,".Jr.....

2010 -acjr

+.rrjr"1,4r'.r

Measuring the Residual Stress by Using New

Experimental Iethods Depend on the Stress
Concentration.
Dr Sabah Kh Hussein

Suhad D Salman

Dr Kadhim K Resan

Al Mustansanya

Unv
^DA33'

1 =

=:J41 L_

W`

L_ 1 _1: 41

_
h Lc_
I JI 1 -1 LI

^, 4

, ,

=:^14

JL

it ,Jl

1 l

PIFl t
1 3

4 1 ' IL 4
. 1

ri tt
1

)'Jtt , LUl

11

'

=:J4! L l l
11^ L

14

IYI

ILL

3. b=:J
L L II^

1 1 h
Abstract
The residual strcss is that which remttns in a body thatis stationary
and at cquilibHum with its surrclundings lt can be vcry dct
thc pcrfollllancc of a matcrial or thc lifc of a componcnt

incntal to

Altcrnativcly,beneflcial rcsidual strcsscs can be introduccd

dclibcratcly Residual strcsscs arC mOrc difflcult to prcdict than thc in

cxpeHrnental incthods to scrvicc stresscs Thcrc arc rnany different

measure thc rcsidual strcss such as holc dHlling and x

ray ln this

paper,the new incthod is proposed to mcasurc thc residual strcss in

weldment,this mcthod dcpends on thc strcss concentration factor for

210

:^

1
,3 Io

tt^ LJ.

circle hole with different distance from weld line for different
specimens (five specimens) , the hole is the region of measure the
residual stress .Also , the heat treatment is applied for different
specimens as same condision and dimention to measure the decrease
in the stress .The results are in a good agreement with another
method.

Keywords : residual stress , stress concentration factor,welding

Introduction
Residual stresses are those which are not required for an engineering
structure to maintain equilibrium with its environment. Although they
can have many different origins, residual stresses are always the result
of some form of misfiU either between different parts, different regions
within the same part, or even different phases within a microstructure
t'1. Welding residual stresses arise as a consequence of the
heterogeneous application of energy and localized fusion.
With modern analytical and computational techniques it is often
possible to estimate the stresses to which a component is subjected in
service. This in itself is not sufficient for the reliable prediction of
component performance. Indeed, in many cases where unexpected
failure has occurred, this has been due to the presence ofresidual
stresses which have combined with the service stresses to seriously
shorten component life. On the other hand, compressive stresses are
sometimes introduced deliberately, as in shot peening which is used to
improve fatigue resistance. Furthermore, in natural or artificial
multiphase materials, residual stresses can arise from differences in
thermal expansivity, yield stress, or stiffness. Considerable effort is
currently being devoted to the development of basic framework within
which residual stresses can be incorporated into design in aerospace,
nuclear, and other critical engineering industriestu.
The residual stress in the vicinity of a weld can have a large
influence on structural integrity. Here the extent to which the
martensite-start temperature of the weld filler metal can be adjusted to
mitigate residual stress distributions in ferritic steel welds has been
inveitigated by Francis and et all2)

Dai and et all3) were study weld residual stresses often approach, or
exceed, the yield strength of the material, with serious implications for
the integrity of engineering structures. It is not always feasible to
measure residual stresses, so integrity assessments often rely heavily on
211

numerical models. In ferritic steels, the credibility of such models

depends on their ability to account for solid-state phase transformations
that can have a controlling effect on the final residual stress state.
Furthermore, a better understanding of weld transformations provides
an opportunity to engineer the weld stress state and microstructure for
improved life. In this paper the complementary merits of synchrotron
X-ray and neutron diffraction are exploited both to verify and refine
weld models and to inspire the development of weld filler metals to

control weld stresses.

Basic stress analysis calculations assume that the components are
smooth, have a uniform section and no irregularities. In practice
virtually all engineering components have changes in section and / or
shape. Common examples are shoulders on shafts, oil holes, key ways
and screw threads. Any discontinuity changes the stress distribution in
the vicinity of the discontinuity, so that the basic stress analysis
equations no longer apply. Such 'discontinuities' or 'stress raisers' cause
local increase of stress referred to as 'stress concentration'.

