SAE International SAE Transactions: This Content Downloaded From 194.27.186.225 On Sat, 10 Dec 2022 10:46:09 UTC

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

Design Data and Buckling of Laminated Composite Triangular Plates

Author(s): R. R. Valisetty and A. D. Reddy


Source: SAE Transactions, Vol. 94, Section 4 (1985), pp. 673-677
Published by: SAE International
Stable URL: https://www.jstor.org/stable/44729709
Accessed: 10-12-2022 10:46 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

SAE International is collaborating with JSTOR to digitize, preserve and extend access to SAE
Transactions

This content downloaded from 194.27.186.225 on Sat, 10 Dec 2022 10:46:09 UTC
All use subject to https://about.jstor.org/terms
850890

Design Data and Buckling of Lamin


Composite Triangular Plate
R. R. Valisetty
Aero Structures Inc.
Arlington, VA
A. D. Ready
School of Aerospace Engineering
Georgia Institute of Technology
Atlanta, GA

ABSTRACT section at the Georgia Institute of Technology in


sponsorship with the Lockheed-Georgia Company. The
Design studies have been performed for results of this study, reported in Reference (l), confirm
isosceles
triangular plates subjected to uniaxial and thebi-axial
possibilities for tailoring the skin layups, stiffening
grids and stiffener sizing all at once in relation to the
loading. The material system used in this application
is AS4/3502 graphite/epoxy with (0, 90) , (0,overall90, +applied
45) loading. In this study, it has been found
that
(0, + 60) , (+ 45, 0, 90) , (+ 45) and (+ 4J, + 45) local skin buckling is an important design criteria.
layups.
The problem has been solved by using a GalerkinOne of the earliest studies on triangular plate
buckling was by Woinowsky-Krieger (2) . He obtained
procedure. The buckling load results are normalized
with respect to a rectangular plate of correspondingan exact solution for the buckling of an isotropic skin
base and height dimensions and compared.ofThe equilateral
data triangular shape. A Galerkin procedure
suggests that certain layups are preferred over and aothers
collocation technique were, respectively, used in
in particular design applications. The local References
skin (3) and (4) to consider isosceles triangular
buckling load with triangular stiffening pattern plates underisdifferent loading conditions.
about 2.5 to 9 times greater than the load with This paper describes a Galerkin procedure for the
rectangular stiffening pattern. buckling of isosceles triangular shaped laminated
plates under uniaxial and bi-axial loading conditions.
The edge fixity conditions are conservatively assumed
COMPOSITE MATERIALS are now being extensively to be simply supported. The actual constraint lies
used for primary load bearing aerospace structures. somewhere between the clamped and simple boundary
Efficiency in these applications requires that the conditions and is controlled by the relative stiffnesses
structures be stiffened in some way. Experience of the skin and stringers and the mode or shape of
gained with metallic stiffened structures was relied overall deformation. Different layups are considered
upon in earlier years to design composite stiffened for the skin. These are (0, 90) (0, 90, + W, ±
structures. Individual parts are often fabricated and 60) , (0, + 45, 0, 90) , (+ 45) , and (+ 45, + 45Ķ. The
assembled with fasteners. The stiffening patterns material system is AS4/3502 graphite/epoxy. An
included familiar axial stringers, circular rings and isotropic
a case with aluminum as the material was also
combination of both. This approach basically has two studied as a reference case. Buckling load is obtained
drawbacks. First, it does not fully utilize the for each of these skins and rectangular plate of
directional nature of the composite materials. corresponding base (2a) and height (h) dimensions. The
Secondly, it is recognized that fastening leads to local effect of varying h/a ratio is also investigated. The
strength degradation in composite structures. Recent results indicate that triangular stiffening is superior.
advances in automated composite manufacturing The data suggests certain layups are preferred over
methods and co-curing of stiffeners and skin hold great others in particular design and loading situations. The
potential for new stiffening concepts. Triangular local skin buckling load with triangular stiffening
stiffening pattern is unique among these concepts. By pattern is about 2.5 to 9 times greater than for the
controlling size, shape and direction of the basic grid, rectangular stiffening pattern.
overall stiffness of the structures can be changed to
take the multiaxial overall loads more effectively.GALERKIN PROCEDURE
Certain design concepts involving isogrid (equilateral
triangular stiffening), orthogrid (stiffeners of some Consider an isosceles triangular plate of base
geometry in both axial and hoop directions) width, and 2a, and height, h, referred to a x-y coordinate
generalized orthogrid (different sized stiffeners in
axial and hoop directions) were studied for cost*Numbers in parentheses designate references at end
effectiveness on a 665.48 cm long C-130 fuselage of paper.

4.673
0096-736X/86/9404-0673$02.50
Copyright 1986 Society of Automotive Engineers, Inc.

