139

CHAPTER 5

BEHAVIOUR OF HPSFRC PLATE ELEMENTS

5.1 GENERAL

Most of the research carried out on plates under combined loads has
been conducted on metal plates. Although concrete plates supported along all
the edges (4 edges) and subjected to combined uniaxial or biaxial in-plane and
transverse loads are encountered in the structural applications such as wall of
offshore structures, hulls of concrete barges and pontoons, floating bridges
and bridge decks, no such published test data is available for steel fiber
reinforced concrete (SFRC) composite plates subjected to combined in-plane
and transverse loads. In this chapter, the behavior of plate elements and an
analytical model for predicting the out-of-plane deflection of such plates in a
reasonably accurate manner are presented. The results reported in this study
partially fill the above gap in this area.

5.2 BEHAVIOUR OF PLATES UNDER COMBINED IN-PLANE

AND TRANSVERSE LOADS
5.2.1 Test Results and Discussion

Table 5.1 lists the compressive and flexural tensile strengths, static

modulus of elasticity, and Poisson’s ratio based on the correlation of data for

each plate specimen at 28 days of testing. Plate parameters and type of

loading on the plate element with all the edges simply supported are shown in
Figure 5.1.
 140

L y=

L x=
z

Figure 5.1 Plate parameters and loading type

Table 5.1 Data of compressive and flexural strengths, elastic modulus,

and Poisson’s ratio of HPSFRC

Poisson’s
Mix Plate RI f’cf frf Ec
designation Specimen ratio (from
(MPa) (MPa) (GPa)
correlation)
FC1*-0 A1 0 52.56 6.21 29.68 0.19
FC1*-1 A2 2.58 56.01 7.73 30.92 0.20
FC1*-1.5 A3 3.88 57.42 8.19 31.78 0.21
FC1**-0 B1 0 55.70 6.84 30.89 0.20
FC1**-1 B2 2.58 60.21 8.64 32.75 0.22
FC1**-1.5 B3 3.88 61.17 9.28 33.14 0.22
FC3*-0 C1 0 63.86 7.40 34.14 0.23
FC3*-1 C2 2.58 68.91 9.32 36.52 0.24
FC3*-1.5 C3 3.88 69.67 10.13 36.53 0.25
FC3**-0 D1 0 64.28 8.16 34.39 0.23
FC3**-1 D2 2.88 69.74 10.32 36.56 0.25
 141

FC3**-1.5 D3 3.88 70.32 11.08 36.79 0.25

Table 5.2 shows the properties of each specimen as well as the
maximum transverse load per unit area, and the uniform in-plane load per unit
width applied in the x-direction. The variation of in-plane load during the
application of transverse was less than 1 percent. The total transverse load
calculated in kN was applied through load cell placed over the load
distribution assembly, which then distributed uniformly on the specimen. The
in-plane load is given in kN/m as the hydraulic load divided by 0.6 m.

Table 5.2 Summary of experimental program and test results

Applied
Plate Specimen size f’cf q
Lx / h Nx Nx / f’cf.h
Specimen (mm) (MPa) (kN/m) (kN/m2)

A1 600 x 600 x 30 20 52.56 100 0.063 92.22

A2 600 x 600 x 30 20 56.01 100 0.060 125.27
A3 600 x 600 x 30 20 57.42 0 0 93.06
B1 600 x 600 x 30 20 55.7 100 0.060 106.11
B2 600 x 600 x 30 20 60.21 100 0.055 129.44
B3 600 x 600 x 30 20 61.17 0 0 95.83
C1 600 x 600 x 30 20 63.86 100 0.052 115.28
C2 600 x 600 x 30 20 68.91 100 0.048 133.33
C3 600 x 600 x 30 20 69.67 0 0 99.44
D1 600 x 600 x 30 20 64.28 0 0 65.83
D2 600 x 600 x 30 20 69.74 100 0.048 142.78
D3 600 x 600 x 30 20 70.32 100 0.047 159.86

Note: Lx = Ly = 600 mm; h = thickness of plate = 30 mm; aspect ratio = Ly / Lx = 1;

Lx / h = slenderness ratio; Nx = in-plane load per unit width; q = transverse load
per unit area.
 142

Specimens A1 and A2 were tested for combined compressive

in-plane and transverse loads, while the specimen A3 was subjected to
transverse load only to give indication of the reduction in capacity for
specimens within the same series, for comparing the crack patterns. In the
specimen A1, fracture occurred at a central deflection of 4.63 mm and failure
was sudden and explosive. Further, concrete crushing lines were clearly
observed and cracks were more prevalent in x-direction. The pattern of
cracking and crushing is shown in Figure 5.2. In the specimen A2, cracks
were first observed at the bottom of the specimen in the direction of
in-plane load (x-direction) at a transverse load of 125.27 kN/m 2. The central
deflection at maximum transverse load was 11.32 mm. Final failure occurred
at a central deflection of 14.53 mm.

