Composites: Part B 68 (2015) 51–58

Contents lists available at ScienceDirect

Composites: Part B
journal homepage: www.elsevier.com/locate/compositesb

An engineering formula for the stress concentration factor of orthotropic

composite plates
Michele Zappalorto ⇑, Paolo Andrea Carraro
University of Padova, Department of Management and Engineering, Stradella San Nicola 3, 36100 Vicenza, Italy

a r t i c l e i n f o a b s t r a c t

Article history: An engineering formula for the theoretical stress concentration factor of orthotropic notched plates under
Received 19 June 2014 tension is provided, as a function of the material elastic constants and the Kt of the corresponding isotro-
Received in revised form 31 July 2014 pic case. The accuracy and limits of applicability of the new solution are discussed by comparison to data
Accepted 4 August 2014
from the literature and results from FE analyses on notched geometries of practical interests. The pro-
Available online 11 August 2014
posed solution represents a very useful tool to estimate the stress concentration factor of notched ortho-
tropic plates, composite orthotropic laminae, orthotropic unidirectional laminates and homogenised
Keywords:
orthotropic composite laminates.
B. Stress concentrations
A. Laminates
Ó 2014 Elsevier Ltd. All rights reserved.
C. Analytical modelling

1. Introduction The design against fatigue or brittle failure of blunt notched

Geometrical variations, such as notches, grooves or holes,
unavoidably exist in engineering components, being often respon-
sible for crack formation under static and cyclic conditions.
This is the reason why the study of stress distributions around
notches has received a large attention in the past and recent
literature. Blunt cracks in isotropic plates have been analysed by
Creager and Paris [1] who were able to give the closed form
solution for the local stress fields for Mode I, II and III loadings.
The above mentioned solution was later extended to blunt
V-notches, including the effect of the notch opening angle, both
under linear elastic [2–5] and elastic plastic conditions [6,7]. It is
also worth mentioning that a large number of solutions for central
holes in isotropic plates is present in the book by Savin [8].
Moving to orthotropic plates, comparatively few works can be
found in the literature, mainly oriented to refine the classical anal-
ysis for anisotropic plates with elliptical holes by Lekhnitskii [9]
and to make it applicable to composite laminates (see, among
the others, Bonora et al., [10,11], Chern and Tuttle [12]). More
recently, Ukadgaonker and Rao [13] carried out an analytical study
of the stress distributions in an orthotropic plate with triangular
holes, while the case of an irregular shaped hole has been later con-
sidered by Ukadgaonker and Rao [14] and by Ukadgaonker and
Kakhandki [15], where an excellent literature review on the topic
can be found, as well.
⇑ Corresponding author.
E-mail address: michele.zappalorto@unipd.it (M. Zappalorto).
engineering components is often based on strength criteria,
according to which the stress values in the stress concentration
regions are compared to the fatigue or static strength of the base
material, respectively. This process is complicated by the fact that
the stress state close to a notch is inherently multiaxial and, under
such a stress state, the fatigue behaviour of composite materials
might be very complex [16–21].
In the engineering practice, the maximum stress at a blunt
notch root is correlated to the nominal stress using the theoretical
stress concentration factor. However, different from isotropic
materials, in orthotropic plates Kt depends not only on the geome-
try but also on the elastic material properties [8,9,22–24]. This
hampers the possibility for engineers to use design charts or
approximated relationships, as those provided in Refs. [25–30],
allowing a rapid evaluation of Kt.
Although advanced computational technology has made it pos-
sible to calculate the stress concentration factor for any notched
geometry and material, practical expressions, providing a rapid
evaluation of this parameter, remain very useful in the engineering
practice. To this end, starting from some analytical derivations
based on classical solutions of the orthotropic theory of elasticity
[9], an engineering formula to estimate the theoretical stress con-
centration factor of orthotropic notched plates under tension is
provided, involving the material elastic properties and the Kt of
the corresponding isotropic case, which is geometry-dependent,
only.
Accuracy and limits of applicability are discussed by comparing
the approximate solution to numerical results from the literature

http://dx.doi.org/10.1016/j.compositesb.2014.08.020
1359-8368/Ó 2014 Elsevier Ltd. All rights reserved.
52 M. Zappalorto, P.A. Carraro / Composites: Part B 68 (2015) 51–58

