different ratios of
fillet size to plate thickness. The small
values of allowable flaw size so calcu-
lated are indicative of the very large
effect w h i c h toe cracks or undercut
16-s I JANUARY 1977
250 10 T
2 3 45
^, —^ i i it 5
4
- h
L -C:
3 u z
S>
a _ -fa Defect height
CD
2 -c:
e
-*—
o
~~b~
!
2 54
s
o
CU
-a
CJ 01
•
N
a
X Y&
13
0-25 o
!
001 =
5 :
w 4
3 j
0-0012
in 0 025
2 3 45
to
fig. 3—Reduction in defect height permitted tor "short" delects
may have o n fatigue life.
Recategorization of planar defects
in terms of size may also be necessary
if adjacent flaws w i l l interact. Harrison
(Ref. 9) has shown that interaction
cannot occur if the length of sound
metal between the ends of t w o adja-
cent defects is greater than: (a) 2.25
times the thickness of the material and
(b) 1.25 times the length of the larger
defect.
It is, therefore, stipulated that pro-
vided these limitations apply, each
defect may be treated separately. If,
however, either of these limitations
are exceeded, the Draft specifies that
the defects should be recategorized in
size by a rectangle which includes the
interacting defects. This rectangle
should be drawn such that its sides are
parallel and at right angles to the plate
surfaces. This rule also applies to adja-
cent, discontinuous slag inclusions.
Defect Acceptance U n d e r Low
Cycle Fatigue Conditions a n d in
Aggressive Environments
The majority of the test data from
which the rules were developed were
obtained under high cycle (>10 3 )
fatigue conditions. Bearing in mind
that the Standard had to cater for
defects in all types of structures,
including pressure vessels, it was
necessary to demonstrate that the
methods of deriving acceptable defect
levels w o u l d cover weld defects oper-
ating under low cycle fatigue condi-
tions. This held particularly for defects
situated in regions of high stress
concentration such as nozzle joints in
pressure vessels. There was also
virtually no published data dealing
1 ""IT M M
a .•0-D1
it/
0-K
n?
\ \ys// //
Defect length
M*/ / / / ;
WT
7\/\/
A/y y
\
/ T
/A\y
1 11 > I I -L i i i l l l l
3 4
5 007 2 3 4 5 01 2 3 4 5 10
Defect height
Actual ratio
Thickness
w i t h the effects of other than ambient
temperature air environments on the
fatigue behavior of joints containing
weld defects; consequently, no guid-
ance could be given as to h o w to
assess the behavior of defects in other
environments.
One of the most important func-
tions of the research worker in this
area is to supply the required data
obtained from small-scale laboratory
tests. Another function is to generate
confidence by demonstrating that the
information can be safely applied to
real structures. W i t h these objectives
in mind, The Welding Institute has
carried out a major project (Ref. 14),
sponsored by a number of industrial
concerns, to provide the data and
analyses necessary to assess the behav-
ior of w e l d defects under low cycle
and aggressive environmental condi-
tions. This project was outlined by
Archer (Ref. 12), prior to its comple-
tion, but is n o w described below w i t h
test results and analyses relevant to the
derivation of acceptance levels for
defects operating under the aforesaid
conditions. Test results are given in SI
units, but a list of conversions to
customary U.S. units is included in
Table 4.
Low Cycle Fatigue
The problem of assessing weld
defects under low cycle fatigue condi-
tions was tackled using t w o methods.
Butt welded specimens containing
actual buried defects, (porosity, slag
and lack of penetration) were fabri-
cated from a mild steel to BS 1501-161
and a low alloy steel to BS 1501-271.
These materials were chosen to cover
the range nf steels in current use for
 330
T a b l e 4—List of C o n v e r s ons f r o m Sl t o U.S. C u s t o m a r y U n ts

Unit SI U.S. c u s t o m a r y Conversion

Length mm inch 25.4 m m = 1 i n c h

Stress N/mm'3 Ksi 6 . 8 9 N / m m - = 1 Ksi
Stress I n t e n s i t y N m m '•"- Ksi v i n c n 34.7N m m - 1 " = 1 K s i V i n c h

