Professional Documents
Culture Documents
h i g h l i g h t s
a r t i c l e i n f o a b s t r a c t
Article history: One of the most common and serious damages due to an earthquake in precast buildings is the failure of
Received 3 April 2014 the beam-to-column connection. This work investigates the shear behavior of RC beam-to-column dowel
Received in revised form 3 July 2014 connections. A FEM model of the connection, detailed at the material level, is provided; it is validated by
Accepted 15 July 2014
experimental evidences. Several parametric analyses are performed in order to investigate the influence
Available online 12 August 2014
of different variables on the connection behavior and the reliability of some literature formulae for the
prediction of the dowel connection strength.
Keywords:
Ó 2014 Elsevier Ltd. All rights reserved.
RC beam-to-column dowel connection
FEM analysis
Shear strength
Precast buildings
Experimental test
http://dx.doi.org/10.1016/j.conbuildmat.2014.07.036
0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
272 G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284
2. Experimental test
2.1. Setup
180
160
140
120
Force [kN]
100
80
60
40
experimental curve
20 elaborated cuver
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Displacement [mm]
Fig. 4. Dowels restrained at the top of the beam by means of steel plate, nut and Fig. 6. Force–displacement curve of the monotonic test (dashed blue line) and
washer. ‘‘elaborated’’ curve (solid red line). (For interpretation of the references to color in
this figure legend, the reader is referred to the web version of this article.)
the Teflon sheets). The force is the horizontal action recorded by the horizontal the dowel diameter, a concrete failure is expected due to the tensile stresses in
actuator and the displacement is the relative displacement between the beam the cover. Moreover, since the frontal cover is larger than the lateral one, the col-
and the column. During the experimental test the maximum shear force reaches lapse should start at the lateral cover. The described prevision corresponds to the
a value of 169.99 kN when the horizontal displacement is equal to 0.83 mm. experimental evidence: the first crack forms at the lateral cover in the column
Removing the setup unwanted frictional strength from the recorded results, the (Fig. 9a), corresponding to the peak strength of the force–displacement curve, i.e.
maximum shear strength becomes 161.76 kN. The curve shows a brittle behavior to the failure mechanism of the connection. After this point, increasing the horizon-
of the dowel connection: after the achievement of the maximum strength, a signif- tal displacement, the cracks also propagate in the frontal concrete cover (Fig. 9b).
icant strength decay is recorded without any ductile reserve of the connection. The collapse mechanism of the connection is also confirmed by the records of the
According to Vintzeleou and Tassios [12], given the direction of the applied load installed instrumentations (Fig. 10): when the force achieves the maximum value
(Fig. 7) and the geometrical features of the connection system, the failure mecha- (t = 560 s) a sudden increase of strains is recorded by the strain gauge on the lateral
nism should be predicted. A more detailed description of such a theory is reported cover of the column (SV2 in Fig. 11) and by the strain gauge on the upper stirrup in
in Section 4. Since the column concrete covers (Fig. 8) are smaller than 6–7 times the column normal to the crack (Y1 in Fig. 11).
2 Ø 12
600 600
540
C 4 Ø 20
100 60
600
(a)
Dowel M27
Class 8.8 600
470 130
SECTION A-A SECTION B-B
Dowels M27 Class 8.8 Dowels M27 Class 8.8
100
A A
100
(b)
Fig. 5. Beam (a) and columns (b) reinforcement details. Length dimensions are in mm.
274 G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284
150
Force [kN]
100
50
force
0
0 200 400 600 800 1000 1200
Time [sec]
0.003 SV1
Strain [-]
Fig. 7. Loading conditions. SV2
0.002
0.001
0
0 200 400 600 800 1000 1200
Time [sec]
0.003 Y1
Strain [-]
X1
0.002
0.001
0
0 200 400 600 800 1000 1200
Time [sec]
Fig. 10. Instrumentations records. From the top: force of the actuator, deformations
of the concrete and deformations of the upper stirrup in the column.
Fig. 9. Phases of connection collapse: (a) first crack during the test and (b) final step of the test.
G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284 275
Failure point in
Stress compression
(peak stress)
Start of inelastic
behaviour
Cracking Unload/reload response
ultimate strain
Idealized (elastic) unload/reload response
cm 13
10 10 cm
Strain
cm cm Cracking failure
10 Softening
SV
2 1
SV
Fig. 13. Mechanical behavior of the smeared crack concrete model for uniaxial
compressive load [17].
occur and the stiffness starts to decrease (Fig. 13). The stress–strain
φ27 curve of the concrete can be evaluated according to several models
(see Section 3.1.2).
φ27 When tension stresses occur, the behavior of the concrete is
Y1
modeled linear elastic until the cracking point, where the cracks
φ8 form: the stress at this point is assumed equal to the medium value
X1 of the concrete tensile strength (see Section 3.1.2), while the strain
is assumed equal to 0.0093% (Fig. 13).
The post-cracking behavior is defined by the ‘‘tension stiffen-
ing’’ model. This model depends on the density of reinforcement,
on the quality of the bond between the rebar and the concrete,
on the relative size of the concrete aggregate compared to the
rebar diameter, and on the mesh refinement. A reasonable choice
for relatively heavily reinforced concrete, modeled with a fairly
Fig. 11. Geometrical layout of the column instrumentations. detailed mesh, is to assume a linear softening post-peak behavior,
according to the model of Vecchio [16]. The ‘‘strain’’ option of ABA-
QUS is used in order to assign this kind of behavior. It specifies two
points of the post-cracking curve: the cracking failure point and
the cracking ultimate point.
