You are on page 1of 5

Deformation and the Strip Necking

Zone in a Cracked Steel Sheet

Moire method is used to measure the relative opening displacements


between crack surfaces and between the upper and lower boundaries of
the strip necking region as well as the strain near this region

by B. J. Schaeffer, H. W. Liu and J. S. Ke

ABSTRACT~When a tensile stress is applied to a thin S w e d l o w et al 2 have calculated stresses and strains
cracked plate, a strip necking region results ahead of a near a crack tip for a t w o - d i m e n s i o n a l plane-stress
crack tip. The relative opening displacement between the case. Th ei r calculation shows a diffused deformed
crack surfaces and between the upper and lower boundaries region in contrast with the strip deformed region as
of the strip necking region were measured by the moir6 assumed by the Dugdale model. D r u ck er and Rice 3
method. The strains ahead of the strip necking region and attributed the difference p r i m a r i l y to the yield con-
the thickness reduction (therein) were also measured. The
ditions used by these t w o calculations. While t he
measured relative opening displacements were compared
Tresca yield condition was used for the Dugdale
with the calculated values using the Dugdale strip necking
model. The thickness reduction in the strip necking region model, the Mises' yield condition was used by S w e d -
is equal to the relative opening displacement. low et al. 2 Consequently, D r u c k e r and Rice s con-
cluded that m a t e r i a l obeying Tresca's yield criterion
should h a v e a strip necking region, while m a t e r i a l s
Introduction obeying Mises' yield criterion should show a diffused
K n o w l e d g e of the state of d e f o r m a ti o n at a crack region of i n t en si v e deformation. The regions of i n-
tip is i m p o r t a n t to d e t e r m i n e the conditions which tensive d e f o r m a t i o n based on these two calculations
led to crack propagation and fracture. Valuable r e- are shown schematically in Fig. 2. In this study, the
sults h a v e been obtained w i t h elastic 1, elasto-plas- r e l a t i v e opening displacements b e t w e e n crack sur-
tic 2,3, and modified-elastic solutions.4, 5 Empirically, faces and b e t w e e n the upper and l o w er boundaries of
compliance calibration has been used to m easu r e the strip n eck i n g region, as w e l l as the strain near
elastic stress-intensity factor 6 and PhotoStress has the strip n eck i n g region, w e r e measured using t he
been used to m e a s u r e the stress field near a crack tip. 7 moir6 method. The m e a s u r e d opening displacements
Recently, the moir~ m e t h o d has been used to m e a - w e r e c o m p a r e d with t h e calculated values. The t h i c k -
sure the d ef o r m a t io n n e a r a crack tip s,9 and t h e ness reduction in the strip necking region was also
photoelastic-plastic t e c h n i q u e has been used to m e a - measured.
sure the plastic-zone size in polycarbonate. 10
When a specimen is v e r y thin, e x t e n s iv e d e f o r m a - Experimental Method
tion occurs in a thin strip region. This strain con-
The r e l a t i v e opening displacements b et w ee n crack
centration is caused by the necking of the m a t e r i a l
surfaces and the upper and l o w e r boundaries of the
ahead of the crack tip. F i g u r e 1 shows the moir6
strip necking region, and the strain near the strip
fringes on the surface of a thin steel specimen u n d er
a tensile load. T h e r e is a D u g d a le strip necking region necking region w e r e m e a s u r e d w i t h the moir6 t e c h-
nique. Two t h i n steel specimens w i t h central slots
ahead of each of the tips of t h e slot. When necking
takes place, the m a t e r i a l above and below the necked w e r e tested. The slot is a j e w e l - s a w cut, 1-in. long
and 0.007-in. wide. The specimens are 6-in. wide and
region deforms little as the load is f u r t h e r increased.
T h e r e f o r e the upper and l o w e r boundaries of the 0.012-in. thick. The tensile stress-strain curve is
shown in Fig. 3. The 0.2-percent offset yield strength
strip necking region can be t r e a te d as "elastic". This
is especially correct if the m a t e r i a l is elastic and p e r - is 91 ksi. Two specimens w e r e tested. One was loaded
fectly plastic. The r e l a t i v e opening displacement b e- to 55 ksi and the d e f o r m a t i o n was measured u n d e r
load. The o t h er was loaded to 62 ksi and the d e f o r m a -
t w e e n the u p p e r and t h e l o w e r boundaries increases
w i t h the applied load. T h e d e f o r m a t io n of such a tion was m e a s u r e d after unloading.
The specimen surface was polished, cleaned and
cracked thin plate was first analyzed by Dugdale 4 and
s u b s e q u e n t l y by Goodier and Field~ and Bilby, Cot- coated w i t h photo-resist. Th e moir~ grille was
printed onto t h e specimen surface w i t h lines parallel
trell and S w i n d e n 2 ~
to the slot. A line density of 13,400 lines per inch (lpi)
was used. The moir~ fringes w e r e obtained by double
B. 1. Sehaeffer, H. W . Liu and ]. S. Ke are Research Fellow, Pro- exposures before and after deformation.
fessor and Research Assistant, respectively, Department of Chemical
Engineering and Metallurgy, Syracuse University, Syracuse, 57. Y.
13210.
This report was produced under contracts sponsored by AISI and Results and Discussion
NASA. The conclusions and recommendations expressed are those oi F i g u r e 1 shows a picture of moir6 fringes. The ap-
the authors and are not necessarily endorsed by the sponsor.