Fig. 4—Design oi specimens with defective

pressure vessel applications. system proposed by Harrison vt al
butt welds to assess low cycle fatigue
The specimen design used is shown (Ref. 10) can be extrapolated to endur- behavior. Dimensions in mm
in Fig. 4, and specimens were tested
under predominantly high strain con- I I I I i I M I M i I I I I i
ditions. It is worth mentioning that • Severe porosity - approximately J'/. -
current pressure vessel code require- by volume
ments permit strain ranges of up to A Low porosity - IV. of projected area
per 25 mm of weld
twice yield to occur in nozzle welds; thickness
to simulate this situation, the strain
range across the defective weld was ^^~7^^__^
:

^ III!'
varied up to twice yield. As a fatigue
crack grows under strain controlled
conditions, the applied load required
to maintain the strain range w i l l fall. As
this is the case, the failure criterion
i i i

:
— — ~ _ ! 1 ~ ~ ~~~—~~2~~—•—i
adopted for these specimens was
J

either the development of a through- z~"~—~i

i i i i
thickness crack, or a 60% fall off in 2 3 i 5 n5 2
I I

3 i 5
I 1 l l l l

upper limit load, whichever occurred Cycles

first.
Fig. 5—Results of tests on specimens containing porosity defects
For tests on specimens containing
non-planar defects, t w o levels of
defect severity were produced: llll llll i i i 11 11 rt
:
1. For porosity, 1 % of the projected • r
est r 'suits
area per 25 mm of weld thickness Alt specimens were
based on radiographic measurements stress relieved
(this being the maximum level allowed
by the ASME c o d e - R e t . 15) and 3% by
volume.
2. For slag, a continuous slag line
: K
i
and 3 slag inclusions approximately 6 ""^•i*" ^-—^
mm in length spaced 30 mm apart so — • • • « « ! _ " * — ~ - ^ ~ _ ^ ~ ~ ~»^~^_
that no defect interaction w o u l d take z
place.
E ^•«-A» :
The results obtained were plotted
on an S-N diagram on which the
quality bands had been linearly extrap- --—.
olated into the low cycle regime. The
S-N diagram utilized differed from that 10 llll III
3 4 5
In,, ^
4 5 3 i S 2 3 4 5
originally proposed in that the stress IO2 10' io- 10* 10"
range was replaced by the total strain Cycles

range, calculated as: Hg- 6—Results of tests on welded specimens with continuous slag inclusions

Ay = A a / r (2)
1 i M llll l l l l i i I I 11 —I
where Ay = strain, A a stress and : -lilt
• r
E = Young's modulus.
A* Denotes failure from
The results so obtained are shown in unintentional defect,
Fig. 5 for specimens containing porosi- or specimen unbroken
All specimens were stress relieved
ty and in Figs. 6 and 7 for specimens
V
containing continuous and intermit- ^ ;
tent slag respectively. As can be seen, Af
v ^ — - •*
all the results fell in quality bands
higher than those proposed (see
I"* £^_ ^ ^ « _ " A A"
^-~~L ~~~~^~^^. ^~~~~~^AA~~~~~~~~^.
j
Tables 1 and 2) for the severity of Z ^~~-^ ^~~-~~^ ^"-"—^ k
A
defect. The net result of this work was
* z
to demonstrate that the quality band

Fig. 7-Results of tests on welded

specimens containing 3 slag Mil 1 1 ""TnT
-,! 2 3 <t 5
defects (approximate
length 6 mm) Cycles

W E L D I N G RESEARCH S U P P L E M E N T I 17-s
 n i l M M 1 1 1 1 1 I I I I i M i
Qualit y I
2 bands
70- • Low a llOY steel tested
_. £_ at ambient temperature :
x Type 316 stainless steel
s tested at 450°C
•^x '•^-^J"~~—
-*• Denotes specimen
3 I /~~~~~^-^ "~^~~-^-Jsr--~ unbroken
2
^-—~^_ ^^»x .. •
S •3
10
1 X
AA" ""~--^_X~*"
5
4

4 I I I I l l l l 1 II 1 I I I I
10 jnz 2
^ j3 t4 5
o *p.j *2 j3 4
** 5u If)** 2
z 3 *4 J5
J
1H-' *2 J3 t 4 u5 ^ 0 <:2 o3 ^ 4o 5 .p -,7
/