This content downloaded from 194.27.186.225 on Sat, 10 Dec 2022 10:46:09 UTC
All use subject to https://about.jstor.org/terms
system as shown in Figure 1. The loading conditions modified Galerkin integral (with respect to A
for the numerical example problem are shown in parameters). In the variational form the GalerRîn
Figure 2. The triangular plate is subjected to a bi- principle appears to be
axial loading. The governing equation for the buckling
loads is obtained under the assumption of linear (small)
displacements (5). If w is the transverse displacement, (jļDll -,xxxx + (2D12 + ^óé^xxyy
then this equation can be written in the following form

D..W
11 +,xx
(2D10
121266
+ 4D¿¿)w
,xxyy 66 22
+ D00w 22
,yyyy + D o o w + N w
22 o o w ,yy + ^x w ,xx

+ Nw + N w =0
X ,xx y ,yy

D-parameters are the flexural rigidities of + Nythe


w>Xy ļplate.
<5 W dxdy
These rigidities depend upon plate layups. The
expressions which relate them to the layer elastic
moduli are given in Reference (6). The development *ļ(IV,xx * Di2»,yy>?7^
here is restricted to balanced symmetric layups for
which D.¿ and are zero# The planform of the 2
plate is also restricted to an isosceles triangle. The
edges are simply supported, therefore, the boundary
+ (D12w,xx + D22w,yy)2~~2 a + n
conditions are

@ y J = 0: w = 0, 9 D..W + D10w = 0 ah )(h <Sw +a ôw ,y ) h


y J 9 11 ,xx 12 ,yy
♦ ah 7^a+h )/łh2
7^a+h y = 0 0 ,y y = 0
@ y = 2h - xh/a: w = 0, (D^w^ + Dļ2w>yy)h2
x = (2h - y)a/h
+ (D12w,XX + D22w,yy)a2

+ ^D^ ah = 0 *1(Dll»,xx*Dl2»,yy'¡27^
@ y = xh/a: w = 0, (D^w ^ + Dļ2Wjyy)h2 2

+ (D12w,xx+d22w,yy)-F- 2
+ (D12w,xx + D22w,yy)a2 a + n

- *°66ah = 0
An approximate solution is presented to this
problem based on a modified Galerkin approach. The
more familiar Galerkin method requires that the 66 ,xy a2 + h2J V f~2Va +y
> >
ha2 =+ nh n 0
assumed solution satisfy all the edge boundary
conditions. It is difficult to come up with such a x = ay/h
solution for the present problem. The following mode
shape The double integral in the above equation has to be
computed over the area, A, of the plate.
w = w = £ ZAmn(sin TT* In the Galerkin approach, the error in satisfying
the governing differential equation is minimized with
respect to the assumed mode shape. In the modified
. niTX
- sin . . miry* n 0 form, the error from the unsatisfied edge conditions is
a . h - -
also considered. The above principle provides a system
m = 1,3,5,7, ... n = 1,3,5,7, . . . , m¿ n of homogeneous equations for the solution of the
amplitude parameters and the requirement that this
solution be nontrivial provides the buckling load. The
numerical results for the examples considered in the
next section are obtained using a six term
approximation for the mode shape. The combination
. n ïïx . m iryx ^ 0
- sin . of m and n considered are (1, 3), (1, 5), (3, 5), (l, 7), (3,
a , h - -
7), and (5, 7).

m - 1,3,5,7, ... n - 1,3,5,7, . . ., m ^ n RESULTS AND DISCUSSION

In order to demonstrate the effectiveness of


triangular stiffening, various layups are considered
which is assumed for the buckled deformation satisfies
the triangular plate. These are isotropic, (0, 90)
all the edge displacement boundary conditions and the
moment boundary condition on the edge y = 0. The90, + 45) , (0, + 60) , (0, + 45, 0, 90)s, (+ 45) and Ć+
+ 45) . These layups are also considérée? from
equations necessary for the solution of mode amplitude
parameters, Amn, are obtained by minimizing the viewpoint of skin and stiffener sizing in the design

4.674

This content downloaded from 194.27.186.225 on Sat, 10 Dec 2022 10:46:09 UTC
All use subject to https://about.jstor.org/terms
850890

fuselage section subjected to multiaxial loading in on the height to semi-base width of the
Reference (l). The following material properties v isosceles triangular plate in load Case 2.
typical of a graphite/epoxy composite, are used for the ii) v Certain ply orientations like (0, 90)s, (0, 90, +
individual lamina. 45)s and (0, + 60)§ are very efficient
compared to the metallic isotropic material
in load Case 2 at h/a ratios above 3.
Ejj = 18.5 X lO^psi
ACKNOWLEDGEMENT

E22 = E33 = 1#6^ X 1()6Psi This work was supported by the Manufacturin
Branch, Lockheed-Georgia Company. We acknowled
= ^*55 " ^66 ~ ^.87 x ^ Ps* Dr. Roy Scruggs, Program Manager, for his support
cooperation.