and from FE analyses carried out by the authors, considering In Eq. (5) Tij equate the terms of the compliance matrix, Sij, for
notched plates of practical interest. plane stress. In this case, invoking the engineering elastic
The proposed solution represents a useful tool to estimate the constants:
stress concentration factor of notched orthotropic plates, compos-
1 1 m21 m12 1
ite orthotropic laminae, orthotropic unidirectional laminates and S11 ¼ S22 ¼ S12 ¼ ¼ S66 ¼ ð6Þ
E1 E2 E2 E1 G12
homogenised orthotropic composite laminates.
it results:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
2. Material behaviour u sffiffiffiffiffi
u E1 E1
t
f ¼ b1 þ b2 ¼ 2  2m12 þ ð7Þ
Although the notch tip stress state in a thick plate is always, by E2 G12
very nature, three-dimensional, the adoption of two dimensional
hypotheses, such as plane stress or plane strain, allows to remove For plane strain conditions instead, Tij = Bij. It is worth noting that
many difficulties encountered in the three-dimensional anisotropic for an isotropic material b1 = b2 = 1, so that f = 2.
elasticity theory. Moreover under particular conditions, plane Invoking the theoretical stress concentration factor for the cor-
hypotheses can be representative of the actual three-dimensional responding isotropic case [33,34]:
rffiffiffiffi
behaviour [31,32]. Accordingly, in the present work notched ortho- a
e tg ¼ 1 þ 2
K ð8Þ
tropic plates under plane stress or plane strain conditions are con- q
sidered only.
For plane stress problems (r3 = s13 = s23 = 0) the elastic ortho- Eq. (4) can be conveniently re-written in the following form:
tropic stress–strain relationships can be formulated on the basis f e
of four independent elastic constants: K tg ¼ 1 þ ð K tg  1Þ ð9Þ
2
8 9 2 38 9
>
< e1 >
= S11 S12 0 > < r1 >= One should note that in the case of a wide orthotropic composite
e2 ¼ 64 S12 S22
7
0 5 r2 ð1Þ e tg ¼ 3 and Eq. (9) gives:
plate with a central hole K
>
: > > > vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
c12 ; 0 0 S66
:
s12 ; u sffiffiffiffiffi
u E E
K t ¼ 1 þ t2
1 1
For plane strain problems (e3 = c13 = c23 = 0), instead, Hooke’s law in  2m12 þ ð10Þ
E2 G12
terms of compliance matrix would read as:
8 9 2 38 9 in agreement with the expression reported in Ref. [35] and widely
>
< e1 >= B11 B12 0 > < r1 >= used for composite materials.
e2 ¼ 6 7
4 B12 B22 0 5 r2 ð2Þ
>
: > > >
c12 ; 0 0 B66
:
s12 ; 3.2. The edge notch in a semi-infinite plate under tension

where the constants Bij can be expressed in terms of the complianc- With reference to edge notches in semi-infinite orthotropic
es Sij: plates under tension Chiang [36] proved that the theoretical stress
S11 S33  S213 S12 S33  S13 S23 concentration factor can be written as:
B11 ¼ B12 ¼ Z 0
S33 S33 f ðnÞ
ð3Þ K tg ¼ 1 þ f dn ð11Þ
S22 S33  S223 n
B22 ¼ B66 ¼ S66
S33 where f is the function describing the notch boundary. Accordingly,
Accordingly, from the mathematical point of view, the plane stress the function f0 (n)/n does not depend on the material elastic proper-
and plane strain problems are identical except for the values of elas- ties, so that Eq. (9) holds valid also for semi-infinite plates with
tic constants entering into the reduced strain–stress relations. shallow edge notches of any shape.

3.3. The deep hyperbolic notch

3. Analytical remarks
The stress concentration factor, referred to the net section, for a
3.1. The elliptical hole in an infinite plate under tension
tensioned orthotropic plate weakened by two symmetric deep
hyperbolic notches is [9]:
Consider an elliptical hole in an infinite orthotropic plate under
sffiffiffiffi
tension, the direction of the far applied uniaxial tension equating h b2  b22
the 1 principal elasticity direction. The stress concentration factor K tn ¼  qffiffiffi1  qffiffiffi ð12Þ
q b ArcTan b h  b ArcTan b h
for this problem, referred to the gross section, is [9]: 1 1 q 2 2 q
rffiffiffiffi
a where 2 h is the cross-sectional width and q is the notch root
K tg ¼ 1 þ ðb1 þ b2 Þ ð4Þ
q radius.  qffiffiffi
When the notch is sharp, (q/h ? 0), ArcTan bi qh tends
where a is the notch depth (major semi-axis of the ellipse), q is the towards p/2, and Eq. (12) can be approximated by:
notch root radius and u1;3 ¼ ib1 and u2;4 ¼ ib2 are the conjugate sffiffiffiffi
roots of the following equation [8,9]: 2 h
K tn ffi ðb1 þ b2 Þ ð13Þ
T 11 l4 þ ð2T 12 þ T 66 Þl2 þ T 22 ¼ 0 ð5Þ p q

Eq. (5) is the characteristic equation linked to the governing equa- Under the same conditions, the stress concentration factor for the
tion of the plane orthotropic theory of elasticity to be satisfied by corresponding isotropic case is [34]:
sffiffiffiffi
the Airy stress function. Since li always occur in conjugate pairs,
it is possible to arrange, without loss of generality, that b1 and b2 e tn ¼ 4
K
h
ð14Þ
are real and positive [9].
p q
 M. Zappalorto, P.A. Carraro / Composites: Part B 68 (2015) 51–58 53

σ 3.4. Further remarks

In the previous subsections it has been proven that, under cer-

tain conditions the following equation:
f e
K t ¼ 1 þ ðK t  1Þ ð16Þ
1 2
2b represents a very useful tool to evaluate the theoretical stress con-
centration factor for a notched orthotropic plate under tension as a
function of the corresponding value for the isotropic case and
parameter f which, in turn, depends on the elastic constants of
the material.
2 However, by a strictly theoretical point of view, Eq. (16) is valid
Material reference a only for:
system
W 1. Shallow notches, either lateral or central, in infinite or semi-
infinite plates under tension.
2. Sharp deep lateral notches in plates under tension.

In the following section, instead, Eq. (16) is applied to finite size

σ plates and components of practical interest, with the aim to discuss
its accuracy and range of applicability. Several geometries and
Fig. 1. Orthotropic finite-width plates, containing elliptical edge notches, analysed
by Weixing and Xinlu [37]. materials are considered.

Accordingly, Eq. (13) can be re-written as:

4. Validation by comparison to solutions from the literature
ðb1 þ b2 Þ e
K tn ffi K tn ð15Þ In this section Eq. (16) is compared to some results taken from
2
the literature. Several notched geometries and materials are
which approximately equates Eq. (9), the notch being sharp. considered.