Cycles to failure
Fig. 8—Results of tests on welded specimens containing lack of penetration defects tested at ambient and elevated
temperatures

ances as low as 500 cycles if strain

4> range replaces stress range on the S-N

diagram.
Similar specimens containing cen-
tral buried lack of penetration defects
were also tested under low cycle
fatigue conditions. In this case the
^ < defects were approximately 25 mm
755 long by 8 mm deep. The results
12 5 rod 38 obtained are shown in Fig. 8 where it
can be seen that they fell within

A)- quality band W. The fracture me-

chanics based analysis given in Figs. 2
and 3 w o u l d deem these defects
— 5t-» 343 acceptable to quality Z. Thus, once
again the proposed acceptance stan-
dard was shown to be conservative. It
is worth noting that none of the
present fabrication codes w o u l d deem
these defects acceptable.
Harrison (Ref. 9) showed that it was
possible to predict the behavior of
butt joints containing planar defects
by fracture mechanics analysis
techniques. His analyses referred to
defects operating under high cycle
conditions. To carry out similar anal-
yses of defective joints operating in
the low cycle regime, crack propaga-
tion studies were carried out at very
high stress intensity factor ranges
typical of those which w o u l d be
present at defective nozzle welds.
The type of specimen utilized in
these tests is shown in Fig. 9. It was
f o u n d that fatigue crack growth rates
tended to be lower in the weld regions
than the base metal for the two mate-
rials tested, the growth rate in the HAZ
being the lowest. For this reason it is
suggested that fracture mechanics
assessment of crack growth from

Fig. 9—Design ol crack propagation spec-

imen to measure growth rates at high stress
intensity levels in: a) base metal b) weld
metal; c) HAZ

