You are on page 1of 13

Alexandria Engineering Journal (2016) xxx, xxxxxx

H O S T E D BY
Alexandria University

Alexandria Engineering Journal


www.elsevier.com/locate/aej
www.sciencedirect.com

ORIGINAL ARTICLE

Prevention of premature failures of plate bonded


flexurally strengthened RC slab using end anchor
and connector
Md. Ashraful Alam *, Waleed Mohammed, Suliman Bakkar, Salmia Beddu

Department of Civil Engineering, College of Engineering, Universiti Tenaga Nasional, Malaysia

Received 24 February 2016; revised 7 September 2016; accepted 24 September 2016

KEYWORDS Abstract Flexural strengthening of reinforced concrete (RC) slab using externally bonded plate is
RC slab; found to be the most common and popular in retrofitting of deficient structure. However, prema-
Strengthening; ture failures due to plate end debonding of steel plate and shear could be the major drawbacks of
Debonding; the system to obtain the highest flexural capacities. The aim of this research was to propose a com-
Premature shear; prehensive strengthening technique to eliminate premature failures of flexurally strengthened RC
End anchor; slab. In the experimental programme, five full scale RC slabs including un-strengthened control slab
Embedded connector and flexurally strengthened slabs using steel plates were fabricated and tested. The strengthened
slabs were further strengthened with end anchor and embedded steel bar connector to eliminate pre-
mature end peeling and shear failure respectively. Theoretical and numerical models were also pro-
vided to design and to analyse the structural behaviour of steel plate strengthened slabs. Results
showed that end anchor and embedded connectors prevented premature end peeling and shear fail-
ure of flexurally strengthened RC slabs completely and allowed the slab to fail by flexure with the
maximum strength and ductility, whereas, their corresponding strengthened slabs without end
anchor and embedded connector failed with premature plate end debonding and shear respectively.
Both premature failures were found to be brittle in nature. The proposed theoretical model pre-
dicted the capacities of strengthened slabs satisfactorily. The results based on the numerical analysis
were found to be comparable with the experimental findings.
2016 Production and hosting by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria
University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).

1. Introduction

Strengthening of reinforced concrete slab for flexure is crucial


to enhance the capacity of deficient slab. The slab could be
* Corresponding author. deficient because of underdesign or lack of quality control.
E-mail address: ashraful@uniten.edu.my (M.A. Alam). Furthermore the existing slab might be required to strengthen
Peer review under responsibility of Faculty of Engineering, Alexandria due to increase of design action in the design code or changes
University. of utility of floor. In terms of economy and environmental
http://dx.doi.org/10.1016/j.aej.2016.09.018
1110-0168 2016 Production and hosting by Elsevier B.V. on behalf of Faculty of Engineering, Alexandria University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
2 M.A. Alam et al.

Nomenclature

d effective depth of slab VRd;c shear force resisted by concrete


fbd bond strength of concrete q steel ratio
Ms moment resisting capacity of beam Mpt moment capacity of strengthened slab based on
As cross sectional area of flexural bar tensile strength
fyk yield strength of flexural bar ftk tensile strength of flexural steel bar
fck concrete compressive strength based on cylinder ftk;p tensile strength of steel plate
test Vp shear strength of flexurally strengthened slab
b width of slab Vec shear force need to be resisted by connector
z moment arm fbd bond strength of concrete
x depth of neutral axis r radius of connector
MRC moment resisted by concrete Nc number of connector for single row
Md design moment s spacing of connector
Mp moment need to be resisted by steel plate Mcs failure moment of control slab
t slab thickness Mss failure moment of strengthened slab
Ap required cross sectional area of steel plate Vss shear failure load of strengthened slab with con-
fyk;p yield strength of steel plate nector
Vmax maximum shear resisting capacity of slab

aspects, strengthening of slab is the most acceptable way rather 2. Proposed design model for flexural strengthening of RC slab
than demolish. A significant number of research works have using externally bonded steel plate
been conducted over last decade on strengthening of reinforced
concrete slab using various methods to enhance flexural capac- 2.1. Flexural capacity of un-strengthened control slab
ities of deficient slabs [111]. Amongst the existing methods,
externally bonded plate using steel plate and carbon fibre lam-
The flexural capacity of control slab based on yielding of steel
inate is found to be the most effective and popular method to
bar could be theoretically predicted using Eq. (1) of EC2
strengthen the slab for flexure. It has been investigated that the
(2004):
plate bonding method could achieve the maximum capacity of  
strengthened slab. However, premature plate end debonding 0:588As fyk
Ms As fyk d  1
failure was found to be the major weakness of this method fck b
to obtain the highest flexural strength [12]. In most of the cases
where,
the debonding of plate caused premature failure which was
brittle and catastrophic in nature. As fyk As fyk
x 2
Researchers investigated that because of the discontinuity 0:85fck 0:8b 0:68fck b
of the plate, excessive shear and normal stresses developed at
 
