NASA

Technical

Memorandum

83200

STRESS-INTENSITY FACTOR EQUATIONS FOR

CRACKS IN THREE-DIMENSIONAL FINITE
BODIES

J, C, Newman, Jr, and I. S, Raju

August 1981

(NASA-T,_832 O 0;
ST,{ESS-IhTNSI_
_ FACl CR
_QOAIIONS
FOB CRACKS
I_ THBEE-DIHENSIONAL
FINITE
BODIES
(NASA)
51
_ HC A0_/MF
AOl
CSCL

N81-31578

2OK
G3/39

NASA
National Aeronautics and
Space Administration
Langley Research Center
Hampton, Virginia 23665

0atlas
27371

STRESS-INTENSITY

FACTOR EQUATIONS

THREE-DIMENSIONAL

FOR CRACKS

IN

FINITE BODIES

J. C. Newman, Jr. l and I. S. Raju 2

NASA Langley

Research

Hampton, Virginia

elliptical

corner cracks,
corner
tensile

cracks,

stress-intensity

semi-elliptical

applicable,
equations

thickness

surface

were obtained

The stress-intensity

of these crack

was included

cracks,

give stress-intensity

from current

plate

configurations.

crack front were also included,

estimates.

For all combinations

IResearch

Engineer,

factors

to remote

as a function
and, where

used to develop

the

finite-

The ratio of crack depth to plate

depth

to crack

thickness

on stress-intensity

but were generally

of parameters

NASA Langley

subjected

A wide range of configuration

ranged from O to l, the ratio of crack

The effects of plate width

and quarter-elliptical

three-dimensional

from 0.2 to 2, and the ratio of hole radius to plate

0.5 to 2.

for

quarter-elliptical

thickness,

factors

and previous

in the equations.

equations

The plates were

angle, crack depth, crack length,

analyses

parameters

These equations

hole radius.

factor

surface cracks at a hole,

cracks at a hole in finite plates.

of parametric

element

empirical

semi-elliptical

loading.

23665

SUMMARY

This paper presents

embedded

Center

length

ranged

ranged from

variations

along

based on engineering

investigated,

Research Center,

the empirical

Hampton,

VA

23665

2Assistant Research Professor, The George Washington University,

Jointlnstitute
for Advancement of Flight Sciences, NASA Langley Research
Center, Hampton, VA 23665

the

equations

were generally

within 5 percent

except within a thin "boundary

surface.

However,

These equations
crack-growth

layer" where

the proposed equations

in this region because

of the finite-element
the crack

are expected

front intersects

rates as well as in computing

fracture

loads for these types of crack configurations.

a free

to give a good estimate

of a study made on the boundary-layer

should be useful for correlating

results,

effect.

and predicting

toughness

fatigue-

and fracture

INTRODUCTION
In aircraft

structures,

fatigue failures

usually occur from the initiation

and propagation of cracks from notches or defects in the material that are
either embedded,on the surface, or at a corner.
elliptic

or near-elliptic

crack fronts.

fracture

strength, accurate stress-intensity

these crack configurations.

factor solutions are needed for

Instead, investigators

for three-dimensional cracked bodies are available

One of these, an elliptical

crack in an infinite

jected to uniform tension, was derived by Irwin [l]

sis by Green and Sneddon[2].
approximate analytical

For finite

bodies, all

Enetanya [lO] used the alternating

For a quarter-elliptic

method. Shah [ll]

equation method

corner crack in a
method; Kobayashi and

estimated the stress-

factors for a surface crack emanating from a circular

single corner crack emanating from a circular

Kullgren [12] used a finite-element-alternating
3

used

method, and Heliot,

plate, Tracey [8] and Pickard [9] used the finite-element

intensity

respectively,

[7] used the boundary-integral

to obtain the sameinformation.

thick-

factors along the crack

Raju and Newman[5,6] used the finite-element

Labbens, and Pellissier-Tanon

surface crack in a semi-

and Kobayashi [4],

method to obtain stress-intensity

sub-

solutions have required

surface crack in a plate of finite

ness, Smith, Emery, and Kobayashi [3],

front.

solid

using an exact stress analy-

methods. For a semi-circular

solid and a semi-elliptical

the alternating

have used approxi-

factors.

Very few exact solutions

infinite

and

methods, experimental methods, or engineering estimates to

obtain the stress-intensity

in the literature.

To predict crack-propagation life

But, because of the complexities of such problems,

exact solutions are not available.

mate analytical

These cracks propagate with

hole.

For a

hole in a plate, Smith and

method to obtain the stress-

intensity

factors.

finite-element
a plate.

Hechmerand Bloom[13]

method

for two-symmetric

All of these approximate

for ease of computation,

corner cracks emanating

results, except

[6,9] and the corner crack I_9], were

However,

and Raju and Newman

presented

[14] used the

from a hole in

that for the surface

in the form of curves

results expressed

crack

or tables.

in the form of equations

are preferable.
The present paper presents empirical
factors

for a wide

to uniform

equations

variety of three-dimensional

crack configurations

tension as a function of parametric

length, plate thickness,

angle, crack depth,

and hole radius (where applicable),

Figure I.

These crack configurations,

elliptical

crack, a semi-elliptical

crack, a semi-elliptical

surface

shown

surface

dimensional

for example

see

an embedded

crack, a quarter-elliptical

corner

crack at a hole, and a quarter-elliptical

were based on the stress-intensity

finite-element

subjected
crack

in Figure 2, include:

corner crack at a hole in finite plates subjected

The equations

for the stress-intensity

analyses

conducted

to remote tensile
factors obtained

loading.

from three-

herein and from the literature

[5,14], and cover a wide

range of configuration

parameters.

