Nat. Hazards Earth Syst. Sci.
, 10, 1941–1950, 2010
www.nat-hazards-earth-syst-sci.net/10/1941/2010/
doi:10.5194/nhess-10-1941-2010
© Author(s) 2010. CC Attribution 3.0 License.
An experimental investigation for external RC shear wall
applications
M. Y. Kaltakci, M. Ozturk, and M. H. Arslan
Civil Engineering Department, Selcuk University, Konya, Turkey
Received: 6 July 2010 – Revised: 20 July 2010 – Accepted: 3 August 2010 – Published: 16 September 2010
Abstract. The strength and rigidity of most reinforced con-
crete (RC) buildings in Turkey, which are frequently hit by
destructive earthquakes, is not at a sufficient level. Therefore,
the result of earthquakes is a significant loss of life and prop-
erty. The strengthening method most commonly preferred
for these type of RC buildings is the application of RC in-
filled walls (shear walls) in the frame openings of the build-
ing. However, since the whole building has to be emptied
and additional heavy costs arise during this type of strength-
ening, users prefer not to strengthen their buildings despite
the heavy risk they are exposed to. Therefore, it is necessary
to develop easier-to-apply and more effective methods for the
rapid strengthening of housing and the heavily-used public
buildings which cannot be emptied during the strengthening
process (such as hospitals and schools). This study empiri-
cally analyses the different methods of a new system which
can meet this need. In this new system, named “external
shear wall application”, RC shear walls are applied on the
external surface of the building, along the frame plane rather
than in the building. To this end, 7 test samples in 1/2 and 1/3
geometrical scale were designed to analyse the efficiency of
the strengthening technique where the shear wall leans on the
frame from outside of the building (external shear wall appli-
cation) and of the strengthening technique where a specific
space is left between the frame and the external shear wall
by using a coupling beam to connect elements (application of
external shear wall with coupling beam). Test results showed
that the maximum lateral load capacity, initial rigidity and
energy dissipation behaviours of the samples strengthened
with external shear wall were much better than those of the
bare frames.
Correspondence to: M. Ozturk
(muratozturk@selcuk.edu.tr)
Published by Copernicus Publications on behalf of the European Geosciences Union.
Natural Hazards
and Earth
System Sciences
1 Introduction
RC shear wall application is the most preferred method in
the strengthening process of the RC buildings having low
earthquake behaviour, which is a process that started with
the 1992 Erzincan earthquake and intensified after the 1999
Marmara earthquake. This method is applied in various al-
ternating ways. These alternatives can be summarized as the
turning of a partial or complete axle into a shear wall; the
application of welded wire fabric and concrete on the non-
load-bearing brick wall to create a shear wall; or the external
shear wall application performed by applying an additional
shear wall from the outside of the building.
When factors such as existing RC building stock, magni-
tude of earthquakes, damage and loss of life and property
recorded in the settlements close to earthquake hypocentre
are considered, it can be clearly concluded that there is a
great need for the infrastructure required for effective and ap-
propriate strengthening. In this scope, huge investments and
research have been made in Turkey in the last two decades.
The tests performed in the scope of the present study anal-
ysed the changes in the structural performance outputs cre-
ated as a result of external RC shear wall application onto
the plane frames (reflecting the existing building stock in
Turkey which 1) have low-quality material characteristics,
2) fail to meet section and strengthening requirements, and
3) have geometrical scales of 1/2 or 1/3). Moreover, exter-
nal shear wall-frame connection status was analysed in these
tests. In the tests carried out on the plane frame model, out-
of-plane behaviours were prevented or ignored. Earthquake-
simulating reversed-cyclic loading was applied either from
the top storey or, at different rates (triangular load distribu-
tion), from the storey levels representing the real earthquake
behaviour of the building (Sonuvar, 2004; Zhao, 2004; Er-
dem, 2006; Canbay, 2003, 2004; Chan, 2000; Ozcebe, 2003;
Kamanli, 2010). In these tests, during which particularly the
1942 M. Y. Kaltakci et al.: An experimental investigation for external RC shear wall applications
2825
100 1240 150 1240
70
70
70
70
250
3325
Fig. 1. Dimensions and reinforcement layouts of RC(a)frame and external shear wall for the 1st
RC frame
Figure 1. Dimensions and reinforcement layouts of RC frame and external shear wall for
quasi-static loadings were the 1st group
used, loadtests
was applied
(Arslan, 2007as load- 2007) First of all, the RC frames have defects commonly en-
- Kaltakci,
controlled until the yield displacement of the frame system
and, afterwards, as displacement-controlled. Researchers
who conducted the tests agreed that RC shear walls pro-
vided bare frame systems with sufficient rigidity and strength
(Anil, 2007, 2008; Ozcebe, 2003; Erdem, 2006; Zhao, 2004).
Another important issue to be considered at this point is
that strengthening is not only considered as a method that
solely increases system performance. As a matter of fact,
the feasibility of strengthening, its economic analysis, its im-
pacts on the architectural design factors, user satisfaction and
post-strengthening changes in the building functions are as
important as the performance outputs of the strengthening
method. Particularly the external shear wall method, which
claims to be applied without the need to enter into the build-
ing, has a special place among shear wall application meth-
ods.
This study analyses the impacts of external shear wall ap-
Figure 2. Properties of anchor dowels
plication, the most-discussed experimental method, on the
behaviours of the plane frame system. To this end, RC speci-
mens designed in 1/2 or 1/3 geometrical scale and tested un-
der different loading systems were analysed in terms of the
changes observed in their earthquake behaviours. As well as
the study findings and factors that should be considered in
the strengthening process; the effects of dimensions, load-
ing system and similar variables on the test results were dis-
cussed.
2 Experimental study
In the experimental study, two different test groups and their
corresponding results were analysed. Despite the difference
in the scale and loading system of these two groups, the main
objective was to analyse the pros and cons of external shear
wall application. In the tests conducted in the 1st group, ex-
ternal shear wall was placed adjacent to the frame system. In
the second group, on the other hand, external shear wall was
connected to the frame system by using different coupling
beams.
Nat. Hazards Earth Syst. Sci., 10, 1941–1950, 2010
595
100
165
85
150
835
2250
165
2250
835
250 250
250
595
(b) group tests
RC Shear (Arslan, 2007; Kaltakci, 2007).
wall
countered in buildings in Turkey, such as, (a) low concrete
strength (12∼13 MPa), (b) lack of (frequently located) suffi-
cient stirrup volumetric ratio at the beam and column ends,
(c) column ties not extending into beam-column joinings,
(d) use of vertical-hook type binders, insufficient debonding
length of beam bars, (e) construction of along-the-column
longitudinal bars as splices lapped insufficiently at the storey
and base levels, (f) strong beam – weak column was analysed
by strengthening it with an undamaged and non-defective
RC shear wall (placed adjacent to the frame or connected
to the frame via coupling beams) constructed in accordance
with the applicable earthquake regulations of the same frame
(Arslan, 2007; Bal, 2007; Bruneau, 2002; Dogangün, 2004;
Hakan Arslan, 2009; Sezen, 2003).
In the first test group; behavioural changes recorded in
the load-bearing system, when the external shear wall was
placed adjacent to the RC frame, were analysed. To this end,
4 identical RC frames having weak earthquake behaviour
(two-opening, two-storey frames modelled on 1/3 geometri-
cal scale) were produced (Arslan, 2007). These frames were
equipped with the design and construction defects commonly
encountered in the buildings in Turkey. While 2 of these
4 frames were strengthened via external shear wall applica-
tion, no strengthening was performed18 for the remaining two
frames. Then, all frames were tested under reversed-cyclic
lateral loads simulating earthquake loads. Normal force was
applied on the columns of the test elements at such a level
to ensure tensile failure in columns (N = 0.1 × Ac × fc ). At
the end of the tests, behavioural characteristics of the speci-
mens were determined and their lateral load carrying capac-
ity and shear wall efficiency were analysed comparatively.
Dimensions and reinforcement layouts of the specimens are
presented in Fig. 1. In addition, Table 1 lists the general char-
acteristics of the specimens.
As can be seen in Table 1, concrete used in frames
had a compressive strength of approximately 13 MPa and
the concrete used in shear walls of 29 MPa. Side column
dimensions were 85×100 mm and mid column dimensions
www.nat-hazards-earth-syst-sci.net/10/1941/2010/
 compressive
strength compressive
of frames strengthbars
reinforcing of the
streng
used
Axial
Axial
level load
ofload Longitudinal
Longitudinal strength of frames reinforcing
level of Longitudinal strength of frames
strength
and shear wallsof frames reinforcing bars
shearused
reinforcing
frames and wa
Test No Frame type level
columns of barLongitudinal
ratio of and shear walls frames an
M.Test No et al.: Frame
Y. Kaltakci type investigation
An experimental levelfor of
external RC
columns bar ratio of bar
shear ratio
wall of
applications
and shear walls 1943
frames and shear wa
Test No Frame type (MPa)
and shear walls(MPa) (yield/tensil
frames
Test No Frame type columns
columns
columns
bar ( ρ
ratio) of (MPa) (MPa)an (y
No/N
Table 1. Characteristics of the specimens in the 1st( test
columns ( ρ ) (MPa) (MPa) (yield/tensil
r)
program. columns ( ρ ) (MPa) (MPa)Shea(y
( N /N
( No/No r ) r ) columns ( ρ ) Shear
Test No Frame type Axial Longitudinal Average concrete FrameYield and Shear
tensile strength
Shear
Frame
Shea
( N o /N r ) 2825 Wall Wa 595
load 100 bar compressive
1240 FrameFrame
150 of the bars used
1240
Shear
100 Frame Frame
FrameWall Wall Wa
level of ratio of strength of frames in frames and
Frame