end of the plate which caused cracks [13,14]. Finally the cracks 0:588As fyk
z d  0:4x d  3
lead to debond the plate either at the level of concrete adhesive fck b
interface or at the level of internal flexural reinforcement.
Since, this debonding failure was found to be premature and
brittle in nature, elimination of the plate end debonding failure 2.2. Design flexural strength of strengthened slab
is vital to obtain the full strength and ductility of flexurally
strengthened RC slab. Furthermore, in general the slab is The slab could resist the moment until it fails by crushing of
not critical for shear; however, once the slab is being strength- concrete. Thus, the maximum moment resisting capacity of
ened for flexure, it could fail by premature shear. In that case, slab is as follows:
strengthening of slab for shear would be crucial to avoid pre-
MRC Cz 0:85fck b0:8xd  0:4x 4
mature shear failure to obtain the maximum flexural capacity
of strengthened slab. Although the structural behaviour of As per EC2, the maximum depth of neutral axis, x 0:45d.
flexurally strengthened RC slab using externally bonded plates Thus,
is well known, solutions on preventing those premature fail-
ures are seldom found. The main aim of this research work MRC Cz 0:85fck b0:80:45dd  0:40:45d
was to investigate the performances of end anchor and embed- 0:251fck bd2 5
ded connector to eliminate premature plate end debonding and
shear failures of flexurally strengthened RC slab. Comprehen- If Ms < MRC , in that case the slab can be strengthened for
sive theoretical model is proposed to design and analyse the flexure.
slab for flexural and shear strengthening using externally Since, slab could resist moment as long as it fails by
bonded steel plate and embedded connector respectively. crushing of concrete; thus, the maximum moment resisted by
Finally, the slab specimens were modelled using ABAQUS to concrete could be used as the design moment of strengthened
investigate the structural behaviour numerically. slab.

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
Prevention of premature failures of RC slab 3

Table 1 Details of specimens.


Specimen Strengthening plate End anchor plate Connector in shear span
ID
Length Width Thickness Number of Width Length Diameter Spacing
(mm) (mm) (mm) plates (mm) (mm) (mm) (mm)
CS
SS 1200 50 2.7 2
SS-C 1200 50 2.7 2 16 (8) 100
SS-A 1200 50 2.7 2 100 150 16 (8)
SS-A-C 1200 50 2.7 2 100 150 16 (8) 100

3. Proposed design model for shear strengthening of flexurally


strengthened RC slab using embedded connector

3.1. Requirements of embedded connector to prevent premature


shear failure

Maximum shear force on strengthened slab is


MRC 0:251fck bd2
Vmax 10
Shear span Shear span
Shear resisting capacity of slab is
VRd;c 0:12k100qfck 1=3 bd 11
where,
p
Figure 1 Reinforcement details of slab specimens. k 1 200=d 6 2:0 with d expressed in mm 12

As;p
q 6 0:02 13
bd
2.3. Required dimension of steel plate for flexural strengthening
If Vmax > VRd;c , the slab will fail by premature shear. In that
of RC slab
case the slab needs to be strengthened for shear using embed-
ded connector.
Moment needs to be resisted by strengthening plate:
  3.2. Design of flexurally strengthened slab for shear using
0:588As fyk
Mp Md  Ms 0:251fck bd2  As fyk d  6 embedded connector
fck b

Tensile force of strengthening plate is Moment capacity of flexurally strengthened slab is


n h io Mpt Tst zst Tpt zpt
0:588A f
0:251fck bd2  As fyk d  fck bs yk
Tp 7 As ftk d  0:4x Ap ftk;p t  0:4x 14
t  0:18d
where,
where,
As ftk Ap ftk;p As ftk Ap ftk;p
x 15
zp t  0:40:45d t  0:18d 8 0:85fck b0:8 0:68fck b
Required cross sectional area of steel plate is Shear force of flexurally strengthened slab is
h i h i
n h io As ftk d 
0:588As ftk Ap ftk;p
As ftk t 
0:588As ftk Ap ftk;p
0:588A f
0:251fck bd2  As fyk d  fck bs yk Vp
fck b fck b

Ap 9 Shear span
fyk;p t  0:18d
16

Table 2 Properties of materials.


Materials Size (mm) Yield strength (MPa) Tensile strength (MPa) Compressive strength, fcu (MPa)
Flexural steel bar 8 (7.1) 420 500
Flexural steel plate 2.7  50  1200 275 320
Anchor steel plate 2.7  100  150 275 320
Steel bar connector 8 (7.1) 420 500
Concrete 20 (16)

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
4 M.A. Alam et al.