The ratio of crack

depth to plate thickness

(a/t) ranged from 0 to l, the ratio of crack depth

crack length (a/c) ranged from 0.2 to 2, and the ratio of hole radius
thickness

(R/t) ranged from 0.5 to 2.

stress-intensity

variations

The effects of plate width

estimates.

to plate

(b) on

along the crack front were also included,

generally based on engineering

to

but were

SYMBOLS
a

depth of crack

width or half-width

length or half-length

boundary-correction

factor on stress intensity

Fc

boundary-correction

factor for corner crack in a plate

Fe

boundary-correction

factor for embedded

Fs

boundary-correction

factor for surface

crack

in a plate

Fsh

boundary-correction

factor for surface

crack

at a hole in a plate

Fch

boundary-correction

factor for corner crack at a hole in a plate

finite-width

of cracked plate
of crack

correction

angular

function derived

gi

curve fitting

half-length

KI

stress-intensity

Mi

curve fitting functions

shape factor for an elliptical

of cracked

(see Fig. 2)

crack

in a plate

factor

functions

(see Fig. 2)

from embedded elliptical

defined

in text

plate

factor (Mode I)
defined

in text
crack

radius of hole
remote uniform tensile stress
thickness

or half-thickness

Poisson's

ratio

parametric

of plate

angle of the ellipse

(see Fig. 2)

crack solution

THREE-DIMENSIONAL
Three-dimensional
singularity

finite-element

elements were

factor variation
quarter-elliptical

to remote tensile

placements

conditions

The stress-intensity

ences 5 and 15.

method,

In this method,

ahead of the crack front were

The stress-intensity
finite-thickness

the crack were

factors were obtained

yses by using a nodal-force

models,

(see Fig. 2).

is the shape factor

elliptic

The

the same as those

on certain

used

The only
faces of
dis-

(set equal to zero), except

the normal displacements

free.
from the finite-element
of which

the nodal forces

are given

the stress-intensity
the crack

anal-

in refer-

normal to the crack

factor, K I, at any point along

KI = S

complete

the details

used to evaluate

plate was taken

crack at a hole

cracks at holes, the normal

For a corner crack in a plate,

on the two faces that intersect

surface

that were imposed

cracks and surface

and

crack, a

cracks and corner cracks at holes.

on three planes of symmetry were fixed

for the crack surface.

where

loading

elliptical

models used for these confi'gurations were

For embedded

linear-strain

the mode I stress-intensity

corner crack, and a semi-elliptical

were the boundary

the models.

E5,14] using

along the crack front for an embedded

in references 5 and 14 for surface

differences

analyses

ANALYSES

used herein to calculate

in a finite plate subjected

finite-element

FINITE-ELEMENT

plane and

factors.
front in a

to be

F _, c' t'

for an ellipse

(1)

and is given by the square

integral of the second kind [2].

of the

In the finite-element

the w_dth (b) and length (h) of the plate were taken

to be large emough

so that they would have a negligible

correction,
tion

The boundary

of crack depth, crack length, hole radius (where applicable),

system used to define

angle of the ellipse.

the parametric

Very useful empirical

ref. 6).

The expressions

The maximum

was written

corner

for

(see

for_-

(2b)

factor by using these equations

for all values of

a/c.

(Rawe's original

for

equation

a/2c).
factors, F, obtained
elliptical

from the present

to uniform

for various parametric

semi-elliptical

surface

to uniform

finite-

crack and the quarter-elliptic

tension

cracks emanating
tension for

are given

angles and several

3a and 3b give the boundary-correction

finite plate subjected

by Rawe

Q = l + 1.464

crack in a finite plate subjected

symmetric

have been developed

(2a)

in terms of

Tables

for _-<__I

results for the embedded

ratios.

the coordinate

are

error in the stress-intensity

l and 2, respectively,

thick-

angle.

expressions

The boundary-correction
element

Figure 3 shows

plate

Q = 1 + 1.464

was about 0.13 percent

a/t

on stress intensity.

F, accounts for the influence of various boundaries and is a func-

ness, and the parametric

effect

R/t

equal

a/c

factors

from a circular

in Tables
and

for twohole

in a

I and 2, respectively.
s

STRESS-INTENSITY
In the following

sections,

for embedded elliptical

elliptical

subjected
were

polynomials
emanating

semi-elliptical

stress-intensity
surface

surface

are presented.

from systematic

cracks at a hole,

in terms of

The particular

curve-fitting

procedure

a/c, a/t, and angular

from holes, polynomial

equations

factor equations

cracks, quarter-

corner cracks at a hole in finite plates

to remote tension

obtained

the empirical

cracks, semi-elllptical

corner cracks,

quarter-elliptical

FACTOR EQUATIONS

and

(see Figure 2)
functions

chosen

by using double-series

functions

in terms of

of

For cracks

c/R and

were

also

used.

Embedded
The empirical

stress-intensity

crack in a finite plate,

Elliptical

factor equation

Figure 2a, subjected

ting to the finite-element

results presented

limiting behavior

approaches

as

[l] were also used.

a/c

Crack
for an embedded

elliptical

to tension was obtained

herein

(Table l).

zero or infinity,

by fit-

To account

for

the results of Irwin

The equation is

(3)

for

0 <_a/c <__, c/b < 0.5, and

-x <_b <_

provided

that

a/t

satisfies:

a<t,
,,(a+o0)
,oro<a<c
a<
t

for 0.2 < a<

-c-

(4)

The function

Fe

accounts

(a/t), finite width

(c/b), and angular

Fe :

The term in brackets

g = f = 1).

For

MI + H2

gives

The function

bedded elliptical
correction

for the influence

factor.

of crack shape

location

(), and was chosen

+ M3

was taken

from the exact

crack in an infinite solid

The function

(5)

g f fw

factors

[I] and

is a fine-tuning

fw

at

size

as

It)It) l

the boundary-correction
f@

(a/c), crack

@ = _/2

solution

(whe re

for an em-

is a finite-width

curve-fitting

function.

a/c < l:

MI --I

M2 =

(B)

O. 05
0.II + (C/2

(7)

0.29

(8)

M3

0.23 +

g = l

cos

(9)

and

f@ =

cos2 + si

(lO)

The finite-width

correction,

fw:

for

c/b < 0.5.

fw' from Reference

ecI_ c

Is

(11)
_r_l 1 i/2

(Note that for the embedded

of the full plate thickness.)

For

6 was

crack, t

is defined

as one-half

a/c > l:

Ml : _

(12)

and

(13)

sin 2 + cos2 I I/4

The functions

M2, M3, g, and

are given by equations

(7), (8), (9), and

(ll), respectively.
As
equation

a/c

approaches

zero and

equals

_/2, the stress-intensity

factor

reduces to

KI = S _

for

+ 0.455

(14)

+ 1.261

c/b = O.
Equation

(14) is within

a/t < 0.55 and within

l percent of the accepted

3 percent

for

a/t < 0.8.

lO

solution

[16] for

As _ a/c

approaches

infinity and

equals zero,

the equation

reduces

KI:s
Equation

and various

a/t

2 percent

is defined

the proposed

elliptical

crack

angle.