165

85
columns columns 70 and shear walls shear walls
(No /Nr ) (ρ) (MPa) Wall
(MPa) (yield/tensile)

70
387

150
G1-S1 0.1 0.013 Shear Wall
Frame 12.80Frame ----
Shear Wall ----
505

835
387 387 ----
G1-S1 0.1 0.013 12.80
G1-S1 0.1 0.013 12.80 ---- ---- 505
387
505
G1-S1 0.1 0.013 12.80 ----

165
2250
70 505
387 387

70
G1 -S1 0.1 0.013 12.80 – –
G1-S 2 0.1 0.023 13.10 505 ---- ----
505
387
G1-S2 0.1 0.023 13.10 ---- ----

835
G1-S2 0.1 0.023 13.10 ---- 505 387
G1-SG21 -S2
250

0.10.023 0.023– 13.10

387
250

---- 387
505 250

0.1 13.10 –
505 590
505

250
G1-S3 0.1 0.013 13.30 29.75 387
3325
505
387 595 730
590
G1-S3 0.1 0.013 13.30 29.75
505 730
G1-SG31 -S3 0.1
0.10.013 13.30 (a) RC
0.01329.75
frame 387
13.30
505
590
73029.75 (b) RC 387
Shear wa
387
505 wa
G1-S3 0.11. Dimensions and0.013
Figure 13.30
reinforcement layouts 29.75
of RC frame external shear590
387
and
G1-S4 0.1 0.023 14.05 28.05 505 590
st
the0.1
1 group tests (Arslan, 505
387 730
G1G
-S1 -S
4 4 0.1 0.023 0.023 2007
14.05
- Kaltakci, 2007)
28.05 14.05 505 387 28.05590
730
505 730
G1-S4 0.1 0.023 14.05 28.05 387
were 85×150 mm. While 486 longitudinal plain bars were
G -S 0.1 0.023 14.05 28.05 387
505
used in1 the4 side columns and 686 plain bars in the mid
columns of G1 -S1 and G1 -S3 ; 686 longitudinal plain bars
505
were used in the side columns and 688 plain bars in the mid
columns of G1 -S2 and G1 -S4 . Beam sizes were 85×165 mm
and, 388 tensile bars and 288 assembly bars were used in
the beams. Plain bars with a diameter of 4 mm spaced at
70 mm were used as closed ties in columns and in beams.
To ensure the simultaneous movement of the frame and the
shear wall used to strengthen it, deformed bars (8 mm in di-
ameter) were used for anchorage. These anchorage bars were
located at 150 mm intervals, starting at 100 mm from the
base. Holes (each being 80 mm in diameter) were opened on
the side column, complying with 150-mm interval. After the
holes were cleaned, dowels were inserted using Sika Anchor
fix-2 anchoring adhesive. Length of the dowel part inserted
in the shear wall was 20φ. Section dimensions of the ex-
ternal shear wall placed for strengthening were 85×595 mm.
As per TEC-2007 (Turkish Earthquake Code, 2007) require-
ments, broad shear wall edges were formed (in the part from Fig. 2. Properties
Figureof2.anchor dowels of
Properties (Arslan et al.,
anchor 2010).
dowels
base level up to the height of the shear wall) and winded
firmly to prevent plastic hinge effects. Properties of the an- via Loadcell and displacements via LVDT (Linear Variable
chorage dowels is given in Fig. 2. Displacement Transducer) and Dial-gage. Tests of the spec-
All test specimens were tested under reversed-cyclic lat- imens were conducted as load-controlled till the nominal
eral load effects, by using 500 kN-capacity rigid steel loading yield load of the system and, afterwards, as displacement-
frame. Lateral and vertical load measurements were made controlled.

www.nat-hazards-earth-syst-sci.net/10/1941/2010/ Nat. Hazards Earth Syst. Sci., 10, 1941–1950, 2010

1944 M. Y. Kaltakci et al.: An experimental investigation for external RC shear wall applications

Fig. 3. Loading system of the 1st group tests (Arslan, 2007; Kaltakci, 2008).
st
Figure 3. Loading system of the 1 group tests (Arslan, 2007 – Kaltakci, 2008)
A rigid loading frame, used in the tests, was constructed
from different steel profiles and was designed in such a way
as to enable the application of lateral load at upper storey
level. The loading frame was designed as a rigid frame to
keep the value of the horizontal and vertical movements and
rotation of the loading mechanism (constituted by the pump
and load cells) at a value which is close to zero and does
not affect test measurements. Axial load was applied on the
frame columns by using a roller system made up of steel ca-
bles. A loading platform and wall was designed as a rigid
floor plate enabling a fixed support of the test specimens.
Floor system, designed to be quite rigid when compared to
the superstructure, was fixed to the loading platform via but-
tons placed at certain intervals. In this way, the floor was
prevented from rotating due to horizontal or vertical shift.
To prevent out-of-plane movement of the specimen, a sec-
ondary frame was constructed and fixed to the test specimen
via sliding wheels. The test and measurement mechanism is
presented in Fig. 3.
Lateral load-displacement ratio curves and cumulative
energy dissipated-displacement ratio curves of the test spec-
imens are listed in Fig. 4. These curves show that maximum
loads, possibly carried by the bare frames, were 33 kN for
G1 -S1 and 45 kN for G1 -S2 . Maximum load was measured
at 125 kN for G1 -S3 and 170 kN for G1 -S4 . Accordingly,
load carrying capacity of the system was increased to ap-
proximately 3.78 folds after the frames had been strength-
ened. Undoubtedly, a building can stand after an earthquake
as long as it can dissipate sufficient amount of energy during
the earthquake. As can be understood from the energy graph-
ics, when compared to bare frames, the specimens strength-
ened with external shear wall (G1 -S3 , G1 -S4 ) dissipated 3.63
to 4.55 times more energy. The energy dissipation was de-
termined by calculating the areas inside the hysteretic load-
displacement loops for each cycle. The cumulative energy
dissipated was calculated as the sum of area enclosed by all
previous hysteretic loops.
When G1 -S1 , G1 -S2 are considered in terms of hinge for-
mation time, 1st storey column-base connections of both
frames were observed as the first parts to develop hinges. In
G1 -S3 , G1 -S4 specimens, on the other hand, frame damages
started being observed in further cycles. The first cracks
in G1 -S3 were recorded at 60 kN on the support where the
beams rest on the side columns and at 80 kN in G1 -S4 . Sig-
nificant shear cracks developed on the column-beam connec-
tions of G1 -S1 , G1 -S2 . These cracks were recorded at 20 kN
in G1 -S1 and at 25 kN in G1 -S2 . These cracks were observed
in G1 -S3 , G1 -S4 specimens in the last cycles and after the

Nat. Hazards Earth Syst. Sci., 10, 1941–1950, 2010 www.nat-hazards-earth-syst-sci.net/10/1941/2010/

 M. Y. Kaltakci et al.: An experimental investigation for external RC shear wall applications 1945

200 G1-S1
150 60000 G1-S2

Cumulative Energy Dissipation

100 50000 G1-S3
Shear Force (kN) ..

50
G1-S4
40000

(kNmm)
0
30000
-0,06 -0,05 -0,04 -0,03 -0,02 -0,01 0 0,01 0,02 0,03 0,04 0,05 0,06
-50
20000
(a) G1-S1 (b) G1-S2
-100
10000
-150
G1-S1 0
-200 0 0,01 0,02 0,03 0,04 0,05 0,06
G1-S2
Drift Ratio (%) G1-S3 Drift Ratio (%)
G1-S4
(a)
200 G1-S1
150 60000 G1-S2
G1-S1 (b)

Dissipation
100 50000 G1-S3
Force (kN) ..