2.7 mm x 100 mm x 150 mm steel anchor Total shear resisting capacity of embedded connector is
16 mm diameter connector    
dcot45 d
2.7 mm x 50 mm x 1200 mm steel plate Vec fbd t2prNc fbd t2pr fbd t2pr 19
s s
140 mm x 500 mm x 1500 mm concrete slab
Required dimension of connector (single row) is
r V  V
p Rd;c
20
s 2ptfbd d

4. Experimental investigations

4.1. Description of specimens


Support Concrete slab
Connector Adhesive
End anchor Steel plate
In this research, five reinforced concrete slabs with the dimen-
sions of 1500 mm  500 mm  140 mm were fabricated. Slab
CS was prepared as un-strengthened control, while slabs SS,
SS-A, SS-C and SS-A-C were flexurally strengthened using
externally bonded steel plates. Two steel plates with the dimen-
100 1000 100 sions of 1200 mm  50 mm  2.7 mm were used for flexural
mm mm mm strengthening of each slab. Based on the proposed equation
1300 mm (Eq. (9)), the required width of plate was 100 mm, however,
to confirm the sufficient ductility of slab, 50 mm width of plate
Figure 2 Details of strengthened slab. was used for strengthening. Steel anchor plate having the
dimensions of 100 mm  150 mm  2.7 mm was used at both
ends of the flexural steel plates of slabs SS-A and SS-A-C to
Shear force needs to be resisted by embedded connector
eliminate premature plate end debonding failure. Special
Vec Vp  VRd;c 17 embedded connectors were inserted in shear span of end
anchored flexural strengthened slab (SS-A-C) to eliminate pre-
Shear resisting capacity of single embedded connector:
mature shear failure. The dimension and spacing of connector
vec fbd t2pr 18 were obtained from proposed design guideline (Eq. (20)).

(a) End anchored strengthened slab (b) Connector

Figure 3 Strengthening of slab using end anchor and embedded connector.

Figure 4 Instrumentation and test setup.

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
Prevention of premature failures of RC slab 5

(a) Control slab (CS)

(b) Strengthened slab without end anchor (SS)

(c) Strengthened slab with end anchor (SS-A)

(d) Strengthened slab with embedded connector at end of plate (SS-C)

(e) Strengthened slab with embedded connector and end anchor (SS-A-C)

Figure 5 Failure mode of slab specimens.

Embedded connectors were also provided at end of plates in tension face of each of the slab for flexural strengthening using
slab SS-C. The details of all specimens are shown in Table 1. Sikadur-30 normal adhesive. The plates were positioned prop-
erly and were forced gently on the plates until the adhesive
4.2. Fabrication of specimens and properties of materials came out from both sides of the plates to eliminate entrapped
air. After fixing the plates the specimens were left seven days
The slabs were fabricated using four 8 mm diameter steel bars for curing.
as main flexural reinforcement as shown in Fig. 1. It had
420 MPa yield and 500 MPa tensile strengths. All slabs were 4.4. Shear strengthening of slab using embedded connector
cast with 20 (cube strength) grade concrete. The steel plates
that were used for flexural strengthening and anchoring had To eliminate premature shear failure of flexurally strengthened
275 MPa yield and 320 MPa tensile strengths respectively. slab, steel bar embedded connectors were used in shear span of
The flexural reinforcement (8 mm) was also used to prepare slab. Before fixing the plates, the slab was drilled for 16 mm
the embedded connector. The bar with 140 mm length was hole with the spacing of 100 mm centre as shown in
inserted in 16 mm diameter adhesive filled embedded connec- Figs. 2 and 3(b). The holes of the connectors were properly
tor. The properties of materials are summarized in Table 2. cleaned to remove dust and loose particles. The holes were
then filled by Sikadur-30 adhesive and 8 mm steel bars were
4.3. Flexural strengthening of slabs inserted into the adhesive filled holes. The connectors were
fixed prior to fix the steel plates and end anchors.
Before strengthening of slabs SS, SS-A, SS-C and SS-A-C, the
greasy and loose particles were removed from the bonding 4.5. End anchoring
faces of the concrete using scaling hammer. The dust and loose
particles were finally removed by compressed air. In addition, In slabs SS-A and SS-A-C, the flexural steel plates were end
the bonding faces of steel plates were sand blasted according to anchored using 2.7 mm  100 mm  150 mm steel plates as
Swedish standard Sa  2.5 to make sure the accurate bond shown in Figs. 2 and 3. The dimensions of steel anchor plates
of plates can be achieved. Once the surfaces of steel plate were arbitrarily chosen. The surfaces of concrete to fix the
and concrete had been prepared, two steel plates with the anchor plates were scrubbing using scaling hammer to expose
dimension of 2.7 mm  50 mm  1200 mm were fixed on the the coarse aggregate for strong bonding between anchor plate

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
6 M.A. Alam et al.