18-s I J A N U A R Y 1977

Forged cylindrical
— Detective
Fig. 10-Design of spherical test vessel to
study, a) slag; b) porosity; c) lack of pene-
tration defects in stress concentration
regions
planar w e l d defects by fatigue is
carried out using growth data for the
base metal. The scatterband of crack
propagation data, which except at very
high values of AK was relevant to
propagation under plane strain condi-
tions, lay on a linear extrapolation of
data obtained at lower values of AK by
Maddox (Ref. 16) and was essentially
linear up to very high AK values (6000
Nmm-W).
To generate confidence in the use of
the quality band system and small
scale tests to determine acceptable
defect levels, particularly under low
cycle fatigue conditions, tests were
also carried out on model pressure
vessels; these were fabricated from the
same mild and low alloy steels as were
used in the small scale tests.
The vessel design used for studying
the behavior of w e l d defects in stress
concentration regions is shown in Fig.
10. Defects (slag, porosity and lack of
penetration) were situated in the
nozzle welds. The opening was rein-
forced to current code requirements
w i t h the reinforcement in the nozzle
for experimental reasons. A cylinder/
sphere intersection was chosen since it
was more amenable to a theoretical
crack propagation analysis than a
cylinder/cylinder intersection.
The weld design selected involved
only a very small external fillet, about
3 m m . For this design a finite element
analysis of the defect free joint gave a
peak stress concentration factor (SCF)
of about 2.1 in the radial direction
(perpendicular to the weld) and about
1.9 in the meridional direction. In both
cases these were in the weld area. By
increasing the fillet size to a more
realistic 12 m m , these SCFs could both
have been reduced to 1.8. However,
the small fillet size was used in the
tests in an attempt to ensure that
failure w o u l d occur in the defective
nozzle welds.
The vessels were subjected to a
hydrostatic test before cyclic pressure
loading, and the hoop stresses for the
static and cyclic tests were the
IUUU M M
'-
nozzle
weld
- Membrane stress
3 Quality bands
«. X
2 Membrane stress Mud
- Spherical
test vessel
100 ' —^ — ^
:
-^^
5
3 -
2
- * Denotes no failure
from defect
I i i i i i i
m4 2 3 4 5 m5 2 3
Endurance, N, cycles
Fig. 11-Results of tests on nozzle welds containing slag defects compared with
acceptance proposals: VV—as welded, 1.5 mm max. stress relieved, 6 mm max.; X-as
welded, 10 mm max. stress relieved, no maximum; Y—no maximum; Z—no
maximum
maximum w h i c h BS 1515 (Ref. 17)
w o u l d allow for the materials, the
cyclic stresses being 163 and 250
N / m m 3 for the mild and low alloy
steels respectively. Bearing in mind
that intermittent slag had been shown
to be a virtually innocuous defect in
the small scale tests, only continuous
slag defects were introduced into the
nozzle welds. These slag lines ex-
tended around the complete periphery
of the weld. Neither of the nozzle
welds failed from the slag-line defects.
In the mild steel vessel the test was
stopped after more than 100,000 cycles
at w h i c h stage no crack growth from
the defect was evident. In the low
alloy steel vessel, failure occurred from
one of the main seams in the vessel
after 19,000 cycles. This seam weld
failure was associated w i t h joint mis-
alignment.
The t w o results are plotted in Fig. 11
together with the proposed quality
bands. Although these results have
been plotted in terms of vessel
membrane stress, no account being
taken of the localized stress con-
centration due to the nozzle geome-
try, it may be seen that the proposed
acceptance levels are still conserva-
tive.
Two vessels containing severe po-
rosity were tested, one in each of the
materials. The defective nozzle weld
in the low alloy steel vessel failed after
only 1,130 cycles, the nozzle being
completely b l o w n off the vessel. After
failure, the porosity level was found to
be 9.4% by volume. The main reason
for the early failure, however, was the
gross lack of fusion also present which
covered about 50% of the fracture
face; this was presumably due to the
extreme measures taken to produce