At

: 0

and

because

(symbols).

the equation

factors

case.

solid

crack

that the finite-element

1.5 percent

in an infinite

solution

below the exact solution.

for

Because

higher than the finite-element

a/t = 0.2

"percent

This defi-

in some cases vary

exact solution

for an elliptic

{solid curves)

and the finite-element

for that particular

the stress-intensity

a/c = 0.4

{Herein

The dashed

slightly

a/t

Fe, is plotted

7/2, the equation

results

between

value

factor,

4 for

from small to large values along the crack front.)

indicate

as

and the finite-element

is shown in Figure

of the finite-element

by the maximum

is necessary

equation

The boundary-correction

as the difference

results normalized
nition

between

ratios.

the parametric

is within
error"

comparison

for an embedded

against

[16] for this configuration

unity.

A typical
results

(IS>

(15) is the accepted solution

approaches

to

curve shows

[l].

These

is probably

the proposed

equation

results, the equation

the

results
about
is

should be very

accurate.

Semi-elliptical
An empirical
elliptical
obtained
element

surface

equation

for the stress-intensity

crack in a finite plate,

from Reference
results

Surface

6.

Crack
factors

Figure 2b, subjected

This equation was previously

from Raju and Newman

[5] for

II

for a semi-

a/c

values

to tension was

fitted to the finitefrom 0.2 to I.

An

equation for
limiting
[17]

a/c

greater than unity was developed herein.

behavior as a/c

for a single-edge

To account for the

approaches zero, the results of Gross and Srawley

crack were also used.

The equation

is

(16)

KI = S _-_ a Fs lac-'t-'b-'
a c I

for

0 <_a/c <_2, c/b < 0.5, and

satisfies, equations

(4).

0 <_@ <7,

The function

Fs

again, provided

that

a/t

was chosen to be

(17)

For

a/c < 1:

00 (a)

(18)

0.89
M2 = -0"54+

0.2+(a)

(19)

I
M3 = 0.5 -

0.65 + a

+ 14

(20)

and

f@

is given by equation

given by equation

(ll).

(lO).

Equations

The finite-width
(17) through

12

correction,

(21) were taken

fw' is again,
from Reference

For

a/c>

l:

(22)

(23)

(24)

(25)

g = l + [O.l + 0.35

and

and

fw

are given by equations

Figure 5 shows the distribution

the crack front for a semi-elliptical
a/t

ratios.

The proposed equation

finite-element
For

results

equation

These predicted

measurements
Reference

surface

crack with

(solid curves)

results.

a/c = 2

Fs, along

for two

3 percent

surface-crack-growth
crack-growth

made on steel,

patterns

patterns were

of the

under

titanium alloy, and aluminum

(a/t).

13

between

tension

alloy

the

6, the equation

in good agreement

18, the equation was also used to correlate

(a/c) and crack sizes

comparison

In Reference

data for a brittle epoxy material within +_lO percent

shapes

factors,

is within

6 gives a more complete

and the finite-element

was also used to predict

loads.

of boundary-correction

(symbols).

a/c <_ l, Reference

proposed

(13) and (ll), respectively.

fatigue
with

materials.

surface-crack
over a wide

In

fracture

range of crack

Quarter-elliptical
The empirical

stress-intensity

factor equation

corner crack in a finite plate, Figure

by fitting
equation

to the finite-element

0.2 <_ a/c _2,

was

for a quarter-elliptical

2c, subjected

results presented

to tension was obtained

herein

(Table 2).

The

is

KI = S

for

Corner Crack

chosen

Fc

(26)

_, _,

a/t _ I, and 0 <_ <_d2

for

c/b < 0.2.

The function

as

(27)

(Although
tion

the effects of width are significant,

(27) because results on finite-width

parison.)

For

fw

is not included

plates were not available

in equafor com-

a/c ( l:

.,
;,o.
o )

(28)

M2 = -0.44 +

(29)

1.06

(30)

gl
i Ioo8
sin,
o41 3 Ic
(31)

14

(32)
g2 = ] + [0.08

and

is given by equation

For

a/c >I :

+ 0.15

(I0).

., #(,o, oo, )

(33)

(34)

(35)

gl = l +

0.08 + 0.4

(36)

(l - sin )3

(37)

and

f@

is given by equation

(13).

Figure 6 shows boundary-correction

tors for a quarter-circular
0.2) under tension
solid circular
Tracey

loading.

symbols

[8] and Pickard

(b) and half-length

factors obtained by several

corner crack in a finite-thickness

The present

finite-element

and the proposed equation

(h) of their models were equal

15

(a/t =

results are shown

is shown

[9] also used the finite-element

plate

investiga-

as the solid

method,

to the plate

as

curve.

but the width

thickness

(see

dashed and dash-dot lines in the insert).

method, but the

a/c

ratio was 0.98.

Pickard's results were l to 3 percent

higher than the present finite-element

to a width- and length-effect

Kobayashi [lO] used the alternating

results.

Part of the difference

is due

in Pickard's model. And the present results are

expected to be about 1.5 percent below the exact solution.

Near @= 0

and

_/2, Tracey's and Kobayashi's results are 5 to 13 percent higher than the
present results.
mid-point (@ =

All results are in good agreement (within

_/4).

Figures 7 and 8 show the distribution

of boundary-correction

along the crack front for a quarter-circular

(a/c = 0.2) corner

crack, respectively,

The figures show the results

(solid curves)

is generally

sults (symbols), except

surfaces

(@ = 0

and

within

ratios.

subjected

The proposed

of the crack

[19] and this behavior

Semi-elliptical
Two-symmetric

surface

for two-symmetric

hole in a finite plate,

to the finite-element

cracks.-

equation

The maximum

is discussed

Surface

is generally

difference

being about 5

due to a boundary-

in the appendix.

Crack at Hole

The e_npirical stress-intensity

semi-elliptical

surface cracks

Figure 2d, subjected

results presented

is

16

re-

front with the free

Near these points, the equation

results.

Fc,

to tension.

about 2 percent of the finite-element

These low values at the free surfaces are probably

layer effect

equation

a/t

near the intersection

_/2).

factors,

(a/c = l) and semi-elliptical

in a finite plate

for several

higher than the finite-element

percent.

3 percent) at the

factor

at the center of a

to tension was obtained

here (Tables

3a and 3b).

by fitting

The equation

KI = S_

for

0.2 <_ a/c <__2, a/t < l,

-7/2 <_ <_7/2.

thickness.)