The 'theoretical' or 'geometric' stress concentration factor K or Kr is

to relate the actual maximum stress at the discontinuity to the
nominal stresstal.
used

= max direct stress / nominal direct stress and

Ks = rnilX shear stress / nominal shear

stress

......... (l)
....... (2)

In published information relating to stress concentration values the

nominal stress may be defined on either the original 'gross cross
section'or on the'reduced net cross section'and care needs to be taken
that the correct nominal stress is used.

The subscript 't' indicates that the stress concentration value is a

theoretical calculation based only on the geometry of the component
and discontinuity.

This localization of high stress is known as stress concentration,

t5l.
measured by the stress concentration factor (Kt).which is defined as
Kt=o,orrl6oo. ... (tension or bending).

212

2010,ltur

r.rrJl

.............

'..

-liJ.iJl r':ds

alx

The stress concentration factors are obtained experimentally or

mathematically by such means as photo-elasticity, classical method
(tensile test method), precision strain gauges, electrical analogy for
torsion and the finite element method. The experimental work, when
conducted with sufficient precision, excellent agreement is generally
obtained with the mathematical stress concentration factors.
The flow analogy is probably the best single means of visualizing

flow analogy is important because it gives us

physical
picture
why
does stress concentration exist. It can also be
of
a
used as a qualitative tool to leam what to do to decrease stress
concentrationt6l. For the lowest Kt the fillet radius r must be as large as
possible, but r is usually limited by the design of the mating part. A
larger fillet can be used and a lower Kt is obtained by under cutting the
shoulder, sharp edges.
The general progress in both theoretical analysis and experimental
measurements gradually stimulated the interest of the engineering
public in stress concentration .Its understanding was spread wider by
the second world war. In the mean time ,new theoretical results for
stress concentration factors had been found by many mathematicians
and engineers as reported by FrochettTl
Fessler and Woods t8l studied the effects of shape and the position of
load application on the magnitude and position of the peak stress in
stepped bars and a round T-slot were determined. Photo----elastic
models were used for this purpose.. The stress concentration factors for
any practical shape could be quickly determined from these results.
This study gave the lowest peak stress as well as the lowest value of Kt
for any particular fillet radius ratio (r/a) .
Hyde and Webstertel submitted two reports the first dealt with an
experimental method. The results from eleven tests of lead alloy model
shouldered tubes, subjected to constant axial mechanical loads and
cyclic , axisymmetric thermal shocks were described. Strain gqlges
were used to measure stress concentration .The second reporttlol dealt
with the same procedure as that for the experimental work but used the
finite element method.
Okubo and Katotrll studied the stress concentration factors at shallow
grooves in shafts under torsion experimentally and theoretically. The
discrepancies between theoretical and experimental results were
discussed, and the results were also compared with Neubers solution
stress concentration .The

213

.The experimental technique used was based on the observation of the

number and size ofthe grains grown in the copper plating ofthe shafts.
Another investigation was reported by Kuange and freederick tl2l. In
this study circular and sharp notched specimens made from 304
o
stainless steel, had been heated to 650 C .The notch models were
subjected to both constant and cyclic loading.
Navio Hasebe and Mikiya Okumura{lllstudied the problem of two
bonded, dissimilar half-planes containing an elliptical hole on the
interface.
In this paper the residual stress was determined by new method ,this
proposal method depends on stress relief from heat affective zone .

The development of the basic stress equations for tension, compression,

bending, torsion and combined loading involves the analyses of certain
geometrical shapes by using the theory ofelasticity. Thus it is possible
to determine the values of stress-concentration factors.
Also, the stress concentration can be determined by using the finiteelement technique or theory of elasticity equation. This is a very
powerful and relatively a new approach made possible by the advances
in computers and computer-aided design methods in recent years tlal.
The theoretical basis of the finite element method for plane stress
analysis was described in details by Zienkiewicz trsl and the general
principles will not be reiterated here.
The methods to find the stress concentration factor (S.C.F) by using
theory of elasticity equation depend on stress concentration in tip. The
stress at the point away from the hole is practically uniform and the
maximum stress will be induced at the edge of the hole. The maximum
stress is given by(for ellipse hole)tl6l:

d.*

2a

- o(l + b. ).......(4

Where:
a: major length.
b: minor length.
When a/b is large, the ellipse approaches a crack transverse to the load
and the value of Kr become very large .In circular hole a/b =1,
therefore:

ax= 3)