the porosity defects. In view of the
gross porosity and lack of fusion, the
W E L D I N G RESEARCH S U P P L E M E N T 119-s
III I I I I
_
Low alloy steel
-
x ^~~~^~
r -
****—-
I I I I i i , i
4 5 JQ6 2 3 4 5 ;Q7
failure was probably due to over-
loading rather than fatigue.
This result did highlight the dangers,
mentioned earlier, of gross porosity
masking more harmful planar defects.
The defective nozzle weld in the mild
steel vessel failed after 24,000 cycles.
After failure it was found that the
porosity level was 7% by volume. This
result demonstrated that the proposed
acceptance levels were safe providing
that account was taken of the increase
in stress due to the stress concentra-
tion at the nozzle. These results indi-
cate that levels of porosity as high as
7—10% by volume are dangerous in
stress concentration regions because
of the resultant over-stressing of the
remaining sound weld metal. H o w -
ever, there is no doubt that the
extremely low allowable porosity lev-
els in existing codes are unnecessarily
conservative and could safely be raised
to the levels suggested in Table 2(b),
where the maximum allowable level is
5%. Such a level is extremely high and
is never likely to occur in practice.
Three vessels were tested, each
containing a continuous buried lack of
penetration defect around the periph-
ery of the nozzle weld. O n e of these
vessels was in the low alloy steel and
contained lack of penetration n o m i -
nally 3 mm in depth. This vessel failed
from the main seam weld after 21,000
cycles, the defective weld being
uncracked.
The remaining t w o vessels were
fabricated from the mild steel material
and contained lack of penetration
defects 7 and 8 mm in depth respec-
tively. The vessel containing the 7 mm
defect failed after 53,000 cycles, but
the failure had initiated from the w e l d
toe and had joined up w i t h the planar
defect. A second fatigue crack had
then propagated from the planar
 1000 I I I
: Table 5—Value of Constant in Equations
Membrane stress 0 36r m
- ?
Membrane stress
1501 • 1SI
1 -
For crack to propagate to outside surface
For leakage
I I Allowable crack sizes from draft standard (5) based on nominal membrane
I 1 stresses and endurance of W'cycles
m -7~~ I Results from tests on defective vessels with lack of penetration
? lmm in nozzle welds
A 8mm )
w l l l l l l l l
103 2 3 4 5
Endurance. N
12—Comparison ol results for lack of penetration type defects
defect to form a through-thickness
crack. The remaining mild steel vessel,
containing an 8 mm planar defect, was
taken off test at 94,000 cycles w i t h no
crack growth associated w i t h the
defect having occurred, although
cracking from the weld toe had again
occurred.
A theoretical crack propagation
analysis was carried out for these
defective nozzles making use of an
existing finite element analysis pro-
gram. The crack propagation law
assumed in the program, based on
experiments (Ref. 16) was:
d a / d N = (1.035/1013) (AK) 3 " 3
when d a / d N is expressed in m m / c y c l e
and AK is in Nmnrr 3 '' 3 .
The results obtained from this analy-
sis, together with the actual test
results, are shown in Fig. 12. It can be
seen that there is good correlation
between the analytical and experi-
mental results; this indicates that frac-
ture mechanics can be used to predict
the behavior of planar defects w i t h a
reasonable degree of accuracy, even in
complex geometrical regions.
Also plotted in Fig. 12 are allowable
crack sizes derived from the Draft
Standard (Ref. 5). Although these crack
sizes were determined using the n o m -
inal membrane stress, (no account
being taken of the SCF due to the
nozzle geometry) it can be seen that
the Draft proposals are safe for this
nozzle geometry.
Fatigue in Aggressive Environments
As mentioned earlier, it was felt that
a useful contribution to knowledge
could be made by examining the rele-
vance of the Draft Proposals (Ref. 5) to
quantify defects operating in other
than ambient temperature air environ-
20-s I I A N U A R Y 1977
I I I I i '
-*• Denotes no failure
from defect
Bmm rfoo„ _ equation describing S-N
7k ~ ~ . ^ ^ ^
^-^^- mm deep crack
\14mm\
—• - A^ ^
I Results from finite
