See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/317167426

investigation of the behavior of reinforced concrete hollow core thick slabs

Article in Computers and Concrete · May 2017

CITATIONS READS

0 1,238

2 authors:

Adel A. Al-Azzawi Sadeq Aziz

Al-Nahrain University Al-Nahrain University
70 PUBLICATIONS 261 CITATIONS 3 PUBLICATIONS 23 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

soil improvement and deformation View project

reactive powder concrete View project

All content following this page was uploaded by Adel A. Al-Azzawi on 26 May 2017.

The user has requested enhancement of the downloaded file.

Computers and Concrete, Vol. 19, No. 5 (2017) 000-000
DOI: https://doi.org/10.12989/cac.2017.19.5.000 000

Investigation of the behavior of reinforced concrete hollow-core thick slabs

Adel A. Al-Azzawi and Sadeq A. Abeda
Department of Civil Engineering, Al-Nahrain University, Baghdad, Iraq

(Received October 2, 2016, Revised January 10, 2017, Accepted February 2, 2017)

Abstract. This study presents investigation of the behavior of moderately thick reinforced concrete slabs having hollow cores
with different parameters. The experimental part of this investigation includes testing eight specimens of solid and hollow-core
slab models having (2.05 m) length, (0.6 m) width and (25 cm) thickness under two monotonic line loads. Load versus
deflection was recorded during test at mid span and under load. Numerically, the finite element method is used to study the
behavior of these reinforced concrete slabs by using ANSYS computer program. The specimens of slab models are modeled by
using (SOLID65) element to represent concrete slabs and (LINK180) element to represent the steel bars as discrete axial
members between concrete nodes. The finite element analysis has showed good agreement with the experimental results with
difference of (4.71%-8.68%) in ultimate loads. A parametric study have been carried out by using ANSYS program to
investigate the effects of concrete compressive strength, size and shape of core, type of applied load and effect of removing top
steel reinforcement.
Keywords: experimental; finite element; reinforced concrete; hollow-core slab

1. Introduction • bf =bw+dia. of core

• bw: width of web
The one way reinforced concrete slab can be done in • hw: height of web(diameter of core)
three forms of slabs: solid slabs, ribbed slabs and hollow- • hf: height of flange top or bottom
core slabs. Hollow core floors represent a special kind of • bf: width of flange
floor totally made of concrete lightened by leaving The main objective of this studying is to investigate the
longitudinal voids (cores) of suitable size to create webs to behavior of one way reinforced concrete slab with
reduce weight, costs, as fire resistance and for electrical and longitudinal hollow cores. Experimental investigation is to
mechanical purposes. Primarily, hollow core elements are be carried out for these types of slabs and solid slabs then
used as floor or roof deck systems and also have the testing results will be compared through changing two
applications as wall panels, sound barriers, spandrel parameters, size of core and shear span effective depth
members and bridge deck units. Reinforced concrete (a/d) ratio. Theoretical investigation is to be carried out by
hollow-core slab is made with cores in one direction with using finite element program ANSYS V.15 (2013) to
temporary or permanent formwork according the following compare the results obtained from the experimental tests.
points: The accuracy of the finite element solution is checked and
1. Cross section of hollow-core slab was defined as used to study the effect of some additional parameters on
continuous of I-section parts and designed as a one way the behavior of slabs.
ribbed slab with top and bottom flanges.
2. Conventional span length up to 7 m and width of each
panel up to 2.4 m 2. Literature review
3. Assumed moderately thick Plate (thickness to span
ratio ( Pajari (2004) carried out four tests on pre-stressed
hollow core slab units. The slab units, two of them 200 mm
4. Limitations of cross section according to ACI-318
thickness with 4m span and two 400 mm thickness with 6 m
(2014) code as shown in Fig. 1:
span, were subjected to pure torsion and together have 1200
• bw ≥ 100 mm
mm width. In all tests the observed failure mode was the
• hw≤3.5 bw
same as the predicted one, i.e., cracking of top flange in
• s≤ 800 mm
angle of 45º with the longitudinal axis of the slab unit. The
• hf ≥s/2, hf ≥50 mm
predicted torsional resistance was 60% and 70% of the
observed resistance for 200 mm and 400 mm slabs,
respectively, when the lower characteristic value for the
Corresponding author, Assistant Professor
tensile strength of the concrete was used for prediction.
E-mail: dr_adel_azzawi@yahoo.com
a Chang et al. (2008) presented a simple computational
M.S. Student
method to be used in design and modeling the structural
E-mail: sadeq_a.abed@yahoo.com
behavior of hollow core concrete slabs in fires. The
Copyright © 2017 Techno-Press, Ltd.
http://www.techno-press.org/?journal=cac&subpage=8 ISSN: 1598-8198 (Print), 1598-818X (Online)
 Adel A. Al-Azzawi and Sadeq A. Abed