(a) Load vs mid span deflection (b) Load vs flexural bar strain

(c) Load vs concrete compressive strain (d) Load vs flexural plate strain

Figure 6 Effects of end anchor on strengthened slabs.

Table 3 Experimental results.


Specimen First crack First crack load Failure Failure load over Steel bar Plate Concrete Mode of failure
load (kN) over control slab load the control slab (%) strain (l) strain compressive
(%) (kN) (l) strain (l)
CS 18.6 36.2 1321.9 981.3 Flexural
SS 33 77 55.6 53 649.9 1036.6 392.2 Debonding
followed by
shear
SS-C 32 72 62 71 730 1225 620 Debonding
followed by
shear
SS-A 35 88 64.2 77 1070.6 1293.2 600.8 Flexural
followed by
shear
SS-A-C 38 104 75.5 109 1266.6 1504 1093 Flexural

and concrete. The surface of steel anchor plate was sand Strain gauges were also fixed on the strengthening steel plates
blasted using same standard of flexural plate. The anchor and at top of the slab to record the tensile strain of strengthen-
plates were then fixed on the flexural plates using Sikadur-30 ing plate and compressive strain of concrete respectively.
adhesive. LVDT was used at mid span of slab for measuring mid-span
deflection. All specimens were tested under four points loading
4.6. Instrumentation and test procedure with the shear span of 450 mm as shown in Fig. 4. The support
distance of the slab was 1300 mm. The loads were applied
Before casting of concrete, two strain gauges were fixed on the through the loading frame using hydraulic jack. The Data Log-
flexural reinforcements of each slab to record strain of bar. ger recorded all the required data for analysis of the results.

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
Prevention of premature failures of RC slab 7

5. Experimental results had shown elastic behaviour as shown in Fig. 6(b) and (d),
the deflection of that slab was also found to be elastic. It
5.1. Effects of end anchors was observed that the end anchored slab failed at shear span
rather than mid-span before yielding of bar and steel plate at
mid span of slab. It could be the reason to have elastic mid-
Strengthened slab with end anchors had failed with shear with-
span deflection behaviour of end anchored slab.
out debonding of steel plate as shown in Fig. 5(c). It had
Result also showed that slab with end anchor had 64.2 kN
shown flexural failure followed by shear failure, whereas, un-
failure load which was found to be 15.5% and 77% higher as
anchored strengthened slabs had shown debonding of plate
compared to strengthened slab without end anchor and un-
followed by shear failure (Fig. 5(b)). The end anchored
strengthened control slab respectively (Table 3). Since, end
strengthened slab failed by shear because of enhancement of
anchored prevented premature debonding of steel plate, had
flexural capacity. The failure mode of that slab was found to
shown higher failure load. The ductility factor of end anchored
be bit ductile in nature rather than brittle and catastrophic fail-
strengthened slab was higher as compared to that of un-
ure, whereas, un-anchored strengthened slab had shown brittle
anchored slab. It was noticed that the number of cracks of
and catastrophic failure. The mid span deflection of end
end anchored slab was higher than un-anchored slab. Since,
anchored strengthened slab was found to be elastic in nature
the number of crack increased, the crack width of end
as shown in Fig. 6(a). Since, the flexural bar and steel plate

80
80
70 70

60 60

Load (kN)
Load (kN)

50 50
CS
40 40
SS-C
30 30
SS-A CS
20 20
SS-A-C SS-C
10 10 SS-A
SS-A-C
0 0
0 10 20 30 40 0 500 1000 1500
Deflection (mm) Bar strain (micro)
(a) Load vs mid span deflection (b) Load vs flexural bar strain

80 80
70 70
60 60
Load (kN)

Load (kN)

50 50
40 40
30 30 SS-C
CS
20 SS-C 20 SS-A
10 SS-A 10 SS-A-C
SS-A-C
0 0
0 200 400 600 800 1000 1200 0 400 800 1200 1600
Concrete strain (micro) Plate strain (micro)
(c) Load vs concrete compressive strain (d) Load vs flexural plate strain

Figure 7 Effects of embedded connector on strengthened slab.

Table 4 Comparison between theoretical and experimental results.


Specimens Flexural failure load (kN) Shear failure load (kN) Mode of failure
Expt Theory Expt Theory Expt Theory
CS 36.2 33.5 62 Flexural Flexural
SS 78 55.6 62 Debonding followed by shear Flexural
SS-C 62 78 62 62 Debonding followed by shear Flexural
SS-A 78 64.2 62 Flexural followed by shear Flexural
SS-A-C 75.5 78 100 Flexural Flexural and shear

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
8 M.A. Alam et al.