Table 1
e tg , as reported in Ref. [37].
Finite-width plates with elliptical notches. Stress concentration factor for isotropic plates, K

W/(2a)
20 10 5 3.33 2.5 2 1.33
a/b
0.4 1.80685 1.79795 1.79009 1.82442 1.94352 2.19647 2.65195
1 3.05067 3.04176 3.02661 3.03876 3.09993 3.24327 3.54922
1.25 3.57448 3.56479 3.5477 3.55855 3.62701 3.76821 4.06957
2 5.14838 5.1467 5.11717 5.14699 5.2448 5.44393 5.80576
4 9.37817 9.57176 9.61065 9.68197 10.2075 10.7547 11.3404

Table 2
Finite-width plates with elliptical notches. Comparison between (a) results from Ref. [37] and (b) results by Eq. (16). 3.33 < W/2a < 20. Relative deviation: D ¼ K ðbÞ ðaÞ
tg =K tg  1.

W/(2a)
20 10 5 3.33
a/b Ktg(a) Ktg(b) D (%) Ktg(a) Ktg(b) D (%) Ktg(a) Ktg(b) D (%) Ktg(a) Ktg(b) D (%)
Material A1 0.4 1.839 1.832 0.37 1.828 1.823 0.30 1.825 1.815 0.57 1.876 1.850 1.39
1 3.125 3.114 0.34 3.114 3.105 0.29 3.100 3.090 0.33 3.122 3.102 0.64
1.25 3.666 3.654 0.32 3.656 3.644 0.31 3.639 3.627 0.34 3.662 3.638 0.67
2 5.308 5.277 0.57 5.325 5.275 0.92 5.300 5.245 1.04 5.337 5.276 1.14
4 9.779 9.638 1.43 10.036 9.838 1.97 10.205 9.878 3.20 10.273 9.952 3.13
Material A2 0.4 2.049 2.050 0.06 2.039 2.038 0.03 2.031 2.028 0.15 2.068 2.073 0.24
1 3.662 3.668 0.15 3.653 3.657 0.10 3.637 3.637 0.01 3.660 3.653 0.20
1.25 4.341 4.350 0.20 4.336 4.337 0.01 4.320 4.315 0.13 4.347 4.329 0.41
2 6.400 6.397 0.05 6.483 6.395 1.36 6.476 6.357 1.84 6.523 6.396 1.96
4 11.712 11.901 1.61 12.263 12.153 0.90 12.743 12.203 4.24 12.845 12.296 4.27
Material A3 0.4 1.562 1.564 0.10 1.551 1.557 0.42 1.552 1.552 0.01 1.623 1.576 2.89
1 2.423 2.433 0.39 2.415 2.426 0.49 2.402 2.416 0.58 2.418 2.424 0.25
1.25 2.785 2.799 0.47 2.776 2.792 0.57 2.761 2.780 0.70 2.774 2.787 0.47
2 3.881 3.898 0.44 3.867 3.897 0.77 3.839 3.876 0.96 3.849 3.897 1.25
4 6.861 6.853 0.12 6.865 6.988 1.80 6.760 7.015 3.77 6.799 7.065 3.91
Material A4 0.4 2.248 2.242 0.24 2.239 2.228 0.48 2.226 2.216 0.45 2.246 2.269 1.03
1 4.180 4.157 0.55 4.174 4.143 0.74 4.157 4.120 0.90 4.178 4.138 0.94
1.25 4.986 4.963 0.45 4.993 4.948 0.89 4.976 4.922 1.08 5.003 4.939 1.28
2 7.425 7.386 0.53 7.579 7.383 2.59 7.594 7.338 3.37 7.645 7.384 3.41
4 13.269 13.897 4.73 14.149 14.195 0.32 14.924 14.255 4.48 15.060 14.365 4.61
54 M. Zappalorto, P.A. Carraro / Composites: Part B 68 (2015) 51–58

Table 3
Finite-width plates with elliptical notches. Comparison between (a) results from Ref. [37] and (b) results by Eq. (16). 1.667 < W/2a < 2.5. Relative deviation: D ¼ K ðbÞ ðaÞ
tg =K tg  1.

W/(2a)
2.5 2 1.667
a/b Ktg(a) Ktg(b) D (%) Ktg(a) Ktg(b) D (%) Ktg(a) Ktg(b) D (%)
Material A1 0.4 2.018 1.973 2.25 2.284 2.234 2.21 2.734 2.703 1.11
1 3.203 3.165 1.17 3.384 3.313 2.09 3.742 3.628 3.05
1.25 3.746 3.709 0.99 3.927 3.854 1.84 4.280 4.165 2.69
2 5.463 5.377 1.58 5.686 5.582 1.83 6.107 5.955 2.49
4 10.810 10.493 2.93 11.271 11.058 1.90 12.130 11.661 3.86
Material A2 0.4 2.185 2.228 1.96 2.423 2.557 5.52 2.850 3.149 10.51
1 3.741 3.732 0.24 3.920 3.919 0.02 4.270 4.317 1.10
1.25 4.441 4.418 0.52 4.624 4.602 0.49 4.995 4.994 0.02
2 6.712 6.523 2.81 6.967 6.782 2.65 7.450 7.253 2.65
4 13.582 12.980 4.43 14.278 13.692 4.10 15.153 14.454 4.61
Material A3 0.4 1.799 1.659 7.79 2.101 1.836 12.64 2.587 2.154 16.75
1 2.490 2.467 0.93 2.667 2.567 3.76 3.042 2.781 8.58
1.25 2.841 2.835 0.20 3.000 2.934 2.21 3.335 3.144 5.71
2 3.921 3.965 1.13 4.092 4.105 0.31 4.425 4.357 1.52
4 7.052 7.432 5.39 7.389 7.815 5.76 7.898 8.224 4.12
Material A4 0.4 2.328 2.452 5.34 2.521 2.842 12.73 2.913 3.543 21.65
1 4.250 4.233 0.41 4.420 4.453 0.75 4.740 4.924 3.89
1.25 5.098 5.044 1.06 5.281 5.261 0.37 5.637 5.725 1.57
2 7.863 7.534 4.17 8.167 7.841 3.99 8.665 8.398 3.08
4 15.971 15.174 4.99 16.958 16.017 5.55 17.634 16.918 4.06