[ element analysis
I II
W4 2 3 4 5 1Q5 2 10s
ments. Tests were, therefore, made on
both types of specimen shown in Figs.
4 and 9 in an elevated temperature air
environment. There is a considerable
amount of published literature to
show that elevated temperatures can
profoundly alter fatigue crack propa-
gation behavior from that at ambient.
In addition to the mild and low alloy
steels used in the low cycle fatigue
program, a 2'/iCr-1Mo steel to BS 1501-
622 Grade 31 and an austenitic stain-
less steel to AISI Type 316 were also
used. The steels were tested at the
f o l l o w i n g temperatures:
(3) 1. BS 1501-161-350 C (662F)
2. BS 1501-271-350 C (662 F)
3. BS 1501-622-400 C (752 F)
4. AISI Type 316-450 C (842 F)
These temperatures were chosen to
be outside the designated creep range
of the materials to avoid complications
due to crack growth by creep-assisted
processes. In fact, it was found that
rates of crack propagation at elevated
temperature were not substantially
different from those at room tempera-
ture for the mild and low alloy steels
where comparison was possible.
AISI Type 316 was the only steel
tested w h i c h showed an indication
that the elevated temperature environ-
ment had an adverse effect upon crack
growth rate. As this was the case tests
on specimens containing real defects
were confined to one series of spec-
imens containing lack of penetration
defects in butt welds in AISI Type 316
at 450 C (842 F). To be consistent w i t h
the low cycle fatigue program, these
specimens also contained central
buried defects 25 mm in length by 8
mm in depth, and were tested under
both low and high cycle fatigue condi-
tions.
The results obtained are shown in
Fig. 8, plotted on the basis of total
' i I I I l_
of Quality Bands
Value of constant D, in
Lines between data, ( A a ) ' N = D. D
quality bands is given in N' mm-»
v-w 9.12 X 10"
w-x 1.87 X 10'-1
X-Y 3.73 X 10"
Y-Z 7.39 X 10,:<
strain range as defined in equation (2)
against cycles to failure. As can be
seen the results all fell w i t h i n Quality
Band W. As for the similar specimens
tested at ambient temperature, the
Draft Proposals (Ref. 5) w o u l d deem
such defects acceptable to Quality Z.
This work has been reported by the
present author (Ref. 18), who also
showed how it was possible to predict
accurately the fatigue lives of such
joints by taking account of the change
in shape as the crack grew through the
plate thickness. However, such an
analysis is considered to be too
complex for everyday use, and the use
of the simple graphical presentation
given in the Draft Proposals (Figs. 2
and 3) will yield a safe estimate of
required quality.
It is w o r t h comparing the results
shown in Fig. 8. These results were
obtained from similar specimens c o n -
taining similar planar defects. In fact,
the only differences between these
test results were the specimen material
and testing environment. The results
indicate that both joints w o u l d be
acceptable to quality W. This can be
explained by fracture mechanics
techniques.
Comparison of the measured rates
of crack propagation in both materials
and environments revealed that they
were markedly similar. This leads to
the important conclusion that the
proposed methods of defect accept-
ance (Ref. 5) can be used for defects in
other than ambient air environments,
providing that it can be shown that the
fatigue crack propagation behavior is
similar to, or better than, the crack
propagation behavior used to draw
up the acceptance proposals. If the
aggressive environment produces
crack propagation rates w h i c h are
more deleterious to fatigue strength
than that used in the Draft proposals,
the f o l l o w i n g method of deriving
acceptance levels applicable to the
environment in question is sug-
gested.
Derivation of Defect Acceptance Levels for
Specific Aggressive Environments
Harrison originally defined the qual-
ity bands on an S-N plot from a frac-
ture mechanics analysis of results of
tests on joints containing planar
 f i l l to ' / • /1
» / j K/c ffor purposes of example. t 1
1
kic assumed to be 1S00Nmm~3'2
- Ij I
-
1 1
p l'l 1
1 I - //
1 3
Hypothetical
propagation
crack
behaviour
1
1
11 • / ' /