Figure 8 Details of reinforcement, steel plates, anchor and embedded connector of specimens in numerical model.

Figure 9 Mesh distribution of strengthened slab with plate and end anchor in numerical model.

anchored strengthened slab was reduced. The end anchored nectors and control slab respectively as shown in Table 3. The
slab had higher concrete strain at failure load as compared connectors assisted to have the highest strength of strength-
to un-anchored slab as shown in Fig. 6(c). ened slab as shown in Fig. 7(a) and (b). Fig. 7(d) shows that
at failure load the strain of flexural steel plate of embedded
5.2. Effects of embedded connector to eliminate premature shear connector strengthened slab exceeded the yield strength of
failure of flexurally strengthened slab plate, which indicated the maximum capacity of the slab. Fur-
thermore, before failure, the concrete strain of slab with
Results showed that the embedded connectors completely pre- embedded connector was found to be higher as compared to
vented premature shear failure of strengthened slab as shown others (Fig. 7(c)). It could be due to higher capacity of embed-
in Fig. 5(e). The connectors increased the shear resisting capac- ded connector strengthened slab. Although the concrete strain
ity of flexurally strengthened slab; thus, it failed by flexure of that particular slab was found to be higher at failure load, it
rather than premature shear. The failure mode of that slab was found to be lower as compared to the crushing strain of
was found to be ductile in nature. The strengthened slab with concrete (3500 micro). It could be due to under-design of slab
embedded connector had shown higher deflection after around rather than balanced design. Moreover, it was noticed that the
68 kN load as shown in Fig. 7(a). It could be due to yielding of slab (SS-A-C) failed at loading point rather than mid-span
steel bar and steel plate as shown in Fig. 7(b) and (d) respec- which had effects on mid-span concrete strain as well.
tively. Since, the bar and plate yielded before failure of the
slab, had shown higher ductility, whereas, end anchored 5.3. Comparison between theoretical predictions with
strengthened slab without embedded connector showed brittle experimental results
premature shear failure before yielding of steel bar and plate
(Fig. 7(b) and (d) respectively). The theoretical flexural capacities of un-strengthened control
The embedded connectors enhanced 10% and 109% slab and flexurally strengthened RC slab using externally
strength of slab as compared to strengthened slab without con- bonded steel plate could be obtained based on the proposed

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
Prevention of premature failures of RC slab 9

theoretical model as shown in Eqs. (21) and (22) respectively,

ABAQUS
whereas, Eqs. (23) and (24) are proposed to predict the shear

19,769
capacity of flexurally strengthened slab and shear strengthened

3510
892

939
slab with embedded connector. All the equations of this


Plate strain (l)
research are based on EC2.

Experimental
 
0:588As ftk
Mcs As ftk d  21
fck b

1036.6

1293.2
1225

1504
 


0:588As ftk Ao ftk;p
Mss As ftk d 

ABAQUS
fck b
 

24,560

12,756
0:588As ftk Ap ftk;p

1461
2182
As ftk t  22

935
Steel bar strain (l)
fck b

Experimental
Vss 0:12k100qfck 1=3 bd 23
 

1321.9

1070.6
1266.6
649.9
d

730
Vss 2fbd t2pr 0:12k100qfck 1=3 bd 24
s
The theoretical flexural failure load of control slab was
found 8% lower as compared to those of experiment shown
in Table 4. The shear capacity of control slab was higher as

Failure load experiment/


compared to experimental flexural failure load, and thus, did
not show shear failure. The experimental failure load of
strengthened slab without end anchor was 40% and 11.5%
lower as compared to theoretical flexural and shear capacities

ABAQUS
respectively. Since, that slab failed by debonding of steel plate,
it had shown lower failure load compared to theoretical predic-

1.03
0.77
0.85
0.96
0.94
tions. Results also showed that the experimental failure load of
end anchored strengthened slab was close to theoretical shear

ABAQUS
capacity rather than flexural capacity as shown in Table 4,
whereas, the experimental failure load of embedded connector
slab was found to be almost similar with the theoretical flexu-
35
72
73
67
80
Failure load (kN)

ral failure load. The slab with connector failed by flexure thus
Experimental

showed similar flexural failure load of theoretical prediction.


The shear capacity of embedded connector slab was 32%
higher than experimental flexural failure load, thus, did not
36.2
55.6

64.2
75.5
show premature shear failure.
62
Comparison between numerical and experimental results.