172 I A p r i l 1 9 7 1
Fig. 1-Moirà pattern of a steel specirnen: applied stress, L L2 -L
55 ksi; 0.2-percent-offset yield stress, 91 ksi; Young's STRAI N - x 10'
modulus, 32 x 10e psi; 0.012-in. thick; 6-in. wide; slot
length, 1.0 in.; pitch of moir6 grille, 1/13,400 in, Fig. 3ÑTensil stress-strain curve of a thin steel sheet

-
x
z
$2
2,

h-
#a
Y
s
'd

DIFFUSED YIELO ZONE STRIP N E C K I N G R E O I O N 3 l


a
OT
Fig. 2-Schematic diagrams of diffused yield c3
zone and strip necking region 5
u

'010 08 OB 04 02 0 02 04 06 08 10
OISTANCI FROMTHECENTER LINE OF THE SPECIMEN (INCH)

Fig. 4ÑOpenin displacement along crack line: a, 55 ksi;


plied stress is 55 ksi. The slot is 1-in. long. The Dug- c'y, 91 ksi; E, 32 x 108 psi; thickness, 0.012 in.; width, 6
in.; slot length, 1.0 in.
dale strip necking region is clearly visible. Along
each fringe, the displacement in the loading direction
is constant. The relative displacement between two
neighboring fringes is equal to the pitch of the grille higher than the calcuiated values. The deviation is
lines, Le., 1/13,400 in. The relative opening displace- approximately 10 percent. The opening displacements
ments between the crack surfaces and between the in the central region of the crack reflect the extent
upper and lower boundaries of the Dugdale strip and the intensity of the deformation of the cracked
necking region were measured and plotted as the section of the sheet. As indicated by eq (1), the de-
solid line in Fig. 4. In the strip necking region, the formation of the cracked section increases with the
accuracy is ±50 pin. ratio U/Q. The steel strain hardens considerably. The
The crack-opening displacement v along the crack proportional limit is in the neighborhood of 65 ksi
Une has been calculated by Goodier and Field5 using and the 0.2-percent-offset yield stress is 91 ksi. A
the Dugdale model. slightly lower value of u y in the calculation will give
a better agreement between the measurements and
the calculation.
The measured opening displacements are lower
than the calculated values in the strip necking region,
+ COS02 log +
(sin02 sine) 2 The measured and the calculated values at the ends
(sin& - sine) 2 of the slot are 0.86 x 10-3 in. and 1.01 x 10-3 in.,
where respectively. The Dugdale model assumes that the
a deformation is confined to the strip necking region,
cos02 = - and the material outside of the region deforms elas-
1
tically. The stress within the strip is uniaxial, equal
x to UY. This is probably correct if a material is elastic
cos0 = - for 1x1 < l, - If
1
< O < If and perfectly plastic. But most materials strain har-
den. If a material strain hardens, plastic deformation
takes place outside of the strip necking region, before
necking takes place.
When the stresses and the deformations increase in
u and o-y are the applied stress and the tensile yield the cracked section, two opposing effects affect the
stress of the material. a is the half-crack length. l is load-carrying capacity of the material ahead of a
the half-crack length plus the length of the strip crack tip. The load-carrying capacity is increased by
necking region, and E k Young's modulus. The cal- the strain hardening of the material, and it is reduced
culated opening displacements are shown as the by the decrease of the cross-sectional area. For a
dashed line in Fig. 4. Values of 32 x 10%psi and 91 simple tensile test, necking takes place when the
ksi for E and o-y were used for the calculations. The effect of area reduction overtakes the effect of strain
relative opening displacement is 2v. hardening. If a material strain hardens strongly,
Â¥^
In the central region of the crack surface, the mea- necking takes place only after extensive deformation
sured values of the relative opening displacements are occurs throughout the test section of a tensile speci-
Fig. 6ÑSchemati diagram of the
various regions of deformation around
a crack
l1
0 01 0I 05
DISTANCE FROM THE TIPOF THE STRIP NECKING REGIONIINCHI

Fig. 5-Strain ahead of the strip necking region

men. In the case of localized necking ahead of a


crack tip, the situation is complicated by the stress
gradient in this region. When a steep stress gradient
is present, necking may take place at a much lower
strain level in the local region ahead of a crack tip.
Therefore, both the rate of strain hardening and the
stress gradient affect the plastic deformation outside