core slabs with (200-500) mm thick without transverse

reinforcement through a campaign of detailed nonlinear
finite element analysis, matching experiments test data
collected by PAJARI from past program. These members
(49 specimens) characterized with six nominal slab depths,
five hollow shapes with circular and non-circular voids,
different voids ratios, several pre-stressing steel strands
arrangements and levels of initial pre-stress then
comparative with traditional codes. From finite element
results, the proposed numerical approach was validated by
focusing on a single precast prestressed hollow-core unit.
Fig. 1 Dimensions of cross sections of hollow-core slab Haruna (2014) studied the flexural behavior of precast
pre-stressed concrete hollow-core units with cast-in-place
Table 1 Reinforced concrete slabs specimens concrete topping, through load testing of five full-scale
Slab No. Type a/d specimens. The specimens were divided into two groups
Slab-1 Solid slab 2 wide and narrow. A cast-in-place concrete was cast on top
of the hollow-core units selected from the two groups to
Slab-2 Solid slab 2.5
form a composite system. Presence of cast-in-place topping
Slab-3 Solid slab 3 slab improved the behavior of hollow-core units by
Hollow-core slab increasing the flexural crack initiation and maximum load
Slab-4 2
(150 mm core size) capacities as well as the stiffness. As a result of premature
Hollow-core slab loss of composite behavior, the predicted load capacity of
Slab-5 2.5
(150 mm core size)
these specimens assuming a fully composite behavior
Hollow-core slab
Slab-6
(150 mm core size)
3 remained on the non-conservative side. The results obtained
Hollow-core slab in this study suggested that floor system made of cast-in-
Slab-7 2.5 place concrete topping placed over the machine finished
(100 mm core size)
Hollow-core slab surface of precast concrete hollow-core units with no
Slab-8 2.5
(75 mm core size) interfacial roughening is not able to provide the interface
shear strength required to develop a fully composite
behavior.
proposed model consists of a grillage system using beam Lee (2014) studied the web shear capacity of hollow
elements to include the thermal expansion in both directions core slabs (HCS) through a large number of shear tests. The
and to simulate the vertical cracking in the flanges, with the analysis of results indicated that the minimum shear
topping concrete modeled using shell elements. The reinforcement requirement for deep HCS members are too
simulation outcomes show good agreement with the severe, and that the web-shear strength equation in ACI 318
experimental results code does not provide good estimation of shear strengths
Hai-tao et al. (2011) examined the inner force transfer for HCS members. Thus, in this paper, a rational web-shear
mechanism of a column-supported cast-in-situ hollow core strength equation for HCS members was derived in a simple
slab using finite element analysis by (ANSYS) program. manner, which provides a consistent margin of safety on
The dimensions of the floor system were (21.6×21.6m) shear strength for the HCS members up to 500 mm deep.
consist from three panels in each direction, each panel Through previous studies, hollow cores are made in
(7200×7200 mm) with 300 mm thickness. The diameter of prestressed concrete thin slabs while in the present research
hole was 200 mm and tube filler's length was 900 mm with it is made in normal reinforced concrete thick slabs and the
ribs having 100 mm width between holes. The analysis analysis of these hollow core slabs are investigated. In
results of hollow core slab and the corresponding solid slab general, the challenge of using hollow cores in slabs was
were compared. preventing the shear failure of these slabs which may
Rahman et al. (2012) tested full-scale pre-stressed happen due to removing the cores from concrete. Many
precast hollow-core slabs with different shear span to depth previous researches were carried out to study the shear
(a/d) ratio, which were loaded to failure to ascertain the behavior of these thin slabs, numerically and analytically
ultimate load-carrying capacity of these slabs. A total of 15 but few of them carried experimental investigation for
slab specimens, 5 and 2.5 m in span and having three studying the behavior of thick slabs. In this study, the
different depths, 200, 250 and 300 mm were tested to flexural behavior of one way reinforced concrete thick slabs
failure using four-point load test. It was interesting to note with and without longitudinal cores were adopted
that the failure mode of hollow-core slabs changed from experimentally and numerically by creating circular voids at
pure flexure mode to flexure-shear mode for slabs with middle plane of cross sections with varying the core
depth greater than 200 mm. The analysis of the diameter and the ratio of shear span to effective depth (a/d).
experimental results showed that the existing ACI code
equations underestimated the flexure-shear strength of these
hollow-core slabs.
Brunesi et al. (2014) carried out comparison between 3. Details of experimental test
experimental, analytical and finite element method for shear
strength capacity of precast pre-stressed concrete hollow Experimental program comprises casting eight small-
 Investigation of the behavior of reinforced concrete hollow-core thick slabs