60000 G1-S2 (c) G1-S3 (d) G1-S4

G1-S4
Dissipation

50 40000
50000 G1-S3 G1-S1 200 st
Figure 5. Post-test views of the 1 group test specimens (Arslan, 2007)

(kNmm)
50 G1-S4 0 Fig. 5. Post-test views of the 1st group test specimens

Energy
G1-S2 30000
150
40000
-0,06 -0,05 -0,04 -0,03 -0,02 -0,01 0 0,01 0,02 0,03 0,04 0,05 0,06

*
Shear

-50 G1-S3 (Arslan, 2007).

r Force (kN)
100
20000
(kNmm)

40
Energy

30000
,04 0,05 0,06 -100 G1-S4 Cumulative 50
(kN/mm)

10000
30 20000 -150 0
Cumulative

G1-S1 0
shear reinforcements broke. Examination of the test speci-
-50
Shea

20
10000 -200 0 0,01 0,02 0,03 0,04 0,05 0,06
G1-S2
Stifness

G1-S1 0 Drift Ratio (%) G1-S3

mens’ behaviours shows
Drift Ratio that
(%) hinges developed on the beam
-100
-150
G1-S2
10 0 0,01 0,02 0,03 0,04 0,05 0,06
G1-S4 ends and column bases and that vertical cracks developed on
-200
G1-S3 0
Drift Ratio (%) the pile fracture regions of particularly
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 the first storey beams
G1-S4 0 0,01 0,02
(a)
(b) 0,03 0,04 0,05 (160∼170 mm(b) inside Cycle
the support). After theG1-S1 tests, shear wall
G1-S1 G1-S2
60000 G1-S2
Drift Ratio (%) damages were recorded as flexural damages, which concen-
G1-S3
trated mainly within critical shear wall G1-S4 (Fig. 5). No
height
Cumulative Energy Dissipation

50000 G1-S3 G1-S1 200

50 G1-S4 G1-S2 150 separation was observed in the anchorage-type joint between
40000
(c) the frame system(d)
Shea r Force (kN) *

G1-S1 200
G1-S3 100 and shear wall (bases included) until the
(kNmm)

40
30000
3 0,04 0,05 0,06
G1-S2 150
G1-S4 50 end of the tests.
Stifness (kN/mm)

4. Results of the 1st group tests – (a) Load–displacement

Shea r Force (kN) *

G1-S3 Figure 30 20000

100
0 A different type envelope of shear curve wall of specimens
application was analysed
-50
G1-S4 50
20
10000 in the 2nd test serial. When it is impossible to ensure full
G1-S1 (b) Energy dissipation capacity graphs of specimens (c) Stiffness characteristics of specimens
-100
0 0 connection between the external shear wall and the existing
10 0 0,01 0,02 0,03 0,04 0,05 0,06 -150
G1-S2 -50 frame due to architectural reasons (the need to allocate open-
G1-S3 (d) Load
-100 histories of specimens
Drift Ratio (%) -200
0 1 ings
2 3 for
4 5 human
6 7 8 9passage
10 11 12 13as 14 the
15 16ground
G1-S4
0 floor
17 18 19 20 21 is used for com-
-150 0 0,01 0,02 0,03 0,04 0,05 G1-S1
Drift Ratio (%)
mercial purposes) or
Cycle static (the need toG1-S2 a new founda-
lay
-200
(b)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 tion for the external shear wall), inter-element
G1-S3 load transfer
0,04 0,05 (c) G1-S1 can be achieved via RC or a steel
G1-S4
coupling beam, by leav-
Cycle 21
G1-S2
G1-S3 ing some space between the shear wall and frame. To make
G1-S1 200
G1-S4
a behavioural
(d)
analysis of the systems strengthened via this
G1-S2 150
Shea r Force (kN) *

G1-S3 100 st method (which is named “external shear wall application via
G1-S4 Figure 50
4. Results of (d) the 1 group tests – (a) Load–displacement envelope curve of specimens
coupling beam”) under earthquake effect, 3 specimens at 1/2
0
(b) Energy dissipation capacity graphs of specimens (c) geometrical
Stiffnessscale were tested under
characteristics reversed-cyclic lateral
of specimens
-50
tests – (a) Load–displacement
-100
envelope curve of specimens load effect (Ozturk, 2010). Among these three specimens,
(d) Load
-150
histories of specimens the first one was not strengthened and the second one was
raphs of specimens -200
(c) Stiffness characteristics of specimens strengthened via an external shear wall equipped with a steel
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
0,04 0,05 G1-S1
coupling beam. In the third test, on the other hand, system
Cycle G1-S2 behaviour was analysed when the coupling beam (ensuring
G1-S3
G1-S4 load transfer) was a RC beam. One-opening, two-storey test
(d) frame was a RC frame having the same design defects and
failures as the one used in the 1st group tests (strong beam –
Fig. 4. Results of the 1st group tests - (a) load-displacement enve- weak column, low concrete strength, lack of (frequently lo-
lope curve of specimens, (b) energy dissipation capacity graphs of cated) sufficient binders, etc.). A schematic view of the tests
up tests – (a) Load–displacement envelope curve of specimens
specimens, (c) stiffness characteristics of specimens, (d) load histo-
and material characteristics are listed in Table 2.
ries of specimens.
graphs of specimens (c) Stiffness characteristics of specimens

www.nat-hazards-earth-syst-sci.net/10/1941/2010/ Nat. Hazards Earth Syst. Sci., 10, 1941–1950, 2010