of the strip neeking region. The deformation outside
of the strip necking region is evident by examining
the moi& fringes in Fig. 1. The fringe spaeing cor-
responding to the yield strain of the steel is 0.027 in.
It is evident that plastic deformation takes place Fig. 7-Schematic diagram for necking
above, below and ahead of the strip neeking region. zone ahead of a crack
Figure 5 shows the logarithmic plot of the strains
ahead of the strip necking region and dong the crack
line. Two straight-line segments correlate the data
well. Near the tip of the strip necking region, the Rosenfield,14 experimentally, with etching studies on
strains decrease with (r)-0.73, where r is the distance silicon bon.
from the tip of the necking region. The yield strain The strain concentration in the necked region will
of the steel 1s 2.8 x 10-3 inJin., and the plastic-zone increase the fatigue-crack-propagation rate. Liu has
size from the tip of the strip necking region is 0.09 in. also used strain concentration in the necked region to
It is close to the intersection of these two line seg- explain the rapidiy increased fatigue-crack-propaga-
ments. tion rate da/dN as the applied stress-intensity-factor
It is evident that strip necking region is imbedded range, AK, increases. This strain concentration may
in a diffused plastically deformed region as shown lead to a relation that du/dN is proportional to AK4.15
schematically in Fig. 6. This picture of deformation Therefore, the necking process affects both fracture
was first suggested by McClintock.12 The strip neck- toughness of thin sheets and the fatigue-crack-propa-
ing is the result of the reduction of the cross-seetional gation rate.
area ahead of a crack tip, which causes the reduction The simple model in Fig. 7 indicates that the rela-
of the load-carrying capacity of the material element tive opening displacement is equal to the reduction of
regardless of the yield criteria which a material obeys. the thickness. Figure 8 shows the opening displace-
The calculated length of the strip necking region is ments and the thickness reduction in the strip necking
0.36 in., the measured length is 0.14 in., whereas the region. Both the opening displacements and the thick-
total length of the necked region and the diffused ness reduction were measured after unloading. The
plastic zone ahead of it is 0.23 in. The measured total applied stress was 62 ksi. The specimen width, thick-
length of the plastically deformed region is much less ness, and the slot length were 6 in., 0.012 in. and 1.0
than the calculated strip necking region. It is ex- in., respectively. The opening displacements were
pected that the length of a strip necking region using measured by the moirÃmethod. The thickness reduc-
the Dugdale model should be longer than the size of tions were measured with a microscope. The micro-
the diffused plastically deformed region given by scope was focussed first on the specimen surface, then
Swedlow's two-dimensional elasto-plastic plane-stress on the bottom of the necked region. The reduction of
calculation. Therefore, the total length of the strip the specimen thickness is the difference of the read-
necking region and the diffused plastic region in the ings on the micrometer, which is attached to the mi-
steel specimen should be less than the calculated croscope for focus adjustment. Because of the shallow
length of the strip necking region. depth of focussing at high magnification, the thickness
Liu13 has suggested that, when necking takes place, reduction can be measured accuratelv. Both measure-
deformation is the result of shears on two sets of ments agree with each other exceedingly well. The
orthogonal planes, each inclined 45 deg to the loading maximum deviation at the end of the slot is approxi-
axis, as shown schematically in Fig. 7. The 45 deg mately 10 percent. The simple model shown in Fig. 7
inclined shear planes were verified by Hahn and indicates two shear planes orthogonal to each other,
M E A S U R E D AFTER UNLOADING
ture. Hence Kc is proportional to v/c
However, as the
HALF THICKNESS REDUCTION sheet thickness is close to the optimum point of high-
O P E N I N G DISPLACEMENT est fracture toughness, the relation in eq (2) may not
be valid. For most of materials, the optimum point
occurs when a plate is fairly thick. Therefore the
strip necking mode1 is not applicable.