0.5 <_ R/t <_2,

(Note that here

The function

Fsh

is defined

(R + c)/b < 0.5,

as one-half

and

of the full plate

was chosen as

Fsh
- "i+ "2_
For

(38)

_' t'
_- Fsh Ia
a t'
R b'
R b'
c I

+ "3 _

9192f_fw

(39)

a/c < l:

Ml : l

M2

(40)

O. 05
O.ll +

M3 =

0.29
0.23 +

(42)

3/2
a
C

1 - O.15k + 3.46k 2 - 4.47k 3 + 3.52k 4

g2:

(44)
1 + O.08k 2

whe re

17

I + # cos

The function

(45)

(0.9

_)

is given by equation

(I0).

The finite-width

correction,

fw'

was taken as

sec

where

n = l

This equation
tration

is for a single

(46)
+ 2nc _/_)11/2
(4 (b-c)
_T(2R+nc)

crack and

was chosen to account

n = 2

is for two-symmetric

for the effects

at the hole [20] and for crack eccentricity

of width
[16].

cracks.

on stress concen-

For

a/c > I:

Ml : _

The functions

(47)

M2' M3' gl' g2'

(45), and the functions

and

and

fw

are given by equations

are given by equation

(41) through

(13) and (46),

respectively.
Estimates
single-surface
present

for a single-surface

The stress-intensity

factors

crack located at the center of a hole can be estimated

results for two-symmetric

developed

crack.-

by Shah [Ill.

surface cracks by using a conversion

The relationship

was given by

18

between

one- and two-surface

for a

from the
factor
cracks

2tR
4
4_.
+ ac
ac.
+
-_
tR

( KI )
=
one
crack

Shah had assumed

that the conversion

two
cracks

factor was constant

along the crack front; that is, independent

Comparison with another
comparison

between

two-symmetric

semi-circular

cracks emanating

solution.-

(a/c = l) and semi-elliptical

from a hole in a plate subjected

Fsh, as a function of the parametric

by Shah.

The proposed

with the results estimated

angle.

equation

by Shah,

angle.

Figure 9 shows

results and those estimated

results (solid symbols) show the distribution

estimated

for all locations

of the parametric

stress-intensitx

the present

(48)

(KI)

by Shah Ill] for

(a/c = 0.2) surface

to tension.

The present

of boundary-correction
The open symbols

(solid curves)

but the equation

factors,

show the results

is in good agreement

is about 5 percent

(based on peak value) than the present

results

elliptic

of the crack front with the free surface

crack.

Near the intersection

(@ = _/2), the present

this reduction
However,

results

is probably

as mentioned

show a sharp reduction.

due to a boundary-layer

in the Append!x,

causes the stress-intensity

factors

point, but lower at the surface.

interior

region

the equation

0 < 2@/_ < 0.8

to be higher

extrapolated

the limiting behavior

refinement

due to mesh refinement.

19

in this region

very near the intersection

factors

by mesh refinement.

region (2/_ < 0.8) only.

to the surface,

mentioned,

(see Appendix).

Also, the stress-intensity

were unaffected

for the semi-

As previously

effect

further mesh

was fitted in the interior

proposed equation,

in the mid-region

higher

is probably

in the
Therefore,

However,

the

a good estimate

for

The influence of crack shape (a/c) on the distribution

correction

factors is shownin Figure lO.

results from Shah [Ill.

results for

of boundary-

The open symbols showthe estimated

And the solid symbols show the present finite-element

a/c = 2. The solid curves show the results from the proposed

equation for a semi-elliptical

surface crack at a hole with

R/t = 0.5.

The

agreements are very good.

Effects

of crack depth-to-plate

bution of boundary-correction
symmetric
ratios.

semi-circular
The proposed

crack front with

surfacr cracks

results

(symbols),

corner cracks.-

for two-symmetric

(R/t = l) with various

is generally

Here, again,

within

a/t

a few percent
of the

the proposed eq-

for the limiting behavior

results

0.2 < a/c < 2,

in Reference

Fch

a/t < I,

The function

Corner Crack
The empirical

quarter-elliptical

KI = S

0 < _ <_/2.

fron for two-

except near the intersection

to give a good estimate

finite plate, Figure 2e, subjected

for

the distri-

due to

in this region.

Two-symmetric

finite-element

at a hole

(solid curves)

Quarter-elliptical

equation

Figure II shows

Fsh, along the crack

the hole surface ( = _/2).

uation is expected
mesh refinement

factors,

equation

of the finite-element

thickness.-

Fch

at a Hole
stress-intensity

corner cracks

factor

at a hole in a

to tension was obtained by fitting to the

14.

The equation

is

(49)

' t' t' b-' b-' @

0.5 < R/t < I,

was chosen

2O

as

(R + c)/b < 0.5,

and

gl

For

g2 g3 f

fw

(50)

a/c < I:

(a)

(51)

0.89
M2 = -0.54

(52)

0.2+

a
C

l
M 3 = 0.5

0.65 + a

+14

1 -_-a )24

(53)

gl = l + [0.

1 +0.35

(54)

0.15_ + 3.46_ 2 - 4.47_ 3 + 3.52_ 4

(55)

g2 =

1 + 0.13_ 2

where

(56)

1 + _cos

The function

g3

(0.85 )

is given by

(57)

21

Functions
For

and

fw

are given by equations

(I0) and (46), respectively.

a/c > l:

(58)

M2 : 0.2

(_)
4

(59)

Icl
4

(60)

M3 = -O. ll

gl = l + IO.l + 0.35

(l - sin @

(61)

Functions

g2

and

are given by equations

(55) and (56).

The function

g3

is given by

cos
11
o-8
o2
I

_=(_
_ oo_c)I_+o_
_
The functions

f@

and

fw

are, again, given by equations

respecti vely.
Estimates

l'I

(62)

(13) and (46),

for a single-corner

crack.-

The strips-intensity

single-corner

crack at a hole can be estimated

two-symmetric

corner

cracks by using the Shah-conversion

Raju and Newman [14] have evaluated

corner-crack-at-a-hole

from the present

factor

the use of the conversion

configurations.

The stress-intensity

22

factors
results

for _' ',

for

(Eq. (_48]).

factor for some

factor obtained

using the conversion

and Kullgren

[12] for a singl-corner

Effects

of plate thickness

distribution

quarter-elliptical

(a/t) on the distribution

results

are shown as symbols

curves.