σ Table 4
Finite-width plates with circular holes. Comparison between (a) results from Ref. [38]
and (b) results by Eq. (16). Relative deviation: D ¼ K ðbÞ ðaÞ
tg =K tg  1.

W/(2a) e tg [15]
K Material B1 Material B2
(a) (b)
Ktg Ktg D (%) Ktg(a) Ktg(b) D (%)
10 3.035 8.990 9.135 1.62 7.530 7.643 1.50
1 2a 5 3.149 9.110 9.589 5.25 7.670 8.013 4.48
4 3.243 9.240 9.966 7.86 7.790 8.321 6.82
3 3.460 9.550 10.831 13.41 8.080 9.028 11.73
2 4.314 10.710 14.246 33.01 9.150 11.816 29.14
W/(2a) e tg [15]
K Material B3 Material B4
(a) (b)
Ktg Ktg D (%) Ktg(a) Ktg(b) D (%)
2
Material reference 10 3.149 6.270 6.515 3.91 4.800 4.920 2.49
5 3.243 6.380 6.757 5.91 4.900 5.092 3.91
system e 3 3.460 6.650 7.313 9.96 5.160 5.487 6.33
2.5 3.733 6.950 8.013 15.30 5.450 5.984 9.81
2 4.314 7.640 9.505 24.42 6.080 7.045 15.87
W

2. Material A2: E1 = 210 GPa, E2 = 132 GPa, G12 = 44.8 GPa,

σ laminate.
Fig. 2. Orthotropic finite-width plates, containing eccentric circular holes, analysed
by Wang and Song [38].
4.1. Comparison to the results by Weixing and Xinlu [37]: orthotropic
finite-width plates containing elliptical edge notches
Weixing and Xinlu [37] studied the stress concentration factor
of an orthotropic finite-width plate containing elliptical edge
notches using the finite element method. The geometry considered
by these authors is shown in Fig. 1.
Four different material configurations were analysed in Ref. [37]
(plane stress):
1. Material A1: E1 = 132 GPa, E2 = 210 GPa, G12 = 44.8 GPa,
m12 = 0.14 resulting in b1 = 0.5110, b2 = 1.55106 and f = 2.0621,
characterising the behaviour of a [90]n B/Al unidirectional
laminate.
m12 = 0.22 resulting in b1 = 0.6443, b2 = 1.9579 and f = 2.6022,
characterising the behaviour of a [0]n B/Al unidirectional
3. Material A3: E1 = 18.9 GPa, E2 = 90.9 GPa, G12 = 14.6 GPa,
m12 = 0.12 resulting in b1 = 0.5070, b2 = 0.8902 and f = 1.3972,
characterising the homogenised behaviour of a [904/±45]s
Graphite/Epoxy laminate;
4. Material A4: E1 = 90.9 GPa, E2 = 18.9 GPa, G12 = 14.6 GPa,
m12 = 0.58 resulting in b1 = 1.1249, b2 = 1.9504 and f = 3.0788,
characterising the homogenised behaviour of a [04/±45]s
Graphite/Epoxy laminate.
Moreover the authors considered a wide range of geometrical
ratios, with 0.4 < a/b < 4 and 0.05 < 2a/W < 0.6. The values of Kt
for the corresponding isotropic plates are reported in Table 1.
The numerical results from Ref. [37] ([37] page 50, Table 2) are
compared to the predictions based on Eq. (16) in Table 2 (3.33 < W/
2a < 20) and Table 3 (1.667 < W/2a < 2.5). It is evident that when
W/2a > 3.33 (Table 2) Eq. (16) is very accurate independently of
the material configuration and of the notch acuity (a/b), the maxi-
mum deviation D being always lower than 5%.
 M. Zappalorto, P.A. Carraro / Composites: Part B 68 (2015) 51–58 55

Table 5 4. Material B4: E1 = 11.43 GPa, E2 = 11.43 GPa, G12 = 1 GPa,

Finite-width plates with eccentric circular holes. Comparison between (a) results from m12 = 0.06 resulting in b1 = 0.2985, b2 = 3.3498 and f = 3.6483.
Ref. [38] and (b) results by Eq. (16). Material B1. Relative deviation:D ¼ K ðbÞ ðaÞ
tg =K tg  1.