Conclusions
1. For a material that strain hardens, the Dugdale
strip necking region is imbedded in a diffused region
of plastic deformation.
2. The strip yielding region is primarily caused by
necking, Le., the thickness reduction in the area ahead
of a crack tip.
3. The thickness reduction in the strip necking
region is equal to the relative opening displacement.
The strain at the bottom of the necked region is ap-
proximately given by cave = 2 v / t .
O. 1 0.2 0.3 0.4
DISTANCE FROM CRACK T I P (INCH1
Acknowledgment
Fig. 8ÑHalf-thicknes reduction
and opening displacement in the The work was conducted at the Metallurgical Re-
strip necking region of a steel search Laboratories of Syracuse University. The sup-
specirnen. U , 62 ksi port of both American Iron and Steel Institute (to
JSK) Contract No. 123, and NASA (to BJS) Grant
No. NGR-33-022-105 is acknowledged. The assistance
by Mmes. H. Turner and B. Howden in the prepara-
tion of the manuscript is also acknowledged.
and the deformation leads to a stepped surface. This References
picture is shown for the simplicity of presentation.
1 . Paris, P. C . and Sih, G. C., "Stress Analysis of Cracks,"
Shear deformation occurs on two orthogonal sets of Fracture Toughness Testing and Its Applications, ASTM STP 381,
parallel shear planes. The shears on these planes give 30-83 (1965).
a gradua1 thickness reduction to the bottom of the 2. swediow, 1: L., Williams, M. L . and Yang, W . H., "Elasto-
Plastic Stress and Strains i n Cracked Plates," Proc. First Intematl.
necked region, rather than the abrupt thickness re- C o n f . o n Fracture, Sendai, Japan, 1, 259-282 (1965).
duction shown in Fig. 7. 3. Drucker, D. C. and Rice, 1. R., "Plastic Deformation i n
Brittle and Ductile Fracture," Engineering Fracture Mechantes,
If one assumes that the in-plane lateral strain is 1, 577-602 (1970).
zero, the longitudinal tensile strain at the bottom of 4. Dugdale, D. S., "Yielding of Steel Sheets Containing Slits,"
the necked region is equal to the contraction strain in Jnl. o f Mech. and Phys. o f Solids, 8, 100-104 (1960).
5. Goodier, 1. N. and Field, F. A., "Plastic Energy Dissipation
the thickness direction. Hence, the tensile strain is in Crack Propagation," Fracture of Solids, Interscience, N e w York,
given by 103-1 1 8 (1963).
6 . Gross, B., Srawley, 1. E. and Brown, Ir., W . F., "Stress In-
tensity Factors for a Single-Edge-Notch Tension Specimen b y
Boundary Collocathm of a Stress Function," N A S A T N D-2395,
( 1964).
7 . Gerberich, W . W . , "Plastic Strains and Energy Demity h