(5

214

Experimental work
Material properties
The material used in this work is from low carbon steel (mild steel).
The chemical properties are listed in table (1). The percentage of
carbon (Vo) rcfers to that material of low carbon steel .The
microstructure of sample from this material is shown in fig.(l).
Table

(l)

chemical composition of base material carbon steel

Fe

82.9

0.923

Si
0.714

Mn

AI

0.608

0.0374

Co

Cr

2.03

4.37

Cu
0.368

Mo

Ni

1.51

Ti
4.94

V
0.02

1.63

Specilnen preparation
All tested specilnen have been cut according to the dimensions

shown in flg(1)in WhiCh:

L:length of speciinen=20011m
w:width of specilnen=24 1nm
t:thickness of specilnen=5.26 1nln
d=10 111un
x

0,5,10,15,20,25 rrm

Fig(l).

schematic graph
of tested weldment specimen
Each specimen is cut into two equal length specimens and then welded
using Arc welding technical procedure with Butt joint at the mid span
of specimen. One continuous line of weld is used to ensure the uniform
of material properties .Also the welding technique is used at the up and
down of specimen to increase the strength of weldment plate and to
penetrate the weld metal. To compare the tested specimen with
different hole position ,we use five positions of hole centre with respect
to the weld metal

215

centre line as well as one hole at the centre of weld line .fig(l) (where ,
hole diameter d=10 mm) at the weld metal .
The second and third specimen (x=5 mm, l0 mm ) include the effect
of stress concentration which gives stress relive of residual at the heat
affective zone (HAZ). A simple photograph of this specimen is shown

in fig.(2)
Fig.(2) photograph of tested specimen

The other specimens have hole at (x=15,20and25 mm) and the

tensile test are process.The comparison of results is achieved with the
base welded specimen (with out hole).
Many experimental methods are recommended to find the stress
concentration factors in specimens that have notches; the methods were
compared with the theoretical procedure to show the difference
between the experimental and theoretical results. These methods can be
classified in to two types:
l -Destructive method (tensile test method).
2- Indestructive methods(strain gauge, photo elasticity x- Ray ,.etc).
In the present work , the tensile test method was employed.

Tensile test analysis:

The welded specimen has been tested with comparatives tensile test
equipment shown in fig(3 ) .

216

Fig.(3) tensile test equipment

The velocity of equipment is (2 mm,/min ).The results of tensile test
are a stress strain curve for each specimen. A sample of photograph of
this specimens before and after fracture is shown in fig (4).

Fig.(a) fractured specimen with (x=0.0 mm)

This is a simple experimental method for measuring the stress
concentration factors. The method was implemented using a model
material, tested on sharp, rectangular and hole configurations,having
t''i .This
stress concentration factors knewn from classical analysis
method included the measurement of the strength (cu1) without notch
was determined and another specimen with notch was subjected to
tensile force to get (oux) .Then using equationll3l, the stres's
concentration factor was calculated:
217

The tensile strength of un notched model oult T

Nominal tensile strength of notched model oult T
..........(6)
=

Results and discussion

The results of the stress concentration for the circular notch specimen
were presented in Fig(6).The cooling of a localized volume of material
from its melting point to ambient
temperature involves a significant reduction in volume . Since the
weldment structure remains continuous, there must be a mechanism
whereby this difference in strain is accommodated .This achieved in
part through plastic strain of the material. The balance is accommodated
through elastic strain. This results in a stress field distributed
throughout the material after cooling, are known as residual stress.
The calculated values of the stresses concentration factor without heat
treatment specimen are presented with respect to the distance from the
weld line, it shows clearly when the distance is increased the S.C.F is
not clear. The local stress concentration started froml.42 cm and
reaching a maximum value at about 5 cm from the weld line, then it
decreased

until it reached

a constant value.