 Average concrete Yield and tensile
nd
Table 2. Characteristics of the specimens in the 2 test programAverage concrete strengthYield
compressive of theand tensile
Axial load
1946 M. Y. Kaltakci etload
al.: AnLongitudinal Average
experimental investigation compressive
strength concrete
of frames
for Yield
externalreinforcing
RC shear strength
andwall
bars used inof the
tensile
applications
Axialof
level
compressive
and strength
shear wallsof frames strength
frames and of the walls
reinforcing
shear bars used in
Test No Frame Type Axial loadof
level
columns barLongitudinal
ratio of
Table 2. Characteristics of the specimens in the 2nd test program.Longitudinal strength of frames
(MPa)
and shear wallsreinforcing bars used
andinshear walls
(MPa) (yield/tensile)
frames
level of bar (ratio
ρ ) of
Test No Frame Type columns columns and shear walls frames and shear walls
Test No Frame Type (columns
No/Nr ) bar ratio of (MPa) (MPa) (yield/tensile)
Test No Frame type Axial LongitudinalcolumnsAverage
( ρ concrete
) (MPa) Shear Yield and tensile Shear
barcolumns ( ρcompressive
(MPa) (yield/tensile)
load ) Frame strength of the
Frame
level
(ofNo/Nr ratio
) of strength of frames Wall bars used in Wall
(N o/Nr ) Shear Shear
columns columns and shear wallsFrame Shearframes and shear walls Shear
Frame
(No /Nr ) (ρ) Frame
(MPa) Y(MPa)Frame
(yield/tensile)
Wall Wall Wall Wall
Frame Shear Wall Frame 393 Shear Wall
G2-S1 0.1 0.013 13.56 ---- ----
492
393 393----
G2G-S2-S
1 1
0.10.1 0.0130.013 13.56 13.56---- ----492 ----
492
G2 -S1 0.1 0.013 13.56 – 393 –
492
393 472
G2-S2 0.1 0.013 13.56 26.60
492 573
393 472
G2-S2 0.1 0.013 13.56 26.60
492 393573 472
G2-S
G 2-S 0.1
0.1 0.013 0.013
13.56 26.60
13.56 26.60
393 472
2 2 492 573 492 573

G2-S3 0.1 0.013
G -S 0.1 0.013
G22-S33 0.1 0.013
G2-S3 0.1
Columns and beams were constructed with dimensions
160×240 mm and 240×240 mm, respectively, for the speci-
mens. In the fourth column, 12 mm diameter plain bars were
used as longitudinal reinforcement. Six plain bars with a di-
ameter of 12 mm were used as longitudinal reinforcement in
beams. Plain bars with a diameter of 8 mm spaced at 150 mm
were used as closed ties in columns and in beams. No piles
were formed on the beams, therefore, one specific bar ratio
was selected for both support and opening sections. Frame
concrete was produced from low-quality concrete to reflect
the existing building stock of Turkey.
An external shear wall, to be connected to the frame via
a RC or steel coupling beam, was casted horizontally ac-
cording to TEC-2007 (Turkish Earthquake Code, 2007) spec-
ifications and, afterwards, was lifted and placed near the
frame. Finally, different coupling beam details were applied
between the shear wall and frame.
Additional, the external shear wall was 150×1050 mm in
section dimension and RC coupling beam in 150×240 mm
section dimension in G2 -S3 . Longitudinal deformed bars of
1688 mm and horizontal web bars of 88/150 mm were used
in the RC shear wall. Within the critical shear wall height
(the area starting from the base up to the length (lw ) of the
shear wall), special edges were formed and detailed. Lon-
gitudinal deformed bars of 6812 mm were used in the RC
coupling beam. Deformed bars with a diameter of 8 mm
spaced at 70 mm were used as closed ties in RC coupling
beam. To ensure load transfer between the existing frame
and RC coupling beam, quadro group-anchorage was applied
Nat. Hazards Earth Syst. Sci., 10, 1941–1950, 2010
393 472
13.56 26.60
492 573
13.56 26.60 393 472
393 472
13.56 26.60 492 573
492 573
393 472
0.013 13.56 26.60
492 573
on the frame by using deformed bars (14 mm in diameter) at
the level of each storey beam. Since the anchorage would
be applied on a restricted area, the depth to be chosen had to
minimize side distance effect. To this end, suggestions made
in the literature on this issue were complied with and an an-
chorage depth was determined as 9 inches (230 mm) (Can-
non, 1995). The dimension and reinforcement details of G2 -
S3 are shown in Fig. 6. Locations of the anchorages (applied
to the existing frame) on the connection region are shown in
the figure. The only difference in G2 -S2 test was that the cou-
pling beam connecting the frame to the external shear wall
was constructed not of RC but of IPE 240 steel profile. The
cross-section of the used profile is given in Fig. 6.
To simulate earthquake load in a reverse-cyclic manner
and by taking into consideration actual earthquake behaviour
to be shown by the building, test specimens were loaded and
tested with the help of a special mechanism in such a way as
to apply 2 units of load on the upper storey and 1 unit of load
on the lower storey. Success of the strengthening via external
shear wall equipped with a coupling beam depends on the ef-
fectiveness of the anchorage bars connecting the frame and
strengthening elements. Therefore, tests required a loading
method which could reveal debonding effects on the anchor-
age bars. To this end, the plates, fixed on the grooved irons
placed on the existing frame before the concrete pouring pro-
cess, were connected to the main loading system. Reloading
was performed with the help of the tensile forces applied to
the centre of the frame beams.
16
16
www.nat-hazards-earth-syst-sci.net/10/1941/2010/
M. Y. Kaltakci et al.: An experimental investigation for external RC shear wall applications
Fig. 6. Dimensions and reinforcement schemes of the RC frame and external shear wall of the 2nd group tests.
Figure 6. Dimensions and reinforcement schemes of the RC frame and external shear
nd
wall
Total load,of the
and the2firstgroup tests
and second storey loads, were mea-
sured via 3 load-cells. Specimens were fixed on the rigid la-
boratory floor to prevent any movement. The level of the
axial load applied on the frame columns was the same as the
one applied on the 1st group tests (N = 0.1 × Ac × fc ). Ax-
ial load was produced by pushing the steel beam downwards
(located on the columns via simple supports) with the help of
the hydraulic jacks mounted on it and of the bolts connected
to the floor via hinges. The measurement mechanisms of the
loading system and strengthened system are shown in Fig. 7.
Lateral load-displacement ratio curves and cumulative
energy dissipated-displacement ratio curves of the 2nd group
test specimens are listed in Fig. 8. As shown by the curves,
the application of external shear wall equipped with cou-
pling beam significantly increased the lateral load capacity
and energy dissipation capacity of the frame. While the
maximum load carried by G2 -S1 was 33 kN, it was 180 kN
for G2 -S2 and 176 kN for G2 -S3 . Accordingly, load carry-
ing capacity of the system was increased by approximately
5.80 fold after the frames had been strengthened. As can
be understood from energy graphics, when compared to bare
frames, the specimens strengthened with external shear wall
(equipped with coupling beam) dissipated 3.41 to 5.66 times
more energy. G2 -S3 dissipated the highest amount of energy,
in turn, showed the most ductile behaviour (Ozturk, 2010).
Post-test views of the specimens are shown in Fig. 9.
As seen from the views, damages were concentrated on the
nodal points in G2 -S1 and concentrated within the critical
www.nat-hazards-earth-syst-sci.net/10/1941/2010/
1947
shear wall height and on the coupling beam-external shear
wall connections in G2 -S2 and G2 -S3 . No debonding was
observed on the anchorages in G2 -S2 and G2 -S3 , which is an
indication that the quadro-anchorage detailing produced pos-
itive results. However, anchorages formed apparent fracture
cones on the shear wall when trying to debond from concrete.
The height of the fracture cones and their width (starting
from anchorage) was nearly equal to anchorage depth. This
is an important result for implementation since the distance
of the anchorage bars in the upper storey to the upper surface
of the shear wall was significantly less than anchorage depth.
Although no debonding was observed in the present study, a
sudden debonding may develop when fracture cone reaches
the upper surface of the shear wall due to faulty anchorage
workmanship. Therefore, the construction of a shear wall
higher (as much as the anchorage depth) than the frame will
prevent possible debonding.
3 Results and discussion
– Strengthened frames remained within the elasticity lim-
its and no system damage was observed even at a load
level which is two times higher than lateral load carry-
ing capacity of the bare frames. Shear damages were
recorded on the columns and nodal points (in collapse
mode) of the bare frames.
Nat. Hazards Earth Syst. Sci., 10, 1941–1950, 2010
1948 M. Y. Kaltakci et al.: An experimental investigation for external RC shear wall applications