W/(2a) W/(2e) e tg [15]

K Ktg(a) Ktg(b) D (%) The authors considered initially non-eccentric holes with the
5 10 3.187 9.140 9.743 6.60 ratio W/(2a) ranging from 10 to 2.
5 5 3.235 9.200 9.935 7.99 The numerical results for non-eccentric holes from Ref. [38]
5 3.33 3.302 9.310 10.201 9.57 ([38] page 127, Tables 2 and 3) are compared to the predictions
5 2.5 3.406 9.490 10.616 11.87 based on Eq. (16) in Table 4; the Kt for the isotropic case has been
5 2 3.589 9.790 11.349 15.93
estimated from Ref. [25]. It is evident that when W/(2a) is equal to
4 10 3.308 9.300 10.224 9.93 or greater than 4, the accuracy of Eq. (16) is very satisfactory, rel-
4 5 3.390 9.410 10.553 12.15
ative deviation being always smaller than 10%. As noted in the pre-
4 3.33 3.508 9.610 11.026 14.73
4 2.5 3.697 9.930 11.780 18.63 vious subsection a further reduction of the plate size to notch size
4 2 4.035 10.530 13.131 24.70 ratio, W/2a, results in a decrease of the accuracy of Eq. (16).
3 10 3.590 9.690 11.354 17.17 In Table 5, with reference to material B1, the orthotropic finite-
3 5 3.759 9.950 12.028 20.89 width plate with eccentric holes is considered. It is evident that in
3 3.33 4.008 10.410 13.023 25.10 this case, when W/(2a) is greater than 5 and W/(2e) is greater than
3 2.5 4.413 11.230 14.643 30.39
3.33, Eq. (16) still continues to give accurate results (with devia-
3 2 5.151 12.950 17.590 35.83
tions lower than 10%). Decreasing W/(2a) and W/(2e) results in a
decrease of accuracy of Eq. (16).

Further reducing the plate size to notch size ratio, W/2a, the
deviation increases (see Table 3 with 1.667 < W/2a < 2.5). However
in these cases, it has to be noted that the overall accuracy of Eq.
(16) depends on the notch acuity. In particular the equation
remains accurate for sharp or moderately sharp notches
(a/b > 1.25) and for semicircular notches (a/b = 1), while it becomes
inaccurate for the only case of very blunt notches (a/b = 0.4).
4.2. Comparison to the results by Wang and Song [38]: orthotropic
finite-width plates containing eccentric circular holes
Weixing and Xinlu [38] studied the stress concentration factor
of an orthotropic finite-width plate containing eccentric circular
holes using the finite element method and derived explicit expres-
sions to estimate Kt using an approximate analytical method. The
geometry considered by these authors is shown in Fig. 2.
Four different material configurations were analysed by Weix-
ing and Xinlu [38] (plane stress):
4.3. Comparison to the results by Wu and Mu [39]: orthotropic plates
and cylinders with circular holes
Wu and Mu [39] studied the stress concentration factor of
orthotropic finite-width plates with central holes under biaxial
tension and cylinders with circular holes under tension (Fig. 3).
The following orthotropic materials were considered by Wu and
Mu [39] (plane stress):
Material C1: E1 = 20 GPa, E2 = 9.2 GPa, G12 = 5.0 GPa, m12 = 0.341
resulting in b1 = 0.9479, b2 = 1.5555 and f = 2.5034.
Material C2: E1 = 9.2 GPa, E2 = 20 GPa, G12 = 5.0 GPa, m12 = 0.157
resulting in b1 = 0.6430, b2 = 1.0548 and f = 1.6978.
The numerical results from Ref. [39] are compared to the
predictions based on Eq. (16) in Table 6 (holed plate under biaxial
tension) and Table 7 (cylinder with a hole under tension). It is
evident that the accuracy of Eq. (16) is satisfactory.

1. Material B1: E1 = 164.7 GPa, E2 = 7.99 GPa, G12 = 3.03 GPa,
m12 = 0.25 resulting in b1 = 0.6208, b2 = 7.3731 and f = 7.9939.
2. Material B2: E1 = 204 GPa, E2 = 18.5 GPa, G12 = 5.6 GPa,
m12 = 0.23 resulting in b1 = 0.5561, b2 = 5.9715 and f = 6.5276.
3. Material B3: E1 = 207.33 GPa, E2 = 20.65 GPa, G12 = 10.04 GPa,
m12 = 0.32 resulting in b1 = 0.7176, b2 = 4.4154 and f = 5.133.
5. Validation by comparison to results of FE analyses
In addition to the data taken from the literature, in this section
the assessments based on Eq. (16) are compared to the results from
FE analyses carried out on finite size orthotropic notched plates.
Plane stress conditions and remotely applied tension loading are

σ σ

2a 2a
2.5 W

σ σ Material reference
1 system

W W
2

σ σ
(a) (b)
Fig. 3. Orthotropic finite-width plates and cylinders with holes analysed by Wu and Mu [39].
56 M. Zappalorto, P.A. Carraro / Composites: Part B 68 (2015) 51–58

Table 6 are used with a very fine mesh pattern close to the notch tip, in
Finite-width plates with circular holes under bi-axial tension. Comparison between order to get results with high degree of accuracy.
(a) results from Ref. [39] and (b) results by Eq. (16). Material C1. Relative deviation:
D ¼ K ðbÞ ðaÞ Results are listed in Tables 8–10. It is evident that, when applied
tg =K tg  1.
to lateral notches, the accuracy of Eq. (16) is very satisfactory over
W/(2a) e tg [27]
K Ktg(a) Ktg(b) D (%) the entire range of q/a and h/a considered. The maximum devia-
6.667 2.778 3.286 3.225 1.84 tion, D, is 7.7%, only for one case, while in many cases D < 2%.
3.333 3.111 3.714 3.642 1.93 On the other hand, for central U notches, Eq. (16) is found to be
2.000 3.556 4.571 4.199 8.15 accurate only when h/a P 3 (W/(2a) P 4).