Cracks were first observed for specimen A3 on the flexural tension

(bottom) face as it fails by attaining the ultimate flexural tensile strength, at a
transverse load of 93.06 kN/m 2 and center deflection at this load was
5.19 mm. Yield lines were formed on the tension face of the specimen at
maximum load. Cracks were observed along the diagonal on the bottom face
near the corners (Figures 5.2-5.4). In addition, cracks were developed
perpendicular to the diagonal on the top surface near the corners. These
corners resulted from the corners being held down. This indicates the torsion
cracks on the edge of the plate caused due to twisting moment occurred near
the corners. At the ultimate load, all the specimens that were tested under
combined loads was in a state of axial compression at the middle of the plane,
since the principal moments were both positive. For all other series except
series D (i.e., Series B, and C), the same combination of loading was applied
and tested and the observations were noted. In series D, specimen D1 was
considered for transverse load only and other specimens (i.e., D2 and D3) were
considered for combined loads.
 143

Figure 5.2 Fracture pattern of test specimen (Plate A1) after ultimate
failure

Figure 5.3 Fracture pattern of test specimen (Plate A2) after ultimate
failure
 144

Figure 5.4 Fracture pattern of test specimen (Plate A3) after ultimate
failure

5.2.2 Load-Deflection Response

The transverse load versus out-of-plane central deflection

relationships for plate specimens except plate D1, tested under only transverse
(lateral) loads exhibited similar features. The transverse load versus out-of-
plane deflection (P-∆) at the center of the plate specimens is shown in Figures

5.5–5.8. The fibrous concrete (SFRC) specimens exhibited essentially a linear

P-∆ response well beyond the load while the flexural cracks were first
observed during the test. Each diagram is essentially a straight line up to the
start of cracking. Beyond cracking, a rapid change of slope in the load
deflection curve was observed. On further loading, yielding of fibers started
in one or more regions and spread through the areas still elastic. This
continued till the yield line mechanism developed. A small out-of-plane
 145

deflection occurred in several specimens during the application of the in-plane

loads. This deflection ranged from 0-2% of the transverse load. Zero
deflection corresponding to the un-cracked plate in the horizontal position
supporting its own weight was observed.

The level of the in-plane load (Nx/f’c .h) for all the specimens was

moderate, in the sense that there is no possibility of collapse under in-plane

load alone. Table 5.2 gives the ratio of the applied in-plane load per unit
width to the uniaxial strength of concrete multiplied by the panel thickness for
all of the specimens. The test results and predicted yield capacity for
specimens tested under lateral load only are given in Table 5.3.

140
Plate A1 -combined load
120 Plate A2 - combined load
Plate A3 -lateral load
Transverse load (kN/m2)

100

80

60

40
Plate A1, Vf=0%
20 Plate A2, Vf=1%
Plate A3,Vf=1.5%
0
0 2 4 6 8 10 12 14 16

Central deflection (mm)

Figure 5.5 Transverse (lateral) load versus out-of-plane central

deflection for plate specimens (Series A)
 146

140
Plate B1 -combined load
Plate B2 -combined load
120
Plate B3 -lateral load
Transverse load (kN/m2)

100

80

60

40
Plate B1, Vf=0%
20 Plate B2, Vf=1%
Plate B3,Vf=1.5%
0
0 2 4 6 8 10 12 14
Central deflection (mm)

Figure 5.6 Transverse (lateral) load versus out-of-plane central

deflection for plate specimens (Series B)

140
Plate C2-Combined load
120 Plate C3-Transverse load only
Transverse load (kN/m2)

100

80

60

40

20 Plate C2,Vf=1%
Plate C3, Vf=1.5%
0
0 2 4 6 8 10 12 14 16
Central deflection (mm)
 147

Figure 5.7 Transverse load versus out-of-plane central deflection for

plate specimens (Series C)