Cumulative Energy Dissipation (kN.mm)

120000 G2-S1
200
Fig. 7. Loading system of the 2nd group tests (G2 -S3 ) (Ozturk, 2010). G2-S2 nd
Figure 7. Loading system of the 2 100000
group tests (G -S3) (Ozturk, 2010)
G2-S1 2
Cumulative Energy Dissipation (kN.mm)

150 G2-S1
Cumulative Energy Dissipation (kN.mm)

200 120000 G2-S3 120000

200 G2-S2
Cumulative Energy Dissipation (kN.mm)

100
120000
80000
G2-S2
G2-S1
150
200 100000
Force (kN)

150 100000 G2-S3

50 G2-S2
G2-S3
100
150 100000
100 60000
80000 G2-S3
Shear Force (kN)

0 80000
(kN)

10050
-0,05 -0,04 -0,03 -0,02 -0,01 500 0,01 0,02 0,03 0,04 0,05 80000
Shear

60000
40000
Shear Force (kN)

-50 0
Shear Force

50 60000
0
-0,05 -0,04 -0,03 -0,02 -0,01 0 0,01 0,02 0,03 0,04 0,05 40000
60000
-100 -50 20000
-0,05 -0,04 -0,03 -0,02 -0,01 0 0 0,01 0,02 0,03 0,04 0,05 40000
-0,05 -0,04 -0,03 -0,02-150-50
-0,01 -1000 0,01 0,02 0,03 0,04 0,05 20000
40000
-50 0
G2-S1
-100 20000
0 0,01 0,02 0,03 0,04
-200 -150 0
-100 G2-S2
G2-S1 20000
-150 -200 0 0,01 Drift Ratio0,02
(%) 0,03 0,04
Drift Ratio
-150 (%) G2-S3
G2-S2 0
G2-S1 00
-200Drift Ratio (%) G2-S1 0,01 Drift Ratio (%)
0,02 0,03 0,04
G2-S3
G2-S2 0 0,01 0,02 0,03 0,04
-200
G2-S2 Drift Ratio (%)
Drift Ratio (%)
Drift Ratio (%) G2-S3
(a)
Drift Ratio (%)
(a) (a)
G2-S3 (b) (b)
(b)
60 (a) G2-S1 (b)
60 G2-S1
G2-S2 200
50 G2-S2 200
G2-S3 150
60
50 G2-S1
G2-S3
G2-S1 150
Shea r F o rce (kN)

60
Stifness (kN/mm)

100
F o rce (kN)

40 G2-S2
G2-S2 200
Stifness (kN/mm)

40 100
200
50 50
50 G2-S3
G2-S3 150 50
150
30 0
rce (kN)