for aggressive —»j f '/ /

environment
m $ 4 and
where
I
'
j
1 w /
C ±172 xW'15 ' /
1
-?

-
10 ! ' /
/ . From Harrison
' / / -
ref 9

!1 -
- / / C =1-72x10 " "

ill // // environment)
1 /
NPW rtirvp tn nrrnunt for ——+4 \ 1

"/ / accelerated growth under

aggressive environment
j
/
j
' /
L F/_Qm
_ .
Harnson ref 9

3
/ / obtained by a lateral
shtft of Harrison's data I
1
'
I f
j
Q _ pjg x JO~J5
(Ambient air
/ 2
from ref.9. For this curve I 1 environment)
m = 4 and j 1
I I j 1\
// da
i U
dN 10

_ // m 1—^A i r
i~ - propagation
Hypothetical crack
increasir 9/ m behaviour
I C
j decreasing I 1 for aggressive environment ~

1
1 - i
-
i / C
decreasing
i //
// * 1 //
i

1 4K
i 1
II
2
1 1
1 I

200 300 WO 500

ill, JQQQ
Range ot stress intensity ,/lK,
Fig. 13-Diagrammatic representation of hypothetical crack propa-
gation behavior in ambient air and aggressive environments
defects (Ref. 9). He used equation (1)
to describe the rate of propagation of
fatigue cracks and assumed a slope of
m = 4. From other research (Ref. 19) it
is known that the slope of the straight
line, n, representing fatigue data on
log-log S-N plot is related to the slope
of the crack propagation curve by the
relationship:
m = -1/n (4)
Thus the slope of the quality bands
is defined as — VA. The equation of an
S-N curve is given as:
A a "' N D (5)
where D is a constant.
By substituting for m and A a in
equation (5), the value of the constant
D may be determined for the lines
dividing the quality bands in Fig. 1.
These values are given in Table 5.
As has been stated, environmental
effects may increase crack propagation
rates such that the quality band system
based on ambient air test results
becomes unsafe. Consider, for exam-
ple, the crack propagation behavior
depicted diagrammatically in Fig. 13
whereby the environment causes
crack growth to occur at a higher rate
2000 3000 4000 5000
-4
/,' , / : . . ,
300 400 SOO 2000 3000 4000 5000
;
in Nmm' ?
Range of stress intensity, ZlK, in Hmm'3'2
Fig. 14—Diagrammatic representation at lateral shift of ambient air
crack growth curve to account for aggressive environmental
effects
than that analyzed by Harrison. A the fracture toughness, K l r , of the
revised system of quality levels can be material in the environment. This is
derived by operating on the crack depicted diagrammatically in Fig. 14.
propagation law. This hypothetical crack propagation
In equation (1) describing fatigue equation will give an extremely con-
crack propagation, the exponent m servative estimate of crack growth.
describes the slope of the curve and However, it is felt that even grossly
the constant C the position of data conservative defect acceptance stan-
relative to the axes. This is depicted on dards are better than no standards at
the inset to Fig. 13. W h i l e it w o u l d be all. By always maintaining the slope
possible to derive new quality bands in = 4, the quality bands can be
based on a redefinition of m and C, adjusted to account for environment
this w o u l d lead to extremely tedious by simply moving them d o w n the
mathematics; thus a simple solution is abscissa on the S-N plot. This is
proposed whereby the effect of slope achieved by the following simple anal-
m is ignored and a hypothetical crack ysis.
propagation equation is defined which For most welded joints undergoing
has the same slope m = 4 but a new fatigue, the life can be predicted by
constant CE. By this means the crack combining a solution for stress inten-
propagation curve is shifted laterally sity at the crack tip:
across the d a / d N against AK plot to
accommodate the environment accel- AK = Y A a V^a~ (6)
erated crack growth.
w i t h equation (1) describing crack
The position of the hypothetical growth. Combining and rearranging
data w i t h respect to the crack propa- these equations yields an equation
gation data obtained under the aggres- w h i c h may be integrated to obtain a
sive environmental conditions can be prediction of fatigue life:
fixed by stipulating that the curves
intersect at a value of AK equivalent to IZ{ d a / ( Y V l ? a ) m = C ( A a ) ' " - N (7)