160
Plate D1-Transverse load only
140 Plate D2-Combined load
2
load (kN/m2)
(kN/m )

120
100
Transverseload

80
Transeverse

60
40
Plate D1,Vf=0%
20
Plate D2, Vf=1%
0
0 2 4 6 8 10 12 14
Central deflection (mm)

Figure 5.8 Transverse load versus out-of-plane central deflection for

plate specimens (Series D)

Table 5.3 Test results and predicted yield capacity for specimens
tested under lateral loads only

Aspect q (yield-
Plate q (test)
ratio Lx/h line) q test /q yield -line
Specimen kN/m2
(Ly/Lx) kN/m2
A3 1 20 93.06 60.94 1.527

B3 1 20 95.83 69.05 1.388

C3 1 20 99.44 73.95 1.345

D1 1 20 65.83 59.56 1.105

 148

Note: q yield- line was obtained using yield line analysis

 149

5.3 ANALYTICAL MODEL FOR PLATES SUBJECTED TO

COMBINED IN-PLANE AND TRANSVERSE LOADS
5.3.1 Theoretical Background

Several past attempts have been made to apply the methods of limit
analysis to plates subjected to externally applied transverse loads and
internally induced in-plane loads due to the support conditions or externally
applied in-plane and lateral loads. To determine the ultimate transverse load
using the limit analysis, the value of central deflection at the ultimate load
must be known.

Equations for orthotropic plates derived in test books on the theory

of plates (Timoshenko and Krieger 1959) are needed for use in the proposed
analytical model. A typical rectangular plate as considered in the analysis is
shown in Figure 5.9. The plate is of length Lx, width Ly, and uniform
thickness h. The x-, y-, and z- axes are longitudinal, transverse, and vertical
axes, respectively. The uniform in-plane load per unit length Nx and Ny are
applied in the x and y directions to the sides of Ly and Lx respectively. Both
in-plane loads are applied on the thickness of the plates. In addition, there is a
uniform lateral pressure q at any point (x, y), and w represents the out-of-
plane displacements from the flat surface.
 150

Ny

Ly

Nx Nx

Ny

Lx
Lx

Section A-A

Figure 5.9 Plate dimensions, coordinate axes and loading for analytical
model

5.3.2 Analytical model

Based on the Kirchoff theory, the relationship between moments

and deformations for elastic orthotropic plates are given as:

Mx = -Dx (∂2 w/ x2 +υy ∂2 w/ y2) (5.1)

My = -Dy (∂2 w/ y2 + υx ∂2 w/ x2) (5.2)
Mxy = 2Dt (∂2 w/ x y) (5.3)
 151

where

2Dt = (1-√ υx υy) (√Dx Dy) (5.4)

represents torsional rigidity of the plate, while Dx and Dy are the flexural
rigidities in x- and y- directions, respectively. Mx and My are moments per unit
width. Mx is about the y-axis and My is about the x-axis.

The equilibrium equation for an elastic orthographic plate under

combined transverse load q (x,y) per unit area and compressive in-plane loads
Nx and Ny per unit width in the x- and y- directions, respectively is given as:

2Mx/x2 - 2 2Mxy/x y +2My/y2 –Nx 2w/ x2 - Ny 2w/ y2 = - q(x,y)

(5.5)

The preceding equation is independent of the material properties

and hence is valid for both elastic and plastic cases. Substituting the
Equation (5.1) through (5.5), the governing differential equation for the
deflection of the elastic orthotropic plate is written as:

Dx 4w / x4 + 2H 4w / x2 y2 + Dy 4w / y4 = q(x,y) –Nx 2w / x2

- Ny 2w/ y2 (5.6)

where

H = 0.5 (vxDy + vy Dx +4Dt) (5.7)

is called the "effective torsional rigidity" of the plate, while vx and vy are the

Poisson’s ratios in the x and y directions, respectively.

 152

The equations were derived based on the following assumptions:

1. Small transverse deflections are assumed. This means

membrane action is neglected.
2. Plane sections remain plane after bending. This implies that
vertical shear strains are negligible.
3. The stress normal to the mid plane, σz is very small when
compared to other stress components and may be neglected.
4. The plate is initially flat.
5. The material is elastic and orthotropic.

There is experimental evidence to suggest the effective Poisson’s

ratio to be zero, which is reasonable for a cracked concrete plate. For

Poisson’s ratio equal to zero, the equations (5.4) and (5.7) reduce to the

expression given below for the concrete slab can be adopted.