Table 2 Properties of materials

Properties of concrete material

Property Experimental ACI318M (2014)

Compressive strength (ƒ'c) (MPa) 38.1 -

Splitting tensile strength( f 'ct ) (MPa) 3.4 3.09 (0.5√ )

Modulus of rupture ( f r ) (MPa) 3.75 3.83 (0.62√ )
29010.8 29926.2
Modulus of elasticity (Ec) (MPa) 229385.5
(4700√ ) (Wc1.50.043√ )
Properties of steel reinforcement material
(a)
Property Test results

Nominal diameter (mm) 8

Measured diameter (mm) 7.86

Yield stress (ƒy) (MPa) 578.18

Ultimate stress (ƒu) (MPa) 655.74

(b)
scale (1:2) one way solid and hollow-core reinforced
concrete thick slab specimens taking into consideration the
scaling of steel reinforcement by using half diameter of bar
in the slab specimen. Three sizes of circular cores with
three different values of shear span effective depth ratio
(a/d) (ratio of load position to the effective depth of the
slab) was considered in experimental work as shown in
Table 1. The aim was to study the flexural and shear
behavior of these slabs under two static line loads by using (c)
the flexural testing machine.

3.1 Properties of the slab specimens

For the slab specimens, the properties of the hardened

concrete and steel reinforcement which used for
manufacturing the prototype of these slabs are summarized
in Table 2. (d)
The nominal dimensions of slab specimens are (2050
mm) in length with (600 mm) width and (250 mm) Fig. 2 Cross sections of slabs (a) Solid slab (b), (c) &
thickness. These slabs which have span length (1750 mm) (d) Hollow-core slab with core diameter (150,100,75
were tested under two line loads with different shear span mm)
effective depth ratio (a/d). These contains, three one-way
solid slab, three hollow-core slabs with circular core size
(150 mm) and another two hollow-core slabs with circular
core size (100 mm) and (75 mm). The hollow cores of these
slabs are molded by using Poly Vinyl Chloride (PVC) pipes
longitudinal through slabs with (2 mm) thickness.
The reinforcement design of all one way reinforced
concrete slabs was done according to the design procedure
given in ACI 318 (2014) for both solid and hollow-core
slabs. The main longitudinal reinforcement consists from 6
bars with (Ø 8 mm) at top and bottom while the secondary
transverse reinforcement consists from 14 bar (Ø 8 mm)
(with spacing 150 mm) at top and bottom also with concrete Fig. 3 Details of slab testing
cover from all sides 25 mm as shown in Fig. 2.