where t is the sheet thickness. For the same value of Cracked Plates-Part 1-Experimental Technique and Results," EX-
opening displacement, a thinner specimen gives a P E E T A L MECHANICS, 4 ( I l ) , 335-344 (Nov. 1964).
8. Underwood, 1. H. and Kendall, D . P., "Measurement o f
higher strain. This relation is reasonable for a very Microscopie Plastic-stratn Distributions i n the Region of a Crack
thin specimen. Tip," EXPERIMENTAL MECHANICS, 9 (7), 296-304 (July 1969).
9 . Lin, H . W., Gavigan, W . 1. and Ke, 1. S., " A n Engineering
The fracture toughness of a material increases from Analysis of Ductile Fractures," Intematl. Jnl. o f Fracture Mechanics,
the plane-strain fracture-toughness value to an opti- 6 ( 1 ) (1970).
mum value as the plate thickness reduces. As the 10. Brinson, H. F., " T h e Ductile Fracture of Polycarbonate,"
EXPERIMENTAL MECHANICS, 10 (2), 72-77 (Feb. 1970).
thickness is further reduced beyond the optimum 11. B a y , B. A., Cottrell, A. H. and Swinden, K. H., " T h e Spread
point, the fracture toughness reduces.16 Liu17 has of Plastic Yield from a Notch," Proc. Royal Society, 272, Series A
used the concept of strain concentration in the necked (1963).
12. McClintock, F. A., "Discussion o n Fracture Mode Transition
region to explain the reduction in fracture toughness for a Crack Traversing a Plate," Tram. ASME, Jnl. o f Basic Engrg.
as the sheet thickness decreases beyond the optimum 82, 4 2 3 (1960).
13. Liu, H . W., "Discussion o n W . Weibull's Paper o n t h e E f f e c t
point. It was assumed that cave at the crack tip in the of Size and Stress History o n Fatigue Crack Initiation and Propa-
necked region is a constant at fracture. The relative gation," Proc. o f the Intematl. Symp. o n Crack Propagation, Cran-
crack-opening displacement is related to the strain- field, England (1961).
14. Rosenfield, A. B., Dai, P. K. and Hahn, G . T., "Crack Ex-
intensity factor tension and Propagation Under Plane Stress," Proc. First Internatl.
Conf. o n Fracture. Sendai. Iawan. 1. 223-258 (1965).
15. Liu, H . W.; " ~ a t i g & Crack b o p a g a t i o n a n d t h e Stresses and
Strains i n t h e Vicinity o f a Crack," Applied Materiais Research,
(Oct. 1964).
Equations (2) and (3) lead to a fracture toughness, 16. Ir&, G . B., Kies, 1. A. and Smith, H . L., "Fracture Strengths
Relative t o Onset and Arrest o f Crack Prowaeation,"
. - . Proc. Am. Soc.
Testing Materiais, 58, 6 4 0 (1958).
17. Liu, H. W . , "Yielding Initiation Process and Its Effect o n
Fracture of Cracked Plate," Proc. First Intematl. C o n f . o n Frac-
where cc is the critical strain at the crack tip at frac- ture, S q d a i , Japan, 1, 191-212 (1965).