6. Numerical modelling

A three-dimensional nonlinear finite element model was devel-


First crack experiment/

oped using ABAQUS/CAE to analyse the specimens numeri-


cally and to compare the results with experimental findings.
Several constitutive models were developed and analysed and
the most appropriate ones were selected to develop the finite
ABAQUS

element model. The details of modelling procedures used in


this research are briefly illustrated in the following subsections.
1.55
1.65

1.59
1.58
1.6

6.1. Discretization
ABAQUS
First crack load (kN)

Three-dimensional solid elements were employed for the rein-


12
20
20
22
24

forced concrete slab, whereas the strengthening steel plates and


Experimental

anchors were represented using three-dimensional shell ele-


ments (Fig. 8) as recommended by ABAQUS/CAE users
guide. Mesh density was investigated for the concrete and
18.6

the mesh size of 30 mm was chosen in each direction as shown


33
32
35
38

in Fig. 9. In addition, the selected mesh size maintains a bal-


Specimen

ance between accuracy and computational time. Flexural steel


Table 5

SS-A-C

bars modelled as solid elements were employed to represent the


SS-A
SS-C
CS
SS

reinforcement bars in both directions, and connectors were

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
10 M.A. Alam et al.

(a) Interface normal stress (b) Interface shear stress


6 1

5 0.8
SS-A

Shear Stress (MPa)


4
0.6

Stress (MPa)
SS-A SS
3
SS 0.4
2
0.2
1

0 0
0 200 400 600 800 0 200 40 0 600 800
-1 -0.2
Distance from Support (mm) Distance from Support (mm)

Figure 10 Interfacial normal and shear stress (numerical) along the length of soffit plate.

2nd Connector
150
Slab Depth (mm)

100

50 SS-A-C
SS-A
0
-0.5 0 0.5 1
Concrete Shear Stress (MPa) at 67
kN

Figure 11 Concrete shear stress (numerical) at 2nd connector, 250 mm from edge of slab (comparison between SS-A and SS-A-C).

SS-A-C 1st SS-A-C 2nd SS-A-C 3rd SS-A-C 4th


Connector Connector Conncector Connector
160 160 160 160
140 140 140 140
Slab Depth (mm)
Slab DEpth (mm)

Slab Depth (mm)


Slab Depth (mm)

120 120 120 120


100 100 100 100
80 80 80 80
60 60 60 60
40 40 40 40
20 20 20 20
0 0 0 0
-0.5 0 0.5 1 -0.5 0 0.5 1 0 1 -2 0 2 4
Shear Stress (MPa) Shear Stress (MPa) Shear Stress (MPa) Shear Stress (MPa)

Shear stress at Shear Stress at Shear Stress at Shear Stress at


Concrete Concrete Concrete Concrete
Shear Stress at Steel Shear Stress at Steel Shear Stress at Steel Shear Stress at Steel

Figure 12 Shear stress comparison (numerical) between steel connector and concrete for SS-A-C at 67 kN.

modelled similar to reinforcement bars as solid elements. The between steel plates and anchor with the concrete surface of
reinforcement elements were embedded in the concrete using the slab along with connectors and surrounding concrete. Tie
embedded region constrain, and its strains are computed constraints are used to tie two surfaces together for the dura-
from the displacement field of the concrete elements surround- tion of a simulation. For a structural analysis this means the
ing it. In this study, perfect bond was assumed between the translational degrees of freedom are constrained. In this
internal steel reinforcement and the surrounding concrete as research, the steel plates, anchors, and connectors acted as
bond failure had not been observed during the experimental the master surfaces while the concrete surface of the slab acted
tests. Tie constrains were utilized to model the bond area as the slave surface.

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
Prevention of premature failures of RC slab 11

100
60
50 80

Load (kN)

Load (kN)
40 60
30 CS-ABAQUS SS-C-ABAQUS
SS-ABAQUS 40
20 SS-A-ABAQUS SS-A-C-ABAQUS
CS-Exp SS-A-C-Exp
10 20
SS-Exp SS-C-Exp
SS-A-Exp
0 0
0 5 10 15 0 10 20 30
Deflection (mm) Deflection (mm)
(a) Load vs mid span deflection (CS, (b) Load vs mid span deflection (SS-C,
SS, SS-A) SS-A-C)

80
CS-ABAQUS 100
70 SS-ABAQUS
60 SS-A-ABAQUS 80
CS-Exp
Load (kN)

50

Load (kN)
SS-Exp
SS-A-Exp 60
40
SS-C-ABAQUS
30 40 SS-A-C-ABAQUS
20 SS-A-C-Exp
20
10 SS-C-Exp
0 0
0 2000 4000 0 5000 10000 15000
Bar strain (micro) Bar strain (micro)
(c) Load vs flexural bar strain (CS, SS, (d) Load vs flexural bar strain (SS-C,
SS-A) SS-A-C)

70 80
60 70
60
50
Load (kN)

Load (kN)

50
40
40
30 30
SS-ABAQUS SS-C-ABAQUS
20 SS-A-ABAQUS 20 SS-A-C-ABAQUS
SS-Exp SS-A-C-Exp
10 SS-A-Exp 10
SS-C-Exp
0 0
0 400 800 1200 1600 0 400 800 1200 1600
Plate strain (micro) Plate strain (micro)
(e) Load vs flexural plate strain (SS, (f) Load vs flexural plate strain (SS-
SS-A) C, SS-A-C)

Figure 13 Comparison between experimental and numerical results.