Again,

the equation

sults, except near

as mentioned
Further

= 0

previously,

mesh refinement

boundary-correction
surface.

with the results

factors,

Figures

Fch, along

corner cracks at a hole.

and the proposed

equation

is in good agreement
7/2.

is causing

with

The effects

is expected

of crack

is shown as the solid

the finite-element

in the Appendix

re-

effect

low values of boundary-correction

factors near the free surface,

for

The finite-element

Here again the boundary-layer

in this region was shown

Thus, the equation

12 and 13 show the

the crack front

are shown in Figure 12.

and

from Smith

crack at a hole.

and crack shape.-

of boundary-correction

two-symmetric
size

factor were in good agreement

[19],

factors.

to give higher

but lower values at the

to give a good estimate

in these

regions.
The effects
correction

factors are shown

(solid curves)
except

near the intersection

the equations

in Figure 13.

points ( = 0

for all combinations

were within

and

established

the proposed

For

because

crack approaching

stress-intensity

a through crack.

variations

available

to be supported

and

by estimates

The effects

a/t <__0.8,
The

intersects

have not been

for comparison.

along the crack front were

23

(symbols),

results.

the crack front

a/t > 0.8, the accuracy of the equations

their use in that range appears

results

investigated

a few percent of the finite-element

there are no solutions

equation

7/2).

of parameters

error was about 5 percent, except where

free surface.

through

Again,

of boundary-

is in good agreement with the finite-element

In summary,

maximum

of crack shape (a/c) on the distribution

However,

based on a part-

of plate width
also included,

on
but were

generally based on engineering estimates, Table 4 gives the range of applicability

of

, a/t,

a/c,

R/t,

and (R + c)/b

24

for the proposed equations.

CONCLUDING
REMARKS
Stress-intensity

factors from three-dimensional finite-element

were used to develop empirical stress-intensity

variety of crack configurations
following

configurations

elliptical

factor equations for a wide

subjected to remote uniform tension.

were included:

an embeddedelliptical

surface crack, a quarter-elliptical

range of configuration
(a/t)

plates.

The

crack, a semi-

corner crack, a semi-elliptical

surface crack at the center of a hole, and a quarter-elliptical

the edge of a hole in finite

analyses

corner crack at

The empirical equations cover a wide

parameters. The ratio of crack" depth to plate thickness

ranged from 0 to l, the ratio of crack depth to crack length (a/c) ranged

0.2 to 2, and the ratio of hole radius to plate thickness (R/t) ranged from
0.5 to 2.

The effects

of plate width (b) on stress-intensity

variations

along

the crack front were also included, but were based on engineering estimates.
For all

configurations

for which ratios of crack depth to plate thickness

do not exceed 0.8, the equations are generally within 5 percent of the finiteelement results,

except where the crack front intersects

the proposed equations give higher stress-intensity

element results,

a free surface.

factors than the finite-

but these higher values probably represent the limiting

havior as the meshis refined near the free surface.

Here

be-

For ratios greater than

0.8, no solutions are available for direct comparison; however, the equations
appear reasonable on the basis of engineering estimates.
The stress-intensity
solutions
gated.
results.

factor equations were also comparedwith other

reported in the literature

for someof the configurations

investi-

The proposed equations were in good agreementwith someof the reported

For limiting

cases, as crack-depth-to-plate

25

thickness (a/t)

or crack-

depth-to-crack length (a/c) approach limits,

the proposed equations reduce to

exact or accepted solutions.

The stress-intensity

factor equations presented herein should be useful

for correlating

and predicting

fatigue-crack-growth

puting fracture

toughness and fracture

rates as well as in com-

loads for these types of crack con-

figurations.

26

APPENDIX
Boundary-Layer Effect on Stress-lntensity

Factors

Hartranft and Sih [19] proposed that the stress-intensity

thin "boundary layer" near the intersection

of the crack with a free surface

drop off rapidly and equal zero at the free surface.

boundary-layer effect,
considered.

a semi-circular

Three different

To investigate

the

surface crack emanating from a hole was

finite-element

models were analyzed with 8, lO, and

14 wedges. A wedgeis a slice of the finite-element

layer of elements [5].

factors in a very

model used to define a

The width of a wedge is measuredby a parametric angle.

Larger numberof wedges result in smaller wedgeangles and more degrees of

freedom. The 8-wedgemodel had eight equal wedges (A =
models

had non-uniform

near the free surface

wedge models were
obtained

wedges

and were obtained

( = 7/2).

_/48

and

respectively.

from the three models are shown in Figure

the stress intensities

near the free surface

They were higher near the free surface

wedge angles.

However,

Further mesh refinements

intensities
proposed

by mesh

angle for the lO- and 14The stress-intensity

14.

These

results

were affected by mesh

distributions

factors

show that

refinement.

in the interior

refinements.

near the free surface

should give higher stress

near the free surface but lower values at the surface.

equation

model

but lower at the surface with smaller

the stress-intensity

(2@/x < 0.8) were unaffected

The other

by refining the 8-wedge

The smallest wedge

_/180,

_/16).

(solid curve) is expected

limiting behavior due to mesh refinement.

27

to give a good estimate

Thus,

the

for the

REFERENCES
['1]

[2]

Irwin,

G. R.:

The Crack

Plate,

ASME, J.

Green,

A. E,;

Phil.

Smith,

W.;

Factors
J.

for

Appl.

Dec.
[4]

of

Cambridge
F.

1967,

A.
to

Wide

[6]

[7]

[8]

also

Newman,

Heliot,

J.;

in

pp.

Part

4,

4,

1962,

pp.
of

an Elastic

Crack

in

651-654.
Stress

in

Solid,

the

Proc.

159-164.
A. S.:

Stress

2 - Semi-lnfinite

Trans.

ASME, Vol.

6,

1979,

D. M.:
an Arbitrary

R.;

Intensity
Solid,

89,

Series

C.,

Jr..

J.,

Aug.
I.
or

a Surface

Second

ASM, 1976,

pp.

Int.

E,

Cracks

Vol.

in

II,

No.

Flawed
Conf.

Plate

on

I073-I077.

Stress-lntensity

Factors

for

Finite-Thickness
4,

1979,

pp.

a
Plates,

817-829.

1977).
S.:

Analyses

Bending

Loads,

Crack,

Int.

of

Surface

Cracks

NASA TP-1578,

and Pellissier-Tanon,
Surface

pp.

in

Proc.

Surface

Tension

Labbens,

No.

Displacement

Materials,

and Raju,

Under

1 -Semi-Elliptical

pp.

1950,

Mechanics

Jr.;

No.

Dec.