30
Shea r F o rce r(kN)

0
Stifness (kN/mm)

100
Stifness (kN/mm)

40 40 100
-50 -50
20 5050
Shea r F oShea

20
-100-100
30 30 00
10 10 -150-150
-50
-50
20 20 -200-200
0 0 -100
-100 0 10 12 23 34 45 56 6 7 7 8 8 9 9 10101111 1212 13
13 14
14 15 16 17
17 18
18 19
19 2020 21212222232324242525
0 10 0 0,01 0,01 0,02 0,02 0,03 0,03 0,04 0,04 -150
10 -150 G2-S1
G2-S1
Cycle
Cycle
Drift(%)
Drift Ratio Ratio (%) -200 G2-S3
G2-S3
-200
0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23G2-S2
24 25
0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23G2-S2
24 25
0
0
0,01
0,01
(c)
0,02
0,02 0,03
0,03
0,04
0,04 (d)
Cycle
G2-S1
G2-S1
Drift Ratio (%) Cycle G2-S3
Drift Ratio (%) G2-S3
Fig. 8. Results of the 2nd group tests (c) (d)(b) energy dissipation capacityG2-S2
- (a) load-displacement envelope curve of specimens, graphs of G2-S2

specimens, (c) stiffness characteristics of specimens, (d) load histories of specimens.