6.2. Modelling of material 6.3. Boundary conditions and solution algorithm

A damaged plasticity model for concrete available in ABA- A mechanical Displacement/Rotation boundary condition is
QUS/CAE served as the basis for the constitutive model of used in ABAQUS/CAE to model the supports of the RC slab.
the reinforced concrete slabs. This model uses concepts of iso- The slab was designed to be simply supported with one side
tropic compressive and tensile plasticity in combination with pinned by restricting its movement on the Y and Z directions
scalar damage to describe irreversible reductions in stiffness and the other side as a roller support by restricting its move-
during the fracturing process; these values were obtained from ment on the Y direction, where the supports are 100 mm from
the compressive and tensile tests in the laboratory. The elastic the edge of the slab. In addition, the whole slab is restricted to
modulus and Poissons ratio are the two basic parameters to move on the X axis to avoid any displacement in that direction.
define the elastic mechanical behaviour of concrete. An A pressure load is used to simulate the two point loads on the
elastic-perfectly plastic stressstrain model was adopted for RC slab as it is the most similar to the real loading in the lab-
the internal steel reinforcement, steel plates, anchors, as well oratory. The load is applied proportionally by ABAQUS/CAE
as connectors. For steel reinforcement and connector a to simulate the actual load increase. An appropriate
420 MPa yield stress is specified, while for steel plates and incremental-iterative procedure was adopted to solve the non-
anchor have yield stress of 275 MPa and Poissons ratio is 0.3. linear equations. The vertical loads were applied as pressure

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
12 M.A. Alam et al.