2Dt = √Dx Dy; H = 2Dt

Based on the preceding approximations, the differential equation of

the plate problem on hand can be written as:

Dx 4w/ x4 + 4Dt 4w /x2 y2 + Dy 4w/ y4 = q(x,y) –Nx 2w/ x2

- Ny 2w/ y2 (5.8)

The uniform lateral load is represented by the trigonometric series

 153

 
16 q 1 mπ x nπ y
q(x, y) 
π2
 
m 1 n 1 mn
sin
Lx
sin
Ly
(5.9)

The deflection surface can be expressed in the form of the double

sine series which satisfies the boundary conditions for the simply supported
plate without in-plane restrictions.

 
mπ x nπ y
w(x, y)   
m 1 n 1
w mn sin
Lx
sin
Ly
(5.10)

where m and n are positive odd integers (m = 1,3,5, … and n = 1,3,5, …).

Equations (5.9) and (5.10) are usually expressed in infinite series

form. Herein, however, only a finite series is considered to provide some
acceptable tolerance. In this analysis, six terms are used (m=n = 1,3,5).
Substituting this series in equation (5.8), the following expression for the
coefficient is found.

16q
w mn 
 N m2 π2 N yn2π2 
π 2 mnD  1  x 2  
 LxD L2y D 
 
(5.11)

Dxm4π4 2D x D y m 2 n 2 π 4 Dyn 4π4

where D 4
 2 2
 4
Lx Lx Ly Ly

In which m and n are odd integers 1, 3, 5,.... and wmn = 0 if m or n

or both are even numbers.
 154

5.3.3 Parameters considered in Plate Analysis

Material properties
Compressive strength
Flexural strength (modulus of rupture)
Modulus of elasticity (static modulus)
Poisson's ratio

Plate parameters and boundary conditions

Length in x-direction (Lx)
Length in y-direction (Ly)
Plate thickness (h)
Aspect ratio (Ly/ Lx)
Slenderness ratio (Lx/ h)
All the edges are simply supported without in-plane restraints

5.3.4 Solution-Methodology

An analytical model is presented for predicting the out-of-plane

central deflection of HPSFRC plates subjected to combined biaxial in-plane
compressive and transverse (lateral) loads. The plates are assumed to be
simply supported on four edges without in-plane restraints.

A program has been developed using Turbo C++ code for

determining the out-of-plane central deflection of the orthotropic plates
subjected to uniaxial in-plane compressive and transverse loads, is given in
Appendix 3. The output data obtained for plate elements studied using C
programme developed, is also given in Appendix 3.
 155

5.3.5 Comparison of Experimental Results with Analytical

Predictions

The analytical model described was used to predict the out-of-plane

central deflection of HPSFRC plates tested. Table 5.4 compares the out-of-
plane central deflection obtained from the test and the central deflection from
the proposed analytical model. The average test to predicted ratio is 0.900,
and the coefficient of variation is 24.277 percent. The calculation method is
considered satisfactory in view of the nature of the problem being dealt with
and the wide range of variables involved in the test program.

Although the analysis is presented for biaxial in-plane loads, it has

been checked against uniaxially loaded plates only. The analytical results are
compared with experimental values and found that there is reasonably good
agreement between the two results.

Table 5.4 Out-of-plane central deflection at ultimate load:

Experimental versus Analytical predictions

Central deflection Central deflection

Plate Test /
from test from analysis
Specimen Analytical
(mm) (mm)
A1 4.63 7.28 0.636
A2 11.32 9.49 1.193
A3 5.19 6.76 0.768
B1 5.72 8.04 0.711
B2 10.75 9.25 1.162
B3 5.10 6.68 0.763
C1 5.26 7.90 0.666
C2 10.23 8.53 1.199
 156

Central deflection Central deflection

Plate Test /
from test from analysis
Specimen Analytical
(mm) (mm)
C3 4.89 6.29 0.777
D1 3.52 4.42 0.796
D2 10.16 9.13 1.113
D3 10.35 10.15 1.019

Mean 0.9004
Coefficient of variation (%) 24.277
5.4 SUMMARY

Plate specimens were tested for combined loads and transverse load
only cases. Failure pattern of specimens tested were identified and discussed.
The transverse load versus out-of-plane deflection relationships of specimens
tested under in-plane and lateral loads were discussed. The analytical method
described in this chapter is to obtain the out-of-plane central deflection under
combined loads and to compare the test results of central deflection at
ultimate transverse loads with the predictions of the analytical model.