W E L D I N G RESEARCH S U P P L E M E N T I 21-s
 It should be noted that for a given
set of joints or defects a,, a, and Y are
constant and so the integral becomes a
constant. Equation (7) may be re-
written as
I = C(Aff p - N (8)
where
I = J"1]1 d a / ( Y V i a ) ™
From equation (5), D = ( A a p N,
and equation (8) may be further re-
written as:
I = C D (9)
The values of D for the quality bands
are given in Table 4, and the value of C
was given by Harrison (Ref. 9) as
C = 1.72 x 10- 15 , assuming that crack
length was measured in mm and
stresses in N/mm 3 .
To assess defects in aggressive envi-
ronments in terms of quality, the rele-
vant quality for the defect in ambient'
air environment must first be deter-
mined by the use of Table 3 and Figs. 2
and 3, depending upon defect type.
Second, a crack propagation test must
be carried out in the relevant aggres-
sive environment to determine d a / d N
against AK behavior. The crack propa-
gation data must then be adjusted to
obtain an equation w i t h a slope of 4 to
obtain a conservative estimate of crack
growth in the environment, as shown
in Fig. 14 and the new value of the
constant CE determined.
By maintaining the slope of m = 4,
the value of the integral I in equation
(8) is constant regardless of environ-
ment. Thus, for ambient air from equa-
tion (9) and in the aggressive environ-
ment:
1 = C„ D B (10)
and combining equations (9) and
(10):
C D = C, D„ (11)
Values of C, D and CK are known
and so equation (11) can be solved for
D,,. Values of D E so derived may be
entered in equation (5), w h i c h can be
solved in terms of A a and N to obtain
revised quality bands. These quality
bands will simply be downgraded in
terms of allowable stress.
By nature of the fracture mechanics
analysis, itself, whereby only the rate
of crack propagation has been altered,
the limits of acceptable defect levels
(as originally defined) will still be
applicable to the new quality bands.
It must be stressed that the quality
band system, as originally proposed, is
conservative in itself. In many cases it
will adequately account for defective
joints operating in other than ambient
air environments. The methods of
adjusting the quality bands, as de-
scribed previously, should only be
utilized after it has been demonstrated
that the original proposals cannot
22-s | JANUARY 1977
yield acceptable defect sizes, because
of the increase in crack growth rate
caused by the environment in ques-
tion.
Conclusion
This paper has attempted to show
how rationally based acceptance stan-
dards for w e l d defects operating under
fatigue loading conditions have been
derived, based on information gath-
ered together from research carried
out worldwide. Based on this informa-
tion, a Draft British Standard on the
derivation of acceptance levels of
A/eld defects has now been written.
The validity of the concept of defect
acceptance based o n fitness for pur-
pose has been confirmed in a recent
investigation of defective joints oper-
ating in the low cycle regime and at
elevated temperatures. Methods of
adjusting the quality band system to
account for defects operating in
aggressive environments have been
proposed.
O n e point should be made clear to
those w h o regard the defect levels
given in the acceptance proposals and
in this paper as being totally unaccept-
able because they seem to indicate a
reduction in quality. Welding costs
will escalate as NDT of unsophisti-
cated structures becomes more w i d e -
spread because more and more unne-
cessary repairs will be made. In oper-
ating a business for profit, fitness for
purpose is surely a more logical axiom
than quality regardless of cost.
Author's N o t e
Except where direct mention is
made to published research or code
requirements, the views expressed in
this paper are those of the author
alone.
Ac/cnow/edgments
The author is grateful for the help
and advice of his colleagues in the
preparation of this paper—particularly
to Drs. J.D. Harrison and T. R. Gurney
for useful discussions.
References
1. Salter, C. R., and Gethin, |. W., "An
Analysis of Defects in Pressure Vessel Main
Seams," Conference on Pressure Vessel
Standards-The Impact of Change, The
Welding Institute, The Institution of Me-
chanical Engineers, The British Standards
Institute, London, November 1972.
2. Lundin, S., "Some examples of the
consequences of common defects in
welds," Proc. 2nd Conf. on Significance of
Defects, London, May 1968, The Welding
Institute.
3. Dawes, M. C , "Fatigue strength of
ferritic steel shafts reclaimed by welding
and metal spraying," Brit. Weld. /., Vol. 10,
No. 8 August, 1963, pp. 418-427.
4. Wellinger, E.H.K., and Sturm, D., "In-
fluence of weld seam quality on the fatigue
resistance under pulsating internal pressure
in the case of vessels with nozzles," IIW
Document XIII-578-70.
5. "Draft Standard Rules for the Deriva-
tion of Acceptance Levels for Defects in
Fusion Welded Joints," British Standards
Institution, Draft for Public Comment,
Document 75/77081 DC, February 1976.
6. Burdekin, F. M., "Comments on the
Draft British Standard for the derivation of
acceptance levels for weld defects," Paper
7—Conference on Quality control and Non-
Destructive Testing in We/d/ng,"The Weld-
ing Institute, The Institute of Quality
Assurance, The Non-Destructive Testing
Society of Great Britain, London, November
1974.
7. Harrison, J. D., "The basis for an accep-
tance standard for weld defects. Part 1:
Porosity," Metal Construction, Vol. 4, No. 3,
March 1972.
8. Harrison, J. D., "The basis for an ac-
ceptance standard for weld defects. Part 2:
Slag Inclusions," Metal Construction, Vol. 4,
No. 7, July 1972.
9. Harrison, |. D., "The Analysis of fatigue
test results for butt welds with lack of
penetration defects using a Fracture Me-
chanics Approach," Paper 68, Fracture 1969,
Proc. 2nd International Conference on Frac-
ture, Brighton, April 1969, Published by
Chapman and Hall.
10. Harrison, |. D., Burdekin, F. M., and
Young |. G., "A Proposal Acceptance Stan-
dard for Welded Defects based upon Suit-
ability for Service," Paper presented at 2nd
Conf. on Significance of Defects in Welds,
London, May 1968, The Welding Institute.
11. Soete, W., and Sys, A., "Influence of
Weld Defects on High Fatigue Behaviour,"
Proc. 1st Int. Conf. on Pressure Vessel Tech-
nology, Delft, Sept/Oct 1969, ASME.
12. Archer, C. L, "Research on the Signif-
icance of Defects in Weld Regions of Pres-
sure Vessels," Conf. on Pressure Vessel
Standards, The Impact ol Change, BSI,
IMechE, and The Welding Institute, Lon-
don, November 1972.
13. Harrison, J. D., "British Work on the
Significance of Weld Defects," Svetsen,
Vol. 33, No. 2, March 1974 pp 23-38.
14. "The Significance of Defects in
Welded Pressure Vessels," Welding Insti-
tute Group, Sponsored Project, 1970-1975.
15. ASME Boiler and Pressure Vessel
Code, Section III, Nuclear Vessels, ASME,
New York, 1971.
16. Maddox, S. |., "Fatigue Crack Propa-
gation Data obtained from parent plate,
weld metal and HAZ in Structural Steels,"
Welding Research International, 4(1),
1974.
17. British Standard 1515: Part 1 1965
"Fusion Welded Pressure Vessels, Carbon
and Ferritic Alloy Steels."
18. Boulton, C. F., "Fatigue Life Predic-
tions of Welded Specimens containing Lack
of Penetration Defects at Ambient and
Elevated Temperatures," The Welding Insti-
tute, Members Report E /61 /75, March 1975,
and published in Pressure Vessels and
Piping, Vol.4, No. 3, luly 1976.
19. Maddox, S. J., "A Fracture Mechanics
Analysis of the Fatigue Strength of Welded
Joints," Ph. D. Thesis, University of London
1972.