3.2 Testing of the slab specimens circular supports installed with the required span length
(1.75 m) from center to center. A steel plate with width (50
Eight simply supported slab specimens under two line mm) were put over the supports and under two line loading
loads were tested by using hydraulic universal testing with rubber pieces to prevent the crushing of concrete's
machine at the Civil Engineering Laboratory/College of surface. After installation the specimen over the supports, I-
Engineering/ Al-Nahrain University. At first, the solid steel section steel beam was set over the specimen with length (1
 Adel A. Al-Azzawi and Sadeq A. Abed

(a)

3
(b) Fig. 5 Selecting mesh density of sizes (50, 25 and12.5 mm)
Fig. 4(a) Photograph of the mold of hollow-core slab, (b)
Photograph of hollow-core slab testing setup
In the comprehensive sense, any model must comprise all
the nodes, elements, boundary conditions, material
m) to apply the two line load as shown in Figs. 3 and 4. The properties, real constants, and all other features which are
monotonically loading was applied with increased (2.5-5 used to represent the physical system of the current slab
kN) increments and the central deflection is recorded using model (ANSYS User’s Manual 2013).
the dial-gages and the deflection under the loading points
which recorded by the machine. After each increment of 4.1 Element type
loading, the deflection was recorded and monitoring the
bottom face of slab. The cracking and ultimate load at • SOLID65 element was used to model the concrete and
failure stage were recorded with the deflections. It is it has eight nodes with three degrees of freedom at each
noticed that all remarks were recorded during the test and node-translations in the nodal x, y, and z directions. The
the developments of cracks (crack pattern) were marked most important aspect of this element is the treatment of
with a pen at each load increment. nonlinear material properties. The (SOLID65) element has
the capability of plastic deformation, creep, cracking in
three orthogonal directions, and crushing in compression.
4. Material properties and constitutive models • LINK180 element was used to model steel
reinforcement. This element is a 3D spar element and it has
A nonlinear finite element analysis has been carried out two nodes with three degrees of freedom-translations in the
to analyze all tested solid and hollow-core slabs. The nodal x, y, and z directions. Perfect bond between the
analysis was performed by using ANSYS release (15.0) concrete and steel reinforcement considered. However, in
computer program by subprogram ANSYS Parametric the present study the steel reinforcing was connected
Design Language (APDL) for structural analysis problems. between nodes of each adjacent concrete solid element
 Investigation of the behavior of reinforced concrete hollow-core thick slabs

Fig. 6 Convergence of results study (a)

Fig. 7 Meshing of hollow-core slab (dia. 150 mm) (b)

Fig. 8(a) Constraints of plate supports, (b) Distribution of
applied load on nodes
(discrete model), so the two materials shared the same
nodes.
• SOLID185 element was an eight node solid element which formed the reinforcement model was created through
having three degrees of freedom at each node translations in line element connected at the concrete element nodes to
x, y and z directions. The element has capability of plastic form perfect bond between steel bars and concrete
deformation, hyperelasticity, stress stiffening, creep, large elements.
deflection, and large strain. In addition, the element has As an initial and important step in the finite element
mixed nonlinear formulation that can be used for simulating modeling is the selection of the mesh density. A
deformations of nearly incompressible elastoplastic convergence of results is obtained when an adequate
materials and fully incompressible hyperelastic materials. It number of elements are used in a model. The slabs of same
was used for modeling steel supports. Steel plates were material properties, same loading and boundary conditions
added at support and point of loading locations in the finite were modeled with an increasing number of elements for
element models (as in the actual slabs) to provide a more the concrete model. Three types of mesh of reinforced
even stress distribution over the support and point of concrete slab (solid type) are used to find the best mesh size
loading areas. The steel plates were assumed to be linear (2460, 19680 and 157440 elements) for element’s size as
elastic materials. (50, 25, 12.5 mm) respectively as shown in Fig. 5.
• PLANE182 element was used for 2-D modeling of It can be concluded from Fig. 6 that the difference can
solid structures. It was used for the area plane around the be neglected when the number of elements increased from
circular shape of slab cores only. (19680) to (157440), therefore the (19680) model was
adopted in the analysis of all slabs. After creating of
4.2 Modeling and meshing stages volumes and selecting the size of mesh as (25 mm) for each
concrete element, the model for all slabs are divided into
The concrete model that represent the slabs with and (19680) elements by using (SOLID65) element type for
without longitudinal hollow-cores are formulated using concrete and after loading, stresses and strains are
finite elements. The condition of symmetry is not used in calculated at integration points of these small elements as
the present study in order to compare the obtained results shown in Fig. 6..
with the testing results of full experimental slab specimen. The meshing of concrete solid slab was done directly as
The formulation includes drawing areas at plan(x-y) in the cubic elements by (SOLID65) while the meshing of hollow-
first step and subtracts the cores areas and then extrudes core slab with circular and square core shape need fine
them in (z-direction) to form the volume of the slabs. The mesh for the area around the core opening in a way so that
net from longitudinal and transverse reinforcement bars (SOLID65) element can be used for all slabs. The only way
 Adel A. Al-Azzawi and Sadeq A. Abed