used to verify the degree of accuracy and the limits of validity of

Eq. (16).
Different notch geometries are considered: symmetric lateral
U-notches, symmetric lateral V-notches with an opening angle,
2a = 90° and central U notches with q/a ranging from 0.1 to 1
and h/a from 1 to 10 (Fig. 4). Moreover, two different material
systems are considered:
1. Material D1: E1 = 160 GPa, E2 = 10 GPa, G12 = 5 GPa, m12 = 0.3
resulting in b1 = 0.7198, b2 = 5.5572 and f = 6.277, characteris-
ing a typical orthotropic carbon fibre-epoxy lamina with fibres
aligned with the loading direction.
2. Material D2: E1 = 10 GPa, E2 = 160 GPa, G12 = 5 GPa,
m12 = 0.018753 resulting in b1 = 0.1799, b2 = 1.3893 and
f = 1.5692, characterising a typical orthotropic carbon fibre-
epoxy lamina with fibres perpendicular to the loading direction.
Ò
FE analyses are carried out using Ansys version 13 software
package. Parabolic isoparametric elements (PLANE183 in ANSYS)
6. Final remarks on the degree of accuracy of the proposed
equation
In the previous sections the accuracy of Eq. (16) was evaluated
and discussed by comparison to a bulk of results either taken from
the literature or coming from numerical analyses carried out by the
present authors.
As mentioned, by a strictly theoretical point of view, Eq. (16) is
valid only for:
1. Shallow notches, either lateral or central, in infinite or semi-
infinite plates under tension;
2. Sharp deep lateral notches in plates under tension.
In the case of finite components under tension, Eq. (16) is
approximate and its degree of accuracy depends on the local geom-
etry (notch acuity and notch shape), the global geometry (lateral or
central notch, plate or cylinder) and the material elastic properties.

Table 7
Cylinders with circular holes under tension. Comparison between (a) results from Ref. [39] and (b) results by Eq. (16). Relative deviation: D ¼ K ðbÞ ðaÞ
tg =K tg  1.

W/(2a) e tg [27]
K Ktg(a) Ktg(b) D (%)

Material C1 26.66667 2.750 2.878 3.192 10.94

2.666667 3.613 3.918 4.273 9.06
1.6 5.935 6.306 7.183 13.90
Material C2 26.66667 2.750 2.449 2.471 0.88
2.666667 3.613 3.429 3.196 6.79
1.6 5.935 5.510 5.147 6.58

1 Material reference
system

2
σ σ σ

2h 2h 2a

2α=90°
ρ ρ
ρ
a a h

W W W

σ σ σ
(a) (b) (c)
Fig. 4. Notched plates analysed. (a) Symmetric lateral U-notches; (b) V-notches with an opening angle, 2a = 90°; (c) central U notches. Plane stress conditions, tension
loadings.
 M. Zappalorto, P.A. Carraro / Composites: Part B 68 (2015) 51–58 57

Table 8 Table 10
Symmetric lateral U-notches, plane stress conditions. Comparison between Ktg values Central U notches, plane stress conditions. Comparison between Ktg values based on
based on Eq. (16) and FE results. Relative deviation D = KtgEq.(16)/KtgFEM  1. Eq. (16) and FE results. Relative deviation D = KtgEq.(16)/KtgFEM  1.

q/a h/a e tg
K Material KtgFEM KtgEq.(16) D (%) q/a h/a e tg
K Material KtgFEM KtgEq.(16) D (%)