V 12 = V13 = V23 = 0,3 REFERENCES

Isotropic plate, v = 0.3; E = 10 x 10^ psi 1. Reddy, A.D., Valisetty, R.R. and Rehfield, L.W.,
"Continuous Filament Wound Composite
Results are presented in Figures (3) - (11) for the Stiffening Concepts for Aircraft Fuselage
primary buckling load. These are nondimensionalized Structures," presented at the 25th
with respect to the corresponding buckling loads of a AIAA/ASME/ASCE/AHS Structures, Structur
rectangular plate (2a x h) in the same loading situation Dynamics and Materials Conference, Palm
and with the same boundary restraint. The plate is Springs, CA, May 1984. Accepted for publication
subjected to three different loads as shown in Figure 2. in the Journal of Aircraft.
The results are studied on the basis of the ability of
the laminated plate to resist these loads and a size 2. Timoshenko, S., Theory of Elastic Stability,
parameter, h/a, which is the ratio of plate height to its McGraw-Hill Book Company, Inc., 1st Edition, p.
base dimensions.
371, 1936 .
Figures 3-8 illustrate the variation of critical
load ratios of the triangular plate (P A) to 3. Klitchielf, J.M., "Buckling of a Triangular Plate
rectangular plate (P □) with respect to the h/a ratios By Shearing Forces," Quarterly Journal of
for different materials. The case of an isotropic rightMechanics and Applied Mathematics, Vol. IV, Pt.
angled triangle (h/a = 1) indicates that the load 3, pp. 257-259, 1951.
capacity of the skin element can be doubled by
changing the plate from square to triangle in load Case 4. Cox, H.L. and Klein, B., "The Buckling of
3.
Isosceles Triangular Plates," Journal of the
The (0, 90) layup though close to one another in Aeronautical Society, pp. 321-325, May, 1955.
load Cases 1 ands3, the performance increases steeply
for Case 2 at higher h/a ratios. The load ratio is about 5. Timoshenko, S. and Gere, J.M ., Theory of
9.5 at h/a = 4. This means that side angles greater Elastic Stability, McGraw-Hill, 1961.
than 70 provide better skin designs.
The results on three quasi-isotropic layups (0, 90, 6. Jones, R.M., Mechanics of Composite Materials,
+ 45) , (+ 45, 0, 90) and (0, + 60) (Figures 5-7) provide McGraw-Hill, 1975.
somes interesting rlsults. Figuri 5 is comparable to
Figure 7 whereas Figure 6 has almost no resemblance
to either of these. The (+ 45, 0, 90) layup behaves
very similar to the homogeneous isotropic material
case. Unlike the metallic isotropic case (Figure 3), the
other two quasi-isotropic layups have a higher load
carrying ability at h/a ratios beyond 3. This may be a
helpful design input.
The (+ 45) and (+ 45, + 45) layups give the same
results because of the normalization with respect to a
rectangular plate of the same material. This layup is
not superior to the metallic case.

CONCLUSIONS

A study on laminated composite triangular plates


subjected to both uniaxial and bi-axial loading has been
performed. A Galerkin procedure has been employed
to obtain the critical loads of triangular plates of
different shapes and ply layups. There are two main
conclusions that can be drawn from this work:
i) The triangular plate carries a much higher
load than the rectangular plate in general.
This could be as much as 9.5 times depending

4.675

This content downloaded from 194.27.186.225 on Sat, 10 Dec 2022 10:46:09 UTC
All use subject to https://about.jstor.org/terms
5|

Y • Cas* 1
4 - A Case 2

p. ■ Case 3

/ ' h 1 «

12 3 4
L

h
h/a

Figure 1. Dimensions and Coordinate System for Figure 3. Isotropic Case - Aluminum

the Triangular Plate


• Case 1 J
4 - A Case 2 ^
Y i h2
2^2 CASE 1 □ ■ Ca^3^ ^ "
JJ 2 o P2oah 2 o 2 Nx= -P
^¿f a -t h
1 -

Qy

Y OC- CASE2
12 3 4

h/a

X/_p-25ÎL_ Nv=0
a2+h2
Figure 4. Plate with (0, 90) 8 Lay up

-X N»="P # Case 1 /
ft 1 1 1 1 r g
4


-

Case
A Case

3
2

v/7v^ case 3
Y'p
Nx=-P 1 -

1 2 3 *
r 1 1 1 1 'Y '* h/a

Figure 2. Loading Conditions Figure 5. Plate with (0. 90,±45)s Layip

4.676

This content downloaded from 194.27.186.225 on Sat, 10 Dec 2022 10:46:09 UTC
All use subject to https://about.jstor.org/terms
850890

5 ļ

# Case 1
4 - A Case 2
U ■ Case 3

12 3 4

h/a

Figure 6. Plate wi

# Case 1 /
4 - ▲ Case 2 /
□ ■ Case 3 É

1 -

12 3 4

h/a

Figure 7. Plate with (0,±60)s Layup

• Case 1
4 - A Case 2 -

^ ■ Ca*re3^^

12 3 4

h/a

Figure 8. Plate with (±45)s or (±45,±45)g Layup

4.677

This content downloaded from 194.27.186.225 on Sat, 10 Dec 2022 10:46:09 UTC
All use subject to https://about.jstor.org/terms

You might also like