Crack

Bending,

Semi-Elliptical

C.,

Plates

Along

of

NASA TM X-72825,

Finite

Tracey,

No.

or Plate

Fracture

J.

The Distribution

and Kobayashi,

and Newman, J.

Engineering

N.:

Crack-Opening

Tension

S.;

29,

Cracks,
34,

No.

a Part-Through

VoI.

47,

F.;

for

953-959.
S.:

Range of

(See

A.

Vol.

Behavior

I.

Vol.

Emery,

Force

Elliptical

Soc.,

pp.

Kobayashi,

Raju,

I.

Semi-Circular

Mechanical

[5]

a Flat

Mechs.,

Subjected

Mechs.,

and Sneddon,

Neighborhood

[3]

Appl.

Extension

A.:
J.

of

Dec.

Benchmark
Fracture,

in
1979.

Problem
Vol.

15,

R197-R202.

3D Elastic
Crack

Singularity
Front,

Int.

340-343.

28

Element
J.

of

for

Fracture,

Evaluation

of

Vol.

1973,

9,

[9]

Pickard, A. C.:
and Elliptic

[1o]

Stress

Intensity

Crack Fronts-Determined

PNR-90035,

Rolls-Royce

Kobayashi,

A. S.; and Enetanya,

[11]

Stress

Cracks Originating
STP-590, American

[12]

[13]

A. N.:

Intensity

Society

Factors

for Testing
T. E.:

Hole Using

of Testing

pp. 429-459.

and Experimental

Feb. 1977.

of Stress

Stress-lntensity

Intensity
Singularity

Factors

ASTM STP-677,

and Materials,

J. C., Jr.:

Analysis of Finite-Thickness

1976,

ASTM

1977.

Fracture Mechanics,

Raju, I. S.; and Newman,

Society

Growth,

the Isoparametric

Symmetric

Society

of Crack

Laboratory,

J. C., Jr.:

Ed., American

American

from Fastener Holes,

Raju, I. S.; and Newman,

Cracks,

Factor of a

and Part-Through

and Materials,

Determination

Element, Int. J. of Fracture, Oct.

[15]

for Through

Air Force Flight Dynamics

Heckmer, J. L.; and Bloom, J. M.:

Corner

Intensity

ASTM STP-590,

Theoretical

Cracks Emanating

Factors for the Corner-Cracked

[14]

Stress

at Fastener Holes, Mechanics

Analysis of Surface

Methods,

1976, pp. 477-495.

Smith, F. W.; and Kullgren,

AFFDL-TR-76-104,

by 3D Finite Element

of Crack Growth,

and Materials,

Shah, R. C.:

Circular

Limited, May 1980.

Corner Crack, Mechanics

for Testing

Factors for Cracks with

C. W. Smith,

1979, pp. 411-430.

Three-Dimensional

Fracture Specimens,

for Two

Finite-Element

NASA TN D-8414,

May 1977.
[16]

Tada, H.; Paris,

Handbook,

[17]

P. C.; and Irwin, G. R.:

Del Research

Gross, B.; and Srawley,

Edge-Notch
Boundary

Specimens

Collocation

Corporation,
J. E.:

The Stress

in Bending or Combined

29

of Cracks

1973.

Stress-lntensity

of a Stress

Analysis

Factors

Bending

for Single-

and Tension

Function, NASA TN D-2603,

1965.

by

[18]

Newman, J. C., Jr.; and Raju, I. S.:

Factor Equation
Journal,

[19]

for the Surface

Crack, Engineering

Hartranft,

R. J.; and Sih, G. C.:

Journal of Engineering
Howland,

Stress-lntensity
Fracture

Mechanics

1981.

Theory of Plates with Application

[20]

An Empirical

R. C. J.:

Three-Dimensional

to Crack Problems,

International

Science, Vol. 8, No. 8, 1970, pp. 711-729.

On the Stresses

Hole in a Strip Under Tension,

Vol. 229, Jan.

An Approximate

in the Neighbourhood

of a Circular

Philos. Trans. R. Soc. London,

1930, pp.49-86.

3O

Series

A,

TABLEl--Boundary correction factors,

F, for embedded
elliptical
crack in a plate subjected to tension.
(c/b _ 0.2; h/b = l; _ = 0.3)
a/t
a/c

2/_

0.2

0.5

0.8

0.2

0
0.125
0.25
0.375
0.5
0.625
0.75
0.875
l.O

0.450
0.531
0.643
0.750
0.838
0.905
0.951
0.978
0.987

0.473
0.556
0.678
0.794
0.893
0.978
1.042
1.083
1.097

0.514
0.605
0.745
0.884
l.Ol5
1.176
1.329
1.438
1.480

0.4

0
0.125
0.25
0.375
0.5
0.625
0.75
0.875
l.O

0.632
0.656
0.715
0.789
0.857
0.914
0.954
0.978
0.987

0.660
0.685
0.748
0.826
0.900
0.964
l.Ol4
].046
1.056

0.721
0.749
0.821
0.905
0.995
1.105
1.211
1.285
1.312

l.O

0
0.125
0.25
0.375
0.5
0.625
0.75
0.875
l.O

0.986
0.986
0.986
0.986
0.986
0.986
0.986
0.986
0.986

1.009
].009
1.008
1.006
1.006
1.008
l.OlO
l.Ol2
l.Ol3

1.060
1.058
1.050
1.035
1.036
1.059
1.093
l.ll4
1.121

2.0

0
0.125
0.25
0.375
0.5
0.625
0.75
0.875
l.O

0.709
0.703
0.686
0.658
0.622
0.579
0.536
0.503
0.490

0.713
0.707
0.690
0.662
0.625
0.582
0.539
0.506
0.494

0.720
0.714
0.697
0.669
0.633
0.592
0.552
0.522
O.511

31

TABLE 2--Boundary correction factors,

F, for corner
crack in a plate subjected to tension.
(c/b _ 0.2_

h/b = l;