0.1 1 8.40 D1 23.09 24.22 4.9 0.1 1 9.23 D1 23.91 26.83 12.2
0.2 1 6.15 D1 16.78 17.15 2.2 0.2 1 6.99 D1 17.59 19.80 12.7
0.5 1 4.19 D1 10.83 11.01 1.6 0.5 1 5.13 D1 11.67 13.96 19.6
1 1 3.25 D1 7.89 8.06 2.2 0.1 3 7.9 D1 21.32 22.65 6.5
0.1 3 8.01 D1 22.12 23.00 4.0 0.2 3 5.88 D1 15.61 16.31 4.5
0.2 3 5.86 D1 16.08 16.27 1.2 0.5 3 4.1 D1 10.19 10.73 5.3
0.5 3 3.98 D1 10.37 10.36 0.1
0.1 10 7.61 D1 21.05 21.74 3.3
1 3 3.05 D1 7.55 7.43 1.5
0.2 10 5.64 D1 15.39 15.56 1.1
0.1 10 7.98 D1 21.56 22.91 6.2
0.5 10 3.90 D1 10.02 10.10 0.8
0.2 10 5.84 D1 15.67 16.18 3.2
0.5 10 3.96 D1 10.10 10.30 1.9 0.1 1 9.23 D2 7.58 7.46 1.6
1 10 3.04 D1 7.35 7.40 0.7 0.2 1 6.99 D2 5.85 5.69 2.6
0.5 1 5.13 D2 4.50 4.24 5.8
0.1 1 8.40 D2 6.92 6.80 1.7
0.1 3 7.9 D2 6.56 6.41 2.2
0.2 1 6.15 D2 5.14 5.04 2.0
0.2 3 5.88 D2 4.91 4.83 1.6
0.5 1 4.19 D2 3.67 3.50 4.5
0.5 3 4.10 D2 3.49 3.43 1.6
1 1 3.25 D2 3.00 2.76 7.7
0.1 10 7.61 D2 6.34 6.19 2.4
0.1 3 8.01 D2 6.43 6.50 1.1
0.2 10 5.64 D2 4.72 4.64 1.7
0.2 3 5.86 D2 4.75 4.82 1.5
0.5 10 3.90 D2 3.29 3.27 0.4
0.5 3 3.98 D2 3.29 3.34 1.6
1 3 3.05 D2 2.58 2.61 1.1
0.1 10 7.98 D2 6.42 6.48 0.9
0.2 10 5.84 D2 4.74 4.79 1.1
0.5 10 3.96 D2 3.28 3.32 1.4
Eventually, when dealing with composite materials, it must be
1 10 3.04 D2 2.56 2.60 1.5
mentioned that Eq. (16) can be applied to composite orthotropic
laminae (0° or 90° laminae) but it does not account for edge effects
arising at the hole/notch edge in cross ply laminates. Moreover, Eq.
Table 9
(16) can be applied to homogenised orthotropic laminates; how-
Symmetric lateral V-notches with 2a = 90°, plane stress conditions. Comparison
between Ktg values based on Eq. (16) and FE results. Relative deviation D = KtgEq.(16)/
ever it has to be accounted for that the maximum stress deter-
KtgFEM  1. mined does not necessarily correspond to any of the actual
maximum notch stresses occurring in the different plies of a mul-
q/a h/a e tg
K Material KtgFEM KtgEq.(16) D (%)
tidirectional laminate.
0.1 1 8.16 D1 23.40 23.47 0.3
0.2 1 6.05 D1 16.78 16.85 0.4
0.5 1 4.17 D1 10.91 10.95 0.4 7. Conclusions
0.1 3 7.78 D1 22.42 22.28 0.6
0.2 3 5.77 D1 16.08 15.97 0.7 Based on some analytical derivations from classical solutions in
0.5 3 3.96 D1 10.45 10.29 1.5
the frame of the orthotropic theory of elasticity, an engineering for-
0.1 10 7.75 D1 21.85 22.18 1.5
0.2 10 5.74 D1 15.67 15.88 1.3 mula for the theoretical stress concentration factor of orthotropic
0.5 10 3.94 D1 10.18 10.23 0.5 notched plates has been provided. Such an expression needs, as
0.1 1 8.16 D2 6.50 6.62 1.8 inputs, the material elastic constants and the Kt of the correspond-
0.2 1 6.05 D2 5.00 4.96 0.8 ing isotropic case. The results given by the new equation have been
0.5 1 4.17 D2 3.64 3.49 4.2 compared to those from a number of FE analyses carried out on
0.1 3 7.78 D2 5.95 6.32 6.2 finite size orthotropic notched plates showing a satisfactory agree-
0.2 3 5.77 D2 4.55 4.74 4.2
0.5 3 3.96 D2 3.24 3.32 2.5
ment, especially for lateral notches. Accordingly, the proposed
0.1 10 7.75 D2 5.94 6.30 6.0 solution represents a very useful tool to estimate the notch stress
0.2 10 5.74 D2 4.54 4.72 3.9 concentration factor of orthotropic plates, injected short fibre spec-
0.5 10 3.94 D2 3.23 3.31 2.4 imens, composite orthotropic laminae, composite unidirectional
laminates and homogenised orthotropic composite laminates.