(a) (b) (c)

Fig. 8 Load deflection curves for solid slab (SS) with a/d (a)2, (b) 2.5, & (c) 3

(a) (b) (c)

Fig. 9 Load deflection curves for hollow core slab (150 mm) (CCS) with a/d (a) 2, (b) 2.5 & (c) 3

Table 3 Materials properties of slab

Material model
Element type Materials properties
number
Linear isotropic
Young’s modulus 29385.5 MPa
Poisson’s ratio 0.2
Multi-linear isotropic
Point No. Strain Stress (MPa)
Point 1 0.00039 11.43
Point 2 0.001 25.6
Point 3 0.0015 33.1
Point 4 0.002 36.9 Fig. 10 Load deflection curves for hollow-core slab (CCS)
Point 5 0.0026 38.1
Concrete Properties with core diameter (a) 100 mm (b) 75 mm
1 SOLID65
Open crack coefficient 0.8
Closed crack coefficient 0.95
Uniaxial cracking stress 3.40 to do that is by drawing symmetrizes four squares area to
Uniaxial crushing stress 38.1 form a square area greater than the radius or the dimension
Biaxial crushing stress 0 of the core opening then drawing circle or square at the
Hydrostatic pressure 0 center of these square areas and subtract it to form the
Hydro biaxial crushing stress 0 circular or square shape of core opening. The areas were
Hydro uniaxial crushing stress 0 limited between the out square area and voided circular or
Tensile crack factor 0 middle square shape will be meshed using concentrated
Linear isotropic
Young’s modulus 199100 MPa
meshes by (PLANE182) element then extrude it to be a
LINK180 Poisson’s ratio 0.3 volume and then meshing it with (SOLID65) element as
2
(Discrete) Bilinear isotropic shown in Fig. 7.
Yield stress 578MPa
Tang modulus 1991MPa
Linear Isotropic
4.3 Loading and boundary condtions
3 SOLID185 Young’s modulus 2000000 MPa
Poisson’s ratio 0.3 The finite element models were loaded at the same
locations of the tested slabs. In the experiment, the loading
 Investigation of the behavior of reinforced concrete hollow-core thick slabs

Table 4 Test results of slabs

Load (kN) Deflection (mm)
Slab Slab Mode of
Cracking Ultimate Deflection at Deflection at
type no. failure
load load cracking load ultimate load
%
(Pcr) (Pu) (Δcr) (Δu)
Flexural-
Slab-1 2 145.50 293.76 49.5 8.560 64.87
shear
Solid Slab-2 2.5 100.23 232.05 43.1 8.822 56.28 Flexural

Slab-3 3 78.10 196.34 39.8 4.143 33.38 Flexural

Flexural-
Slab-4 2 88.76 269.89 32.9 5.610 54.79
shear
Hollow
Flexural-
core Slab-5 2.5 86.47 184.31 46.9 4.811 88.38
shear
150 mm
Flexural-
Slab-6 3 65.23 147.85 44.1 4.847 71.52
shear
Fig. 11 Deflection contours (UY) of solid slab at Hollow
Flexural-
core Slab-7 2.5 90.21 195.65 46.1 5.03 64.25
ultimate load (a/d=2.5), (Slab-2) 100 mm
shear
Hollow
core Slab-8 2.5 100.02 219.31 45.6 5.84 56.98 Flexural
75 mm

Table 5 Comparison the of experimental and finite element

results
Ultimate load Ultimate deflection
(kN) (mm)
Slab type Slab no.
Exp. FEM Diff. (Pu) Exp. Exp. FEM Diff.
(Pu) (Pu) % (Pu) FEM (∆u) (∆u) %
Slab-1 293.76 310.60 5.73 0.946 64.87 60.43 6.84