_ = 0.3)
a/t

a/c

2/_

0.2

0.5

0.8

0.2

0
0.125
0.25
0.375
0.5
0.625
0.75
0.875
l.O

0.555
0.633
0.753
0.871
0.973
1.055
l.ll5
1.159
1.156

0.761
0.840
0.988
1.141
1.277
1.397
1.495
1.580
1.610

1.288
1.340
1.522
1.705
1.850
2.008
2.118
2.263
2.450

0.4

0
0.125
0.25
0.375
0.5
0.625
0.75
0.875
l.O

0.791
0.774
0.824
0.893
0.964
1.026
1.075
l.ll7
1.132

0.990
0.952
0.997
1.067
1.140
1.210
1.273
1.334
1.365

1.397
1.297
1.310
1.346
1.384
1.458
1.528
1.627
1.788

l.O

0
0.125
0.25
O. 375
0.5
0.625
0.75
0.875
l.O

l.162
l.Ill
l 079
l 064
l 059
l 063
l 078
l I09
l 159

l 275
l 207
l 160
l 134
l 121
l 123
l.140
l.176
1.233

l. 487
I. 378
l .290
l.219
l.180
l.191
l .231
l. 301
l .416

2.0

0
0.125
0.25
0.375
0.5
0.625
0.75
0.875
l.O

0.800
0.787
0.756
0.722
0.683
0.640
0.600
0.579
0.586

O. 826
O.811
0.776
0.738
O. 697
0.653
0.612
O. 590
0.597

0.862
0.837
0.793
0.750
O. 704
O. 660
0.624
O.611
0.625

32

TABLE 3--Boundary correction factors,

F, for surface crack
at center of hole in a plate subjected to tension.
((R + c)/b _ 0.2;

h/b > 1.6;

(a)

v = 0.3)

R/t = l
a/t

a/c

2@/_

0.2

0.5

0.8

0.2

0
0.125
0.25
0.375
0.5
0.625
0.75
0.833
0.917
0.958
l.O

0.641
0.692
0.836
l.Oll
1.196
1.405
1.651
1.905
2.179
2.288
1.834

0.607
0.662
0.775
0.905
1.032
1.178
1.362
1.583
1.885
2.121
1.958

0.593
0.643
0.771
0.919
1.094
1.293
1.528
1.765
2.050
2.336
2.329

0.4

0
0.125
0.25
0.375
0.5
0.625
0.75
0.833
0.917
0.958
l.O

1.030
1.076
1.202
1.376
1.578
1.804
2.040
2.238
2.396
2.376
1.844

0.872
0.912
1.007
1.131
1.275
1.452
1.667
1.891
2.141
2.255
1.923

0.840
0.872
0.959
1.074
1.234
1.426
1.668
1.914
2.201
2.411
2.224

l.O

0
0.125
0.25
0.375
0.5
0.625
0.75
0.833
0.917
0.958
l.O

2.267
2.276
2.301
2.343
2.404
2.481
2.566
2.620
2.622
2.468
1.950

1.806
1.818
1.851
1.905
1.980
2.079
2.206
2.321
2.415
2.370
1.957

1.615
1.619
1.630
1.646
1.730
1.852
2.049
2.250
2.452
2.512
2.203

2.0

0
0.125
0.25
0.375
0.5
0.625
0.75
0.833
0.917
0.958
l.O

1.944
1.931
1.897
1.840
1.763
1.669
1.580
1.498
1.426
1.313
1.042

1.606
1.600
1.582
1.553
1.514
1.468
1.434
1.404
1.387
1.321
1.082

1.394
1.389
1.377
1.357
1.333
1.313
1.310
l .313
1.332
1.294
1.077

33

TABLE3--Boundary correction factors,

F, for surface crack
at center of hole in a plate subjected to tension.
((R + c)/b _ 0.2;

h/b > 1.6;

(b)

= 0.3)

R/t = 2
a/t

a/c

2/_

0.2

0.5

0.8

0.2

0
0.125
0.25
0.375
0.5
0.625
0.75
0.833
0.917
0.958
l.O

0.800
0.864
1.046
1.272
1.508
1.766
2.041
2.279
2.474
2.439
1.791

0.680
0.743
0.877
1.037
1.206
1.410
1.662
1.932
2.238
2.375
1.947

0.634
0.690
0.832
1.002
1.213
1.469
1.787
2.109
2.463
2.699
2.380

0.4

0
0.125
0.25
0.375
0.5
0.625
0.75
0.833
0.917
0.958
l.O

1,290
1.346
1.498
1.704
1.932
2.165
2.378
2.516
2.564
2.417
1.776

1.058
1.107
1.227
1.384
1.568
1.785
2.026
2.237
2.418
2.416
1.894

0.972
l.OlO
l.ll8
1.263
1.470
1.722
2.031
2.319
2.595
2.705
2.258

l.O

0
0.125
0.25
0.375
0.5
0.625
0.75
0.833
0.917
0.958
l.O

2.620
2.626
2.642
2.667
2.700
2.732
2.753
2.733
2.643
2.409
1.862

2.188
2.199
2.232
2.280
2.341
2.410
2.483
2.527
2.521
2.381
1.888

1.990
1.996
2.009
2.026
2.121
2.246
2.437
2.599
2.716
2.662
2.192

2.0

0
0.125
0.25
0.375
0.5
0.625
0.75
0.833
0.917
0.958
l.O

2.136
2.121
2.075
2.000
1.899
1.777
1.659
1.552
1.456
1.325
1.041

1.922
1.911
1.879
1.826
1.756
1.671
1.593
1.522
1.463
1.360
1.088

1.712
1.704
1.681
1.643
1.594
1.541
1.499
1.461
1.434
1.351
1.089

34

A
v

N
+
0
V

0
V

0
V

0
V

0
V

c"
0
.f-4-a

o-

.
o
4.a
t)
0
4-

OU

CU

U
0

>_
4.a

e-

e-i
v

4J

Io
4-

I
I

.p.
.r-

u
r"-

e0
r"
0

CO
v

'UD
,-v

_0
c_
v

CO
CO
v

0'_
q:T
I/)
v

LIJ

0
e_
I
I

"j
.I-.}
,--

--J

%-

0
r-

_-.
.,-.

I"-

(i}
_
_
,--"
CL
c-"
-,-"
v
U

....-.,

v
a._
.I-.}
_
,-_

_._
,--0
"I-

"Z

v
,-0
"7"
I

_
._
{,3

p.tJ

cO

-_

35

II

2h

FIG. l-Corner cracks at the edge of a hole in a finite plate

subjected to remote tension.

36

I"---- _ ------

I'-------'--'----

.:.:.....;.;_.;._....'...._.'_ ;: -_:...<;.

_+:.!-;"_
_._...-'..:..i
_:i _.:._..:-i.:._;',.
1
_,_
:-.-_.:.......
".:.:i".:;:._-;\
[

3!-

I
i

i u
_

.[:_':!

,
_
',"-'

;.!:::.::_,:;:_

..,
; ;:'"",.':i.f!".::_-'!:i'-'_!.:.'.
':-'-'.-:.6.,--'-"

'-"
_

.....