The calculated values of stress concentration factor for with heat

treatment specimens are presented with respect from weld line , as
shown in fig(6).It showed the stress concentration started from 1.46
cm,then it decreased until it reached a constant value. The different
values are between the results with and without heat treatment
because ofthe residual stress effect . The residual stresses that
develop in the weldment often exceed the yield strength of the
material. Therefore, the material yields and the remaining residual
stress can be at the level ofthe yield stress , as the full elastic strain

limit has been reached .

ln a multi pass weld the residual

stress field produced is

triaxial

Since the weld metal is sufficiently constrained by the parent metal in

each principal direction , the residual stresses also act in these three
directions .This residual stresses field can considerably decrease the
strengh of a weldment , therefore the stress concentration factor is
effected by residual stress .
ln midstance of specimen the residual stress is compression and the
applied load is tension therefore the value of stress concentration is
decreased.

218

o'm

uB,r,"n."

l?'ffi'" n

.fS ln" tc,ff'@

Fig (6) variation of stress concentration factor

with weld metd distance(x)

500

450

4()0

350

Fig.(
7) Experimental transverse residual stresses for
weldment steel sPecimen.

2t9

mo

Fig.( 8) Experimental transverse residual stresses for heat treatment

weldment steel specimen.
Conclusion :

l-

The new experimental method has obtain a good result in

comparion with traditional methods

2-

The stress concentration factor is alternative value from the weld

line is increasing near the zero distance then decreasing about 15
mm then increasing again because the value of residual stress is

3-

The heat treatment is decreasing the residual stress and changing

the value of stress with distance from weld line , this results are due
to re-crystalline of grain .

alternative

References

l-P. J Wither and H.K.Bhadeshia, " Measurment techniques part I "

materials science and technology vol 17 ,355-365,2N1.
2-J.A. Francis , D. H. Bhadeshia2 and P. J. Withers " ASME pressure
vessels and piping division conferences " paper, San Antonio , Texas
, ppl-8, 2007.
3-H.Dai " Metallurgical and Materials Transactions A 39A ,30703078 , 2008.
220

4-R.E.Peterson " Stress concentration factors " john Wiley Inc, 2000.
5-J.E. Shigley " Mechanical Engineering Design " Mc.Graw -Hill ,

2003.
6-P.A. Thornton and V.J.Colangelo " Fundamental of Engineering
Materials " Prentice-Hall , INC ,Englewood Cliffs , 1985.
7-M.M.Frochet " Photoelstic studied in studied in stress
concentration " mechanical engineering, Vol58, Augest 1980.
8- H. Fessler and P.J Woods " stress concentration Axially Loaded
unsymmetrical projection an flat Bars", Journal of Strain analysis ,

Vol

17 ,

No 1, 1982

9- T.H. Hyde and J.J. Webster " Experimental Thermal Ratchetting

data for a component with a stress concentration " Journal of strain ,
Vol 17 , No 4 ,1982.
l0- T.H. Hyde and J.J. Webster " Finite Element Method to measure
Thermal Ratchetting data for a component with a stress concentration
" Journal of strain , Vol 17 , No 5 ,1982.
I l- H. Okubo and A. Kato " stress Concentration Factors of Hallow
Grooves in Torsion " J. Of Strain Analysis , Vol 12 , No.2 ,1984.
l2- Kuange and Freederick, "enhancing Finite element boundary
Stress for Plane and Axisymmetric Situations " J. Of Strain Analysis

Vol2l , No. 1, 1990.

l3- Navio Hasebe and Mikiya Okumura "Bonded Bimaterial Half
Planes with Semi-elliptical Notch under tension along the interface "
J. of applied mechanics , Vol 59, No.77 ,March 1992.
l4-T.Chandrupatla and A.D. Berlegunda " Introduction of finite
element in Engineering " Prentice Hall , Inc, 1991.
15-O.C. Zienkiewit " Plane Stress and Plane Strain , The Finite
element method " 3'd edition, Mc Graw Hill , New York, 1977 .
,

16- D.Jaseph " Gear Root Stress Optimization using photoelastic

optimization techniques " copyroght 1989 Society of Automotive

Engineering, Inc.

221