load, and for each increment, general static analysis step was of strengthened slab with embedded connector was found to
implemented with 1 s of time and 0.01 as initial increment. be 109% and 10% higher as compared to control slab and
Convergence was successfully achieved at the end of each load end anchored slab without connector respectively. The end
step using this procedure. anchor could prevent premature debonding of steel plate and
had shown ductile mode of failure rather than brittle. It had
7. Numerical results enhanced 77% and 15.5% ultimate flexural capacity of slab
as compared to un-strengthened control and strengthened slab
The results of specimens obtained from numerical analysis are without end anchors respectively. In general, steel plate
summarized in Table 5. The numerical model predicted crack- strengthened slabs had shown better structural performances
ing loads, failure loads, deflection and strain or stress beha- in terms of crack loads, failure loads, ductility, deflection beha-
viour of all slab specimens. It also investigated the effects of viour and crack patterns as compared to un-strengthened con-
end anchors and embedded connectors in strengthening sys- trol slab. The proposed theoretical models predicted the
tems. Results showed that end anchors reduced the interfacial capacities of slab which were found to be comparable with
shear and normal stresses as shown in Fig. 10. The patterns of the experimental findings. The numerical analysis predicted
predicted interfacial stresses were found to be similar with the capacities of strengthened slabs and investigated the effects
existing researches. In general the interfacial stresses are high of anchor and connectors. The results obtained from numeri-
near the end of the plate which one is agreeable with the find- cal analysis were found to be comparable with experimental
ings of numerical results. Since, anchor reduced the interfacial findings. Further research work is required to predict the
stresses, it could delay or eliminate debonding failure of plates. capacity of embedded connector. A comprehensive guideline
The numerical analysis also predicted the shear stresses of con- of connector systems to prevent premature shear failure of slab
crete near the embedded connector shown in Fig. 11. The slab could also be focused on future research.
with connector showed higher shear stress at concrete as com-
pared to that without connector. Because of connector, the Acknowledgements
concrete could resist higher shear stress and, thus, could pre-
vent premature shear failure of slab. Results also showed that The authors would like to express their gratitude to the Min-
the connectors are effective near the load points (Fig. 12). Near istry of Higher Education (MOHE) for providing research
the support the concrete stress was higher as compared to grant (Grant No. 08012012ERGS) to carry out the project.
embedded connector; however, near the load point the stress Thanks are also due to the Department of Civil Engineering
of embedded connector was found to be higher. and Research Management Centre, Universiti Tenaga
Nasional and those who contributed directly or indirectly.
8. Comparison between experimental and numerical results
References
The cracking loads of slabs based on numerical analysis were
[1] G. Loreto, L. Leardini, D. Arboleda, A. Nanni, Performance of
found to be lower than those of experiment. The numerical RC slab-type elements strengthened with fabric-reinforced
cracking load of slab specimens was predicted from the flexural cementitious-matrix composites, J. Compos. Constr. 18 (2014) 3.
bar strain variations of slab as shown in Fig. 13(c) and (d), [2] M.A. Erki, P.J. Heffernan, Reinforced concrete slabs externally
whereas, the experimental crack loads of slabs were measured strengthened with fiber-reinforced plastic materials, in:
through the surface crack of the slab based on visual observa- Proceedings of the Second International RILEM Symposium
tion. Because of the visual observation of surface crack, the FRPRCS-2, 1995, pp. 509516.
experimental crack load is found to be higher than numerical [3] FIB, Bulletin 14. Externally bonded FRP reinforcement for RC
value. The numerical failure loads of all slabs were closely slab structures, Technical Report, Task Group 9.3 FRP (Fibre
comparable with the experimental findings as shown in Table 5. Reinforced Polymer) reinforcement for concrete structures;
2001; ISBN 2-88394-054-1; 130.
The mid-span deflection patterns of all slab specimens
[4] U. Meier, H. Kaiser, Strengthening of flat slab Structures with
obtained from numerical analysis were found to be identical CFRP, in: S.L. Iyer, S. Sen (Ed.), Advanced Composite
as compared to those of experiment shown in Fig. 13(a) and Materials in Civil Engineering Structures, Las Vegas, NV,
(b). The numerical results showed more ductile behaviour of 1991, pp. 224232.
specimens. The mid-span flexural bar strains of all slab speci- [5] J.M. Longworth, L. Bizindavyi, R.G. Wight, M.A. Erki,
mens obtained from numerical analysis are comparable with Prestressed CFRP sheets for strengthening two-way slabs in
those of the experimental findings (Fig. 13(c) and (d)). The flexure, in: Proc. 4th Int. Conf. on Advanced Composite
numerical analysis predicted the crack loads of slabs and yield Materials in Bridges and Structures (ACMBS-IV) (CD-ROM),
strains of steel bar. The steel plate strain patterns of all speci- 2004.
mens were also found to be identical as compared to those of [6] ISIS Canada. Strengthening reinforced concrete slab structures
with externally-bonded fiber reinforced polymers (FRPs).
experiment shown in Fig. 13(e) and (f).
Design manual No. 4, Manitoba, Canada, 2010.
[7] J.W. Wang, K.I. Tan, Punching shear behaviour of RC flat slabs
9. Conclusions externally strengthened with CFRP system, in: Proc., 5th Int.
Conf. on Fibre-Reinforced Plastic for Reinforced Concrete
The embedded connector completely prevented premature Structures (FRPRCS-5), Thomas Telford, London, 2001, pp.
9971005, 2.
shear failure of strengthened slab and allowed it to fail by flex-
[8] O. Anil, N. Kaya, O. Arslan, Strengthening of one way RC slab
ure with full ductility. It had shown the highest capacity with
with opening using CFRP strips, Constr. Build. Mater. 48 (2013)
maximum ductility as compared to others. The failure load 883893.

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018
Prevention of premature failures of RC slab 13

[9] H.A. El-enein, H. Azimi, K. Sennah, F. Ghrib, Flexural [12] M.R. Aram, C. Czaderski, M. Motavalli, Debonding failure
strengthening of reinforced concrete slab column connection modes of flexural FRP-strengthened RC beams, Composites B
using CFRP sheets, Constr. Build. Mater. 57 (2014) 126137. 39 (2008) 826841.
[10] T.El. Maaddawy, K. Soudki, Strengthening of reinforced [13] S.T. Smith, J.G. Teng, Interfacial stresses in plated beams, Eng.
concrete slabs with mechanically-anchored un-bonded FRP Struct. 23 (2001) 857871.
system, Constr. Build. Mater. 22 (2008) 444455. [14] A. Tounsi, T.H. Daouadji, S. Benyoucef, E.A.A. Bedia,
[11] S.T. Smith, S. Hu, S.J. Kim, R. Seracino, FRP-strengthened RC Interfacial stresses in FRP-plated RC beams: effect of
slabs anchored with FRP anchors, Eng. Struct. 33 (4) (2011) adherent shear deformation, Int. J. Adhes. Adhes. 29 (4)
10751087. (2009) 343351.

Please cite this article in press as: M.A. Alam et al., Prevention of premature failures of plate bonded flexurally strengthened RC slab using end anchor and connector,
Alexandria Eng. J. (2016), http://dx.doi.org/10.1016/j.aej.2016.09.018

You might also like