" '

:."i:'..';:',:"_'.'.'.:""..,
_ !:'_!""":_'_'";':!:"::':':"
_:_]!:_'
u

.c:::
fo
4J
so t-

,,j

__

o
L

or,,"1_

op-

e-4__
o_=

,,_

e-

a,_ r--

N _

,-,

!::

-ca

:"i:_:_"::::.'_':!:!:

.%i-.--- "i':_!'-_:__

..-'$'.:
:.?iY.'_:'.-"
'

_o_-

37

T
r..--

,-'t

e-

J
.=.

(..3

g.
4->
D
E

C3
A

r_

L_"-_

:r

_J
ep..

4_
er--

r--i

VI

o
o

(.3

C_

O
v

38

_A
I,

"o
"t3
"o
,,Q
E
cf,_
.Ca
,IJ
.to
L

C3

"

'

v,m_,l

u
m_
T.I_:.
i.: :_i " .!
)- .,.-...-. : T._,._ .:

t_
o

.
c_
c
o

I.-_ .Jo
u_
., QJ (.,)
L
o_'-

'o

"r"

oP,-

m
C::)

39

u
j,--

,---I
+r.i-,..-

a,i
I
r-

S.o
_i

:+J..++l--

-i

c
o
4_J
ro
L
u
to

o
u
%LI.J
r-

C_

I... II
L
0.._
U_

tI" |1
0

O0

U
tO

0
e-

I
0

Ii

4O

r-

5..

r-

r--

"_

m_
c

u
L

o_

<_E

'

..c:

U
L

-g

II

'-'

,
'-

i:i_-:.-:!:::.
...-..:_ i:-::::;J_ ,4-,
:/:.::!::-::.!:.....::;i!_iii_

g..z
t!

1.1

N_
_"

I|

/:_I,__

_g
O

//

!
C)
,--I

,--'t

,A
LI-

4]

,--4

I
I
.Q

C)

cJ
U_

42

1,1

IJJ

!
C)

0
LL

43

eLL

44

u_

sU

o_

_
.

II

r_
_--

II

LY_
r-I

E
Q._ u
$o_-

_0

o
_"

u
c

t_

3_
'- 4-_
o"_
I

o_

.- r_

If-

t-/)
IJ-

45

,-"4

-I_1-

I.
cO
O0

c-

oo

r!

!
0

LL

46

&.

U
!
_J
4_
&.

O"
&0

.Q

C
O
&.

u v
_..-

Y
o_

,3

!o

L.-J

L
0

!!

Om

(t)
U
L
L4-_

(/)

q.)
co

t
4_

r-(23
E
q3

e-

e--

(1)
!
(13
4_
o_=_

0
0

m
r-

1,I_

c
_-=4

U_

!
pr_

!
c3

o
IJ_

,A
U-

47

I
r_

u_

48

-3
"O
O
F::

4-,
(--

C)

"O
O
E

"O
O
F::

"(C}
_.}

"O
"I13
(3..) (13

CO
4.-,

O
,--I

(--

Qrn4]

_,,,I

OO

.._
,--I

LL

-4ol--

T
z',,..(',4

I
I_

C_

r--I

t'-It_

49

1.

Report

No.

2.

Government

Accession

No.

3.

Recipient's

5.

Report

Catalog

No.

NASA TM-83200
4.

Title

and

Subtitle

Date

August

STRESS-INTENSITY FACTOR EQUATIONS FOR CRACKS IN

THREE-DIMENSIONAL FINITE BODIES

6.

1981

PeHorming

Or_nization

Code

505-33-23-02
7,

Author(s)

J.
9

8.

C. Newman, Jr.

Performing

Organization

NASA Langley
Hampton, VA
12

S_nsoring

Agency

Name

and I.
and

and

Supplementary

Work

No.

11.

Contract

13.

Type

No.

Unit

Center

Address

or

of

Grant

Report

Technical
Administration

14.

Sponsoring

No.

and

Period

Covered

Memorandum

Agency

Code

Notes

Presented
California,
16

10.

Report

S. Raju

National
Aeronautics
and Space
Washington,
DC 20546
15

Organzzation

Addre_

Research
23665
Name

Performing

at

the ASTM 14th

June 30 - July

National
2, 1981.

Symposium

on Fracture

Mechanics,

Los Angeles,

Abstract

This paper presents

empirical
stress-intensity
factor
equations
for embedded
elliptical
cracks,
semi-elliptical
surface
cracks,
quarter-elliptical
corner
cracks,
semi-elliptical
surface
cracks
at a hole,
and quarter-elliptical
corner
cracks
at a hole in finite
plates.
The plates
were subjected
to remote tensile
loading.
These equations
give stress-intensity
factors
as a function
of parametric
angle,
crack depth,
crack length,
plate
thickness,
and, where applicable,
hole radius.
The stress-intensity
factors
used to develop
the equations
were
obtained
from current
and previous
three-dimensional
finite-element
analyses
of
these crack configurations.
A wide range of configuration
parameters
was
included
in the equations.
The ratio
of crack depth to plate
thickness
ranged
from 0 to I, the ratio
of crack depth to crack length
ranged from 0.2 to 2, and
the ratio
of hole radius
to plate
thickness
ranged from 0.5 to 2.
The effects
of plate
width
on stress-intensity
variations
along the crack front
were also
included,
but were generally
based on engineering
estimates.
For all
combinations
of parameters
investigated,
the empirical
equations
were generally
within
5 percent
of the finite-element
results,
except within
a thin
"boundary
layer"
where the crack front
intersects
a free surface.
However,
the proposed
equations
are expected
to give a good estimate
in this
region
because of a study made on
the boundary-layer
effect.
These equations
should
be useful
for
crack-growth
rates
as well
as in computing
for these types of crack configurations.
17.

Key

Words

(Suggested

by

Author(s))

correlating
fracture

and predicting
fatiguetoughness
and fracture
loads

18.

Statement

Cracks
Stress
analysis
Fatigue
(materials)
Fracture
Stress-intensity
factor
Finite-element
method
19.

S_urity

Cla_if.

Unclassified

(of

this

report)

Distribution

Unclassified

- Unl'imited

Subject
20.

_curity

Classif.

Unclassified

(of

this

_ge)

21.

No.

of

Pages

50

Price"

A03

22.

" For sale by the NationalTechnical InformationService,Springfield,Virginia 22161

Category

39