nd
Figure 8. Results of thend 2 group tests – (a) Load–displacement envelope curve of
of specimens
Figure 8. Results of the(c)2(c) group tests – (a) Load–displacement envelope
(d)
(d) curve specimens
(b)Earth
Energy dissipation capacity graphs of specimens (c) Stiffness characteristics of specimens
Nat. Hazards
(b) Figure
Energy Syst. Sci., of
8.dissipation
Results the 2nd group
10,capacity
1941–1950,
nd
2010
graphs of –specimens
tests www.nat-hazards-earth-syst-sci.net/10/1941/2010/
(c) Stiffness
(a) Load–displacement characteristics
envelope curveofofspecimens
specimens
Figure(d)8.Load
Results of the
histories 2 group tests – (a) Load–displacement envelope curve of specimens
of specimens
(d) (b)
Load histories of specimens
Energy dissipation capacity graphs of specimens (c) Stiffness characteristics of specimens
(b) Energy dissipation capacity graphs of specimens (c) Stiffness characteristics of specimens
M. Y. Kaltakci et al.: An experimental investigation for external RC shear wall applications
(a) G2-S1
Figure
Fig. 9. Post-test views of the 2nd group test9.specimens
Post-test views of the2010).
(Ozturk,
– Lateral load carrying capacity increased after strength-
ening by approximately 3.78 fold in the 1st group tests
and 5.80 fold in the 2nd group tests. A similar result was
obtained in an energy dissipation capacity. When com-
pared to the 1st group tests, post-strengthening energy
dissipation was recorded as higher in the 2nd group
tests.
– Load carrying capacity of the bare frames started to de-
crease in both groups when a maximum of 2% displace-
ment ratio was exceeded. This shows that strengthening
proves to be ineffective after a 2% global displacement.
Many literature tests were completed at a maximum of
2% horizontal drift ratio (Kara, 2006; Sozen, 1987). In
the tests carried out under the present study, on the other
hand, loading continued to 4–5% displacement level,
permitted by the measurement mechanism.
– The behaviour expected from the strengthened system
was the ductile behaviour that was not as much as for
the reference specimen. In the tests, the lateral load car-
rying capacity for the strengthened shear wall – frame
systems (G1 -S3 , G1 -S4 ) started to lose after an approxi-
mate 2% top displacement ratio whose basic reason was
due to the anchorage reinforcement debonding as a re-
sult of insufficient lap splice length at the foundation –
shear wall joint. Therefore, a considerable capacity loss
was observed for a strengthening process applied by ex-
ternal shear wall after 2% cumulative displacement ra-
tio.
– In the shear wall systems (where shear wall is adja-
cent to the frame), debonding was being observed par-
ticularly after 2% displacement ratio in the shear wall
strengthening anchored to the ground. In the specimens
(equipped with RC coupling beams), debonding was
observed in the coupling beam-shear wall connections.
Therefore, what actually limits shear wall performance
is the anchorage serving as frame-wall connector.
– Although loading mechanisms were different, a general
improvement was observed in the behaviours of both
groups. Since the tests were conducted on the plane
www.nat-hazards-earth-syst-sci.net/10/1941/2010/
1949
(b) G2-S2 (c) G2-S3
nd
2 group test specimens (Ozturk, 2010)
frame, special attention should be paid to the calculation
of the actual torsion effect (estimated in the 3-D system)
which will occur due to shear wall application.
– Shear wall height/length (hw / lw ) ratio was taken as 3
in the tests. This ratio will be apparently much higher
in a real RC building. In this case, shear wall will
reach bending capacity faster than shear capacity, there-
fore, bending mode will be more determined by the be-
haviour.
– Both groups of strengthened systems can be applied on
the non-adjacent RC buildings, the columns of which
are located on the external axes. These systems are pre-
ferred in the 2nd group tests when there is passenger
traffic particularly on the sidewalks and when the lower
storeys are used for commercial purposes.
4 Conclusions
This study analysed the changes observed in the behaviours
of the RC frames (equipped with the design and implemen-
tation defects commonly encountered in the buildings in
Turkey and having weak poor earthquake behaviour) after
they were strengthened with external shear walls with and
without coupling beams. The tests give an overview of the
structural behaviour of bare frames and strengthened frames
with external RC-SW. The study also presents the results ob-
tained at the end of the tests. Mixed system, established by
applying external shear wall as one-side strengthening on the
RC frame-type structures, significantly increases lateral load
25
strength, rigidity and energy dissipation capacity of the bare
frame. Test results showed that maximum lateral load ca-
pacity, initial rigidity and energy dissipation behaviours of
the samples strengthened with external shear wall were much
better than those of the bare frames. In addition to the struc-
tural parameters, the following should be considered in the
strengthening of existing RC buildings;
– Usability of the building after strengthening.
– Extra internal and external damage costs to occur during
strengthening.
Nat. Hazards Earth Syst. Sci., 10, 1941–1950, 2010