However based on the results presented in Sections 4 and 5 it Acknowledgements

can be stated that for lateral notches in finite size plates Eq. (16)
is very accurate over a wide range of cases (W/(2a) P 2).
In the case of central notches or holes the range of accuracy of
Eq. (16) is slightly more limited and errors lower than about 10%
can been obtained when W/(2a) P 4. Indeed, the case of central
notches or holes is more sensitive to the finite size of the compo-
nent and, in this case, as observed by Tan [40], the application of
isotropic finite-width correction factors to estimate the orthotropic
Kt might result in large errors.
Aware of this last mentioned limitation, the results presented in
this work allow to conclude that Eq. (16) represents a useful design
tool for engineers engaged in the stress analysis of notched ortho-
tropic plates and cylinders under tension. Conversely the equation
is valid neither for plates under in plane shear loadings nor for
notched cylinders under torsion.
The financial support to the activity by Veneto Nanotech, the
Italian cluster of Nanotechnology, is greatly acknowledged.
References
[1] Creager M, Paris PC. Elastic field equations for blunt cracks with reference to
stress corrosion cracking. Int J Fract Mech 1967;3:247–52.
[2] Lazzarin L, Tovo R. A unified approach to the evaluation of linear elastic stress
fields in the neighborhood of cracks and notches. Int J Fract 1996;78:3–19.
[3] Zappalorto M, Lazzarin P. Stress fields due to inclined notches and shoulder
fillets in shafts under torsion. J Strain Anal Eng Des 2011;46:187–99.
[4] Zappalorto M, Lazzarin P, Filippi S. Stress field equations for U and blunt V-
shaped notches in axisymmetric shafts under torsion. Int J Fract
2010;164:253–69.
[5] Lazzarin P, Zappalorto M, Berto F. Generalised stress intensity factors for
rounded notches in plates under in-plane shear loading. Int J Fract
2011;170:123–44.
58 M. Zappalorto, P.A. Carraro / Composites: Part B 68 (2015) 51–58
[6] Zappalorto M, Lazzarin P. A new version of the Neuber rule accounting for the
influence of the notch opening angle for out-of-plane shear loads. Int J Solids
Struct 2009;46:1901–10.
[7] Zappalorto M, Lazzarin P. A unified approach to the analysis of nonlinear stress
and strain fields ahead of mode III-loaded notches and cracks. Int J Solids Struct
2010;47:851–64.
[8] Savin GN. Stress concentration around holes. New York: Pergamon Press;
1961.
[9] Lekhnitskii SG. Theory of elasticity of an anisotropic body. Moskow: Mir
Publishers; 1984.
[10] Bonora N, Costanzi M, Marchetti M. On closed form solution for the elastic
stress field around holes in orthotropic composite plates under in-plane stress
conditions. Compos Struct 1993;25:139–56.
[11] Bonora N, Costanzi M, Marchetti M. A computational procedure to calculate
stress–strain field around simple shape holes in composite laminates. Compos
Struct 1994;53:1167–79.
[12] Chern SM, Tuttle ME. On displacement fields in orthotropic laminates
containing an elliptical hole. J Appl Mech 2000;67:527–39.
[13] Ukadgaonker VG, Rao DKN. Stress distribution around triangular holes in
anisotropic plates. Compos Struct 1999;45:171–83.
[14] Ukadgaonker VG, Rao DKN. A general solution for stresses around holes in
symmetric laminates under inplane loading. Compos Struct 2000;49:339–54.
[15] Ukadgaonker VG, Kakhandki V. Stress analysis for an orthotropic plate with an
irregular shaped hole for different in-plane loading conditions—Part I. Compos
Struct 2005;70:255–74.
[16] Khashaba UA, Selmy AI, El-Sonbaty IA, Megahed M. Behavior of notched and
unnotched [0/±30/±60/90]s GFR/EPOXY composites under static and fatigue
loads. Compos Struct 2007;81:606–13.
[17] Carraro PA, Quaresimin M. Damage mechanics in G/E tubes under tension–
torsion cyclic loading. Key Eng Mater 2012;488–489:783–6.
[18] Quaresimin M, Carraro PA. On the investigation of the biaxial fatigue
behaviour of unidirectional composites. Compos Part B: Eng 2013;54:200–8.
[19] Quaresimin M, Carraro PA. Damage initiation and evolution in glass/epoxy
tubes subjected to combined tension–torsion fatigue loading. Int J Fatigue
2014;63:25–35.
[20] Quaresimin M, Carraro PA, Pilgaard Mikkelsen L, Lucato N, Vivian L, Brøndsted
P, et al. Damage evolution under internal and external multiaxial cyclic stress
state: a comparative analysis. Compos Part B: Eng 2014;2014(61):282–90.
[21] Carraro PA, Quaresimin M. A damage based model for crack initiation in
unidirectional composites under multiaxial cyclic loading. Compos Sci Technol
2014;99:154–63.
[22] Xu X, Sun L, Fan X. Stress concentration of finite composite laminates with
elliptical hole. Comput Struct 1995;57:29–34.
[23] Xu X, Sun L, Fan X. Stress concentration of finite composite laminates
weakened by multiple elliptical holes. Int J Solids Struct 1995;32:3001–14.
[24] Xu XW, Yue TM, Man HC. Stress analysis of _nite composite laminate with
multiple loaded holes. Int J Solids Struct 1999;36:919–31.
[25] Peterson RE. Stress concentration factors. New York: John Wiley & Sons; 1974.
[26] Kato A. Design equation for stress concentration factors of notched strips and
grooved shafts. J Strain Anal Eng Des 1992;27:21–8.
[27] Nisitani H, Noda N. Stress concentration of a cylindrical bar with a V-shaped
circumferential groove under torsion, tension or bending. Eng Fract Mech
1984;20:743–66.
[28] Noda N, Takase Y. Stress concentration formula useful for all notch shape in a
round bar (comparison between torsion, tension and bending). Int J Fatigue
2006;28:151–63.
[29] Yang Z. Stress and strain concentration factors for tension bars of circular
cross-section with semicircular groove. Eng Fract Mech 2009;76:1683–90.
[30] Zappalorto M, Berto F, Lazzarin P. Practical expressions for the notch stress
concentration factors of round bars under torsion. Int J Fatigue
2011;33:382–95.
[31] Lazzarin P, Zappalorto M. A three-dimensional stress field solution for pointed
and sharply radiused V-notches in plates of finite thickness. Fatigue Fract Eng
Mater Struct 2012;2012(35):1105–19.
[32] Zappalorto M, Carraro PA. Stress fields at sharp angular corners in thick
anisotropic composite plates. Compos Struct 2014;117:346–53.
[33] Inglis CE. Stresses in a plate due to the presence of cracks and sharp corners.
Trans Inst Naval Architects 1913;55:219–30.
[34] Neuber H. Theory of notch stresses. Berlin: Springer-Verlag; 1958.
[35] Agarwal BD, Broutman LJ, Chandrashekhara K. Analysis and performance of
fiber composites. 3rd ed. Hoboken-New Jersey: John Wiley & Sons; 2006.
[36] Chiang CR. Stress concentration factors of edge-notched orthotropic plates. J
Strain Anal Eng Des 1998;33:395–8.
[37] Weixing Y, Xinlu Y. Stress concentration factor for an orthotropic finite-width
plate containing elliptical edge notches. Compos Sci Technol 1991;41:47–53.
[38] Wang QZ, Song XL. An expression of stress concentration factor for a circular
hole eccentrically located in a tensile finite-width strip of orthotropic material.
J Mater: Des Appl 2008;222:123–9.
[39] Wu H-C, Mu B. On stress concentrations for isotropic/orthotropic plates and
cylinders with a circular hole. Compos: Part B-Eng 2003;34:127–34.
[40] Tan SC. Finite-width correction factors for anisotropic plate containing a
central opening. J Compos Mater 1988;22:1080–97.