Solid Slab-2 232.05 243.21 4.81 0.954 56.28 52.95 5.92

Slab-3 196.34 210.19 7.05 0.934 33.38 33.02 1.08

Slab-4 269.89 288.28 6.81 0.936 54.79 49.69 9.31

Hollow core
Fig. 12 Deflection contours (UY) of hollow core slab 150 mm
Slab-5 184.31 193.00 4.71 0.955 68.38 67.91 0.69

(150 mm), at ultimate load (a/d=2.5), (Slab-5) Slab-6 147.85 160.690 8.68 0.871 71.52 69.87 2.31
Hollow core
Slab-7 195.65 211.50 7.49 0.925 64.25 62.92 2.08
100 mm
Hollow core
Slab-8 219.31 234.00 6.69 0.937 56.98 53.50 6.10
75 mm

steel plate was restrained in X and Y direction (UX=0,

UY=0) with a single line of supports (see Fig. 8(a)).
Two steel plates at loading locations with the same
size of concrete mesh in order to avoid stress concentration
problems. This will provide a more even stress distribution
Fig. 13 Variation of (a/d) ratio with cracking and ultimate over the support area. The external loads were applied on
load for (a) Solid slab, (b) Hollow core slab (Ø 150 mm) two steel plates over the surface of concrete slabs with the
required locations. These loads were applied in the form of
concentrated loads on all top nodes of plates (75 node per
plate) as a ( ) for each node as shown in Fig. 8(b) to
simulate the real loads which adopted in the experimental
work. The application of the loads up to failure was done
incrementally as required by the Modified Newton-Raphson
method. Therefore, the total applied load is divided into a
series of load increments (load step). Within each load step,
maximum of (200) iterations were permitted.

4.4 Model parameters

The finite element models adopted in this study have a

Fig. 14 Variation of core diameter with reduction of number of parameters which can be classified into three
weight and strength categories as shown in Table 3.
1. Concrete property parameters
2. Steel Reinforcement property parameters
and support dimensions were approximately (50×600) mm. 3. Steel Plate property parameters
Two steel plates of (10 mm) thickness are modeled using
(SOLID185) elements, were added at the support. The right
steel plate was restrained in Z-direction (UZ=0) and the left 5. Experimental and numerical results
 Adel A. Al-Azzawi and Sadeq A. Abed

Experimental Experimental

Finite element analysis Finite element analysis

Fig. 15 Crack pattern at ultimate load for (Slab-2) Fig. 18 Crack pattern at ultimate load for (Slab-8)

5.1 Load-deflection curves

The deflection (vertical displacement) in Y-direction

(UY) are obtained at the center of mid span and under the
Experimental load of the bottom face of the slab. The load versus
deflection plots obtained from the numerical and the
experimental study are presented for comparison in Fig. 8 to
Fig. 10. Deflection contour of finite element analyzed slabs
due to the applied loading is shown in Fig. 11 and Fig. 12.
In Fig. 13, the comparison presents between solid and
hollow-core slab with core diameter (150 mm) by variation
(a/d) ratio with cracking and ultimate load. It was concluded
Finite element analysis that increasing (a/d) ratio the cracking load and ultimate
Fig. 16 Crack pattern at ultimate load for (Slab-5) load will be decrease identity for solid slab but the cracking
load in hollow core slab decreases with low rate up to (a/d)
ratio 2.5.
The selection of optimum core diameter presents in Fig.
14 by comparison the core diameter with reduction of
weight and reduction of strength. The curves are
intersecting in optimum core diameter (91 mm) in which
the reduction in weight equals the reduction in strength. The
Experimental obtained optimum core diameter can be used for the slab
scale 1:2 only and it will be change if the scale changed to
1:1 but the ratio of cores area to total section area remains
the same.