1950 M. Y. Kaltakci et al.: An experimental investigation for external RC shear wall applications
– Difficulties to be experienced during strengthening ap-
plication, necessity of the user to leave the building for
a long time during strengthening.
– Duration of strengthening, etc.
This study showed that strengthening and system improve-
ment of the RC buildings, via low-cost external shear wall
application with and without coupling beam, not only brings
about ease of construction and use but also provides the sys-
tem with significant behavioural improvements, strength and
rigidity.
Acknowledgements. This study was supported financially by Sel-
cuk University BAP (project no: 2007-07101033 and 2004-143).
Edited by: M. E. Contadakis
Reviewed by: two anonymous referees
References
Anil, Ö. and Altin, S.: An Experimental study on reinforced con-
crete partially infilled frames, Eng. Struct., 29(3), 449–460,
March 2007.
Anil, Ö., Altin, S., and Kara, M. E.: Strengthening of RC nonductile
frames with RC infills-An experimental study, Cement Concrete
Comp., 30(7), 612–621, August 2008.
Arslan, M. H. and Korkmaz, H. H.: What is to be learned from
damage and failure of reinforced concrete structures during re-
cent earthquakes in Turkey?, Eng. Fail. Anal., 14, 1–22, 2007.
Arslan, M. H.: Strengthening non-ductile reinforced concrete
frames by columns failure in tension by appending external shear
wall, Ph.D. thesis, Selcuk University, Konya, Turkey, 2007.
Arslan, M. H., Yuksel, I., and Kaltakci, M. Y.: An Investigation on
global ductility of strengthened 2 story 2 bay RC frames, Struct.
Build. J., 163(SB3), 177–194, 2010.
Bal, I. E., Crowly, H., Pinho, R., and Gülay, F. G.: Structural char-
acteristics of Turkish RC building stock in Northern Marmara
Region for loss assesment applications, IUSS Press, 2007.
Bruneau, M.: Building damage from the Marmara, Turkey Earth-
quake of August 1999, J. Seismol., 6, 357–77, 2002.
Canbay, E., Ersoy, U., and Ozcebe, G.: Contribution of reinforced
concrete infills to seismic behaviour of structural systems, ACI
Struct. J., 100(5), 637–643, September–October 2003.
Canbay, E., Ersoy, U., and Tankut, T.: A three component force
transducer for reinforced concrete structural testing, Eng. Struct.,
26, 257–265, 2004.
Cannon, R. W.: Straight talk about anchorage to concrete-Part 1,
ACI Struct. J., 92(6), 580–586, 1995.
Nat. Hazards Earth Syst. Sci., 10, 1941–1950, 2010
Chan, C. M., Ning, F., and Mickleborough, N. C.: Lateral stiffness
characteristics of tall reinforced concrete buildings under service
loading, Struct. Des. Tall Buil., 9, 365–383, 2000.
Dogangün, A.: Performance of reinforced concrete buildings during
the May 1. 2003 Bingol Earthquake in Turkey, Eng. Struct., 26,
841–856, 2004.
Erdem, I., Akyuz, U., Ersoy, U., and Ozcebe, G.: An Experimental
study on two different strengthening techniques for RC frames,
Eng. Struct., 28(13), 1843–1851, November 2006.
Hakan Arslan, M.: Application of ANN to evaluate effective pa-
rameters affecting failure load and displacement of RC buildings,
Nat. Hazards Earth Syst. Sci., 9, 967–977, doi:10.5194/nhess-9-
967-2009, 2009.
Kamanli, M. and Balik, F. S.: The behaviour of roof gable walls
under the effect of earthquake load, Nat. Hazards Earth Syst. Sci.,
10, 251–263, doi:10.5194/nhess-10-251-2010, 2010.
Kara, M. E. and Altin, S.: Behavior of reinforced concrete frames
with reinforced concrete partial infills, ACI Struct. J., 701–709,
2006.
Kaltakci, M. Y., Yavuz, G., and Arslan, M. H.: The effect of column
cross section to frame ductility in RC frames having poor seismic
performance, J. Eng. Appl. Sci., 2(10), 1524–1532, 2007.
Kaltakci, M. Y., Arslan, M. H., Yilmaz, U. S., and Arslan, H. D.: A
new approach on the strengthening of primary school buildings
in Turkey: An application of external shear wall, Build. Environ.,
43, 983–990, 2008.
Ozcebe, E., Ersoy, U., Tankut, T., Erduran, E., Keskin, O., and Mer-
tol, C. H.: Strengthening of brick infilled RC frames with CFRP,
Technical Report, Middle East Technical University (METU),
Structural Laboratory, TUBITAK-METU Press, 2003.
Ozturk, M.: Strengthening of reinforced concrete frames of insuffi-
cient earthquake resistance by applying external shear wall with
coupling beam, Ph.D. thesis, Selcuk University, Konya, Turkey
2010.
Sezen, H., Whittaker, A. S., Elwood, K. J., and Mosalam, K. W.:
Performance of reinforced concrete buildings during the August
17. 1999 Kocaeli Turkey Earthquake and the seismic design and
construction practice in Turkey, Eng. Struct., 25, 103–14, 2003.
Sonuvar, M., Ozcebe, G., and Ersoy, U.: Rehabilitation of re-
inforced concrete frames with reinforced concrete infills, ACI
Struct. J., 101(4), 401–500, July–August 2004.
Sozen, M. A.: Toward a behaviour based design of reinforced con-
crete frames to resist earthquakes, in: Proceedings of The Ninth
Technical Conference of Turkish Society of Civil Engineering,
V1, Ankara-Turkey, 1–47, 1987 (in Turkish).
TEC-2007: Turkish Earthquake Code, Regulations on structures
constructed in disaster regions, Ministry of Public Works And
Settlement, Ankara, 2007.
Zhao, Q. and Astaneh, A. A.: Cyclic behaviour of traditional and
innovative composite shear walls, J. Struct. Eng.-ASCE, 130(2),
271–284, February 2004.
www.nat-hazards-earth-syst-sci.net/10/1941/2010/