5.2 Loads at failure

The test results are presented in Table 4. In this Table,

Finite element analysis the mode of failure for each slab depends on monitoring the
Fig. 17 Crack pattern at ultimate load for (Slab-7) cracking paths during applying the loads until failure stage
and based on which crack is the dominant. A comparison
between the ultimate loads of the experimentally tested
Experimental work and numerical solution (i.e., finite slabs (Pu)Exp at failure stage and the final loads obtained
element method) results which are obtained are presented from finite element models (Pu)FEM is shown in Table 5. The
and then discussed. Experimental measurements which are final loads for the finite element models are the last applied
carried out by testing eight slabs including the ultimate loads before the solution starts to diverge due to numerous
loads, load-deflection curves are given. Comparisons cracks and large deflection in Y-direction (UY). The
between the experimental and the numerical results are ultimate loads obtained from numerical models are in
made to verify the application of the suggested numerical excellent agreement with the corresponding values from the
idealization on the tested slabs. ANSYS program was used experimentally tested slabs. The numerical results show that
to analyze the same slabs and study the effect of some greater ultimate load with smaller deflection at the ultimate
additional parameters on the behavior of these slabs. load stage as compared to experimentally results.
 Investigation of the behavior of reinforced concrete hollow-core thick slabs

Fig. 19 Effect of concrete compressive strength on the Fig. 22 Effect of core shape on the behavior of hollow
behavior of hollow core slab with 150 mm core size core slab

Fig. 20 Effect of core size on the behavior of hollow Fig. 23 Effect of loading Type on the behavior of
core slab with square core shape hollow core slab

Fig. 21 Effect of core size on the behavior of hollow Fig. 24 Effect of removing top steel reinforcement on
core slab with circular core shape the behavior of hollow core slab

5.3 Crack pattern or crushing in concrete elements. Fig. 15 to Fig. 18 show

the location of cracks from experimental test and finite
As expected, the main cracks for all tested slabs element analysis along the solid and hollow-core slabs with
commenced at middle third (flexural cracks) from the core diameter (150, 100 and 75) respectively.
bottom face as a line with the width of the slab and all slabs
exhibited ductile flexural failure. It was noted that some of
shear cracks appeared in slabs which tested with (a/d) ratios 6. Parametric study
(2 and 2.5) especially in hollow-core slab due to position of
applied loads and presence the holes of hollow-core slab but The parameters that may affect the behavior of the
the failure of these slabs are flexural failure. ANSYS hollow-core reinforced concrete slabs under the same
computer program displays circles at locations of cracking conditions and not tested are studied here in:
 Adel A. Al-Azzawi and Sadeq A. Abed

6.1 Effect of compressive strength of concrete increasing the diameter of the cores and also with
increasing the ratio of shear span to effective depth (a/d).
The reinforced concrete hollow-core slab having core • Reducing the own weight of the moderate thick
diameter (150 mm) under (a/d) equal 2.5 was selected to reinforced concrete slabs by about 23.6% with longitudinal
study the influence of the grade of concrete on the behavior hollow cores (dia.=150 mm) lead to reduce the ultimate
of load-deflection curve. It has been reanalyzed using strength by about 20.6% while reducing the weight by about
different values of concrete compressive strength as (25, 38, 15.71% with hollow cores (dia.=100 mm) lead to reduce the
and 45) MPa respectively (This means increasing the ultimate strength by about 15.68% and reducing the weight
stiffness of concrete). The Fig. 19 shows the response of the by about 8.84% with hollow cores (dia.=75 mm) lead to
considered R.C hollow-core slab for different concrete reduce the ultimate strength by about 5.49%.
compressive strength. • Increasing the ratio (a/d) from 2 to 3 lead to reduce the
ultimate strength by about 33% in solid slab with
6.2 Effect of shape and size of hollow cores decreasing the deflection. While 45% reduction in ultimate
load fore hollow-core slab with increasing the deflection
The hollow core slabs having circular core shapes and due to reducing the stiffness of slab with removing the
square core shapes were analyzed with equivalent areas of concrete volume of hollow cores.
three sizes for each core shape under same loading and • In hollow-core reinforced concrete slab, the circular
support conditions. The circular shape types were presented core shapes have cracking and ultimate strength greater than
and analyzed previously while the equivalent square core the square shape by about 13.4% and increasing in
shapes having dimensions (133×133 mm, 89×89 mm and deflection by about 39.5%. The increase in core size for
66×66 mm). The analysis results (load-deflection) curves circular core shape caused a reduction in ultimate strength
and comparison with sizes and shapes of hollow cores were with increasing the deflections while increasing the core
presented at Figs. 20, 21 and 22. size in square core shape cause a reduction in ultimate
strength with reducing deflections.
6.3 Effect of loading type • When the compressive strength of the concrete
increases from (38MPa) to (48MPa), the ultimate strength
The reinforced concrete hollow-core slab with circular increase by 23.6% and when the compressive strength
core (diameter 150 mm) was analyzed under uniform load decrease from (38MPa) to (25MPa), the ultimate strength
on all surface nodes of the concrete elements by ANSYS reduces to about 28.7%.
program. The analysis results compared with the results of • It was found that the ultimate strength of the hollow-
the same slab under two point loads with the ratio (a/d=2.5). core slab increases by about 80% for the case of uniform
The results indicate that the ultimate total load of this slab load and a reduction in deflection by (28.5%) compared to
was increased from (193 kN) for two point loads to (347 two point loads with (a/d) equal 2.5.
kN) under uniform load type with decreasing in deflection • It was concluded that removing the top steel
by about (28.5%) as shown in Fig. 23. It can be seen from reinforcement in hollow-core slab reduces the ultimate
this figure that the polyline appeared clearly in load- strength by about 28% due to crushing failure of top flange
deflection curve for load values between 120 kN and 160 of concrete so that it is recommended to use this layer
kN which refers to the transformation stage of numerical prevent this failure.
solution from linear to non-linear behavior (cracking occur). • Results of comparison between experimental and finite
element results show that the difference range was (4.71-
6.4 Effect of top layer of reinforcement 8.68)% in ultimate load and (0.69-9.31)% in deflection.
• It is recommended that the optimum core diameter in
Fig. 24 shows the effect of using top steel reinforcement hollow-core slab for scale 1:2 is (91mm) because of that the
with bottom one. The hollow-core slab with core diameter reduction in weight and strength are equal. In addition,
(150 mm) under two-point loading with (a/d) equal 2.5 was reducing the core diameter will increase the ribs and that
analyzed first with top and bottom reinforcement as cause increasing in strength of the slab.
presented before then analyzing the same slab with
removing the top reinforcement. It was noted that the
ultimate load capacity of the slab will decreased by about References
28% with removing the top reinforcement. This is may be
due to the existence of top reinforcement will distribute the ACI318-14 (2014), Building Code Requirements for Structural
stresses around the core and prevent crushing. Concrete (ACI 318-14) and Commentary, Detroit, U.S.A.
ANSYS 15.0 Inc (2013), ANSYS User’s Manual, SAS IP, Inc.,
Version 15.0, U.S.A.
Brunesi, E., Bolognini, D. and Nascimbene, R. (2014),
7. Conclusions “Evaluation of the shear of precast-prestressed hollow core
slab: Numerical and experimental comparison”, Mater. Struct.,
Based on the analysis results of the experimental and 48(5), 1503-1521.
numerical investigations of the solid and hollow-core slabs, Chang, J., Buchanan, A.H., Dhakal, R.P. and Moss, P.J. (2008),
the following conclusions can be drawn: “Simple method for modeling hollow core concrete slabs under
• The cracking and ultimate strength of the moderately fire”, University of Canterbury, New Zealand.
thick solid and hollow-core slab lead to be decreased with Hai-tao, L., Deeks, A., Liu, L., Huang, D. and Su, X. (2011),
 Investigation of the behavior of reinforced concrete hollow-core thick slabs

“Moment transfer factors for column-supported cast-in-situ

hollow core slabs”, J. Zhejiang Univ., 13(3), 165-173.
Haruna S.I. (2014), “Flexural behavior of precast pre-stressed
concrete hollow core slabs with cast-in-place concrete topping”,
M.S. Dissertation, Atılım University, Turkey.
Lee, D.H., Park, M., Oh, J., Kim, K.S., Im, J. and Seo, S.
(2014), “Web-shear capacity of prestressed hollow core slab
unit with consideration on the minimum shear reinforcement
requirement”, Comput. Concrete, 14(3), 211-231.
Pajari, M. (2004), Pure Torsion Tests on Single Hollow Core
Slabs, Espoo VTT Tiedotteita, Research Notes 2273, 29-28.
Rahman, M.K., Baluch, M.H., Said, M.K. and Shazali, M.A.
(2012), “Flexural and shear strength of pre-stressed precast
hollow-core slabs”, Arab. J. Sci. Eng., 37(2), 443-455.

CC

View publication stats