In order to completely define the smeared crack concrete
model, the failure ratios should be fixed so that the cracking sur-
face is described. The failure surface is defined by four ‘‘failure
ratios’’:
mean cubic compressive strength (Rcm = 49.91 N/mm2), the other Table 2
properties of the unconfined concrete are evaluated according to Confined concrete mechanical properties.
the Eurocode 2 [18] and reported in Table 1 in terms of: character- Column Beam
istic cubic compressive strength of the concrete at 28 days (Rck), Esec 0
f cc eccu Esec f0 cc eccu
characteristic cylinder compressive strength of the concrete at (N/mm2) (N/mm2) (–) (N/mm2) (N/mm2) (–)
28 days (fck), mean value of the concrete cylinder compressive 13,159 48.69 0.0207 12,282 50.35 0.0128
strength (fcm), mean value of axial tensile strength of the concrete
(fctm), mean value of concrete Young Modulus (Ecm), compressive
strain in the concrete at the peak stress (ec0) and ultimate compres-
sive strain in the concrete (ecu). The stress–strain law of Popovics
[19] is adopted for the unconfined concrete (red curve in Fig. 14). On the contrary, the steel dowels are modeled as three-dimen-
Table 2 shows the mechanical properties of the confined sional elements and interface elements (see Section 3.3) are
concrete for the beam and the column in terms of: confined defined in order to take into account the bond slip phenomena.
concrete secant modulus at the peak stress (Esec), confined concrete
compressive strength (peak stress) (f0 cc) and ultimate concrete 3.2.2. Steel material
compressive strain (eccu). The constitutive law of Mander et al. For both the reinforcement rebars and the dowels, experimental
[20] is adopted for the beam and column confined concrete (green tensile tests were performed. Concerning the reinforcement, for
and blue curve, respectively, in Fig. 14). each diameter used in the setup a tensile test was performed on
It has been observed that the use of the confined concrete does three specimens. One tensile test was also performed on a
not significantly change the analysis results; hence, in order to 27 mm diameter threaded bar, i.e. a dowel with the same geomet-
reduce the numerical effort, the unconfined concrete material is rical and mechanical characteristics of the tested dowels. Fig. 15
used for the whole cross-section of beam and column. shows the experimental curve (blue curve in Fig. 15) and the effec-
tive curve for the steel dowels (red curve in Fig. 15). The effective
values are the stresses and strains obtained taking into account the
3.2. Steel elements reduction of section area and the local elongation of the steel spec-
imen during the tests [21]. Fig. 16 shows the experimental curves
3.2.1. Numerical model for the reinforcement bar with the cross section diameter equal to
The specimen contains two groups of steel elements: the rein- 8 mm: for the other diameters, experimental tests confirmed the
forcement and the steel dowels. same mechanical properties.
By the software ABAQUS [11], the reinforcement can be mod-
eled through a discrete or a continue technique. In the former case,
truss or beam elements are created and a perfect correspondence
of deformations with the concrete is assumed. In the latter case, 1200
a distribution of thin layers with an equivalent thickness is
included in the model and the reinforcement is linked to the con- 1000
crete in order to describe the mutual interaction between the two
Stress [MPa]
materials. In the model of tested beam and column, truss elements 800
are adopted in order to introduce longitudinal and transversal rein-
600
forcement, considered embedded in the concrete. The beam web
rebars (2/12 in Fig. 5) are not introduced in the model, since they 400
negligibly influence the structural element strength.
200 experimental curve
effective curve
Table 1
0
Unconfined concrete mechanical properties. 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Rck fck fcm fctm Ecm ec0 ecu Strain [-]
(N/mm2) (N/mm2) (N/mm2) (N/mm2) (N/mm2) (–) (–)
Fig. 15. Stress–strain relationships for the steel dowels: experimental curve (blue
49.91 33.62 41.62 3.12 33,744 0.002 0.0035
curve) and effective curve (red curve). (For interpretation of the references to color
in this figure legend, the reader is referred to the web version of this article.)
700
60
600
50
500
Stress [MPa]
40
Stress [MPa]
400
30 300
20 200
test 1
confined concrete (column) test 2
100
10 confined concrete (beam) test 3
unconfined concrete 0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
0
0 0.005 0.01 0.015 0.02 0.025 Strain [-]
Strain [-]
Fig. 16. Experimental stress–strain relationships for steel reinforcement
Fig. 14. Concrete stress–strain relationships. (u = 8 mm).
G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284 277
180
160
ΔF=20%
140
Fig. 17. Adopted analytical model for cohesive interface elements.
120
Force [kN]
100
80
60
40
esperimental test
20 numerical analysis
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Displacement [mm]
Fig. 19. Comparison numerical analysis (blue line) versus experimental test (red
Fig. 18. Bond-stress versus slip relationship for the deformed bar – confined line) in terms of force–displacement curve. (For interpretation of the references to
concrete contact behavior according to the Eligehausen model [22]. color in this figure legend, the reader is referred to the web version of this article.)
278 G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284
Depth [cm]
to the attainment of the failure mechanism in the concrete. As a 2
consequence of above, the model can be considered suitable, since
0
it is able to catch the two main parameters of the actual behavior:
-2
the initial stiffness and the maximum strength of the dowel
-4
connection.
-6
At the failure step, the local demand of the concrete, of the dow- Column
els and of the stirrups is also evaluated. According to several exper- -8
imental tests [10], the plastic hinges in the dowels form at small -10
-0.3 -0.2 -0.1 0 0.1 0.2 0.3
depths in the column and in the beam (3–4 cm). The deformation
ε [%]
at the internal face of the dowel is reported along the beam and
column depth in Fig. 21; the tensile deformation is assumed posi- Fig. 21. Deformation of the dowel at the failure step versus the beam and column
tive and the compressive deformation is assumed negative. The depth.
higher deformations are recorded at a depth of 3 cm in the column
and in the beam; this value of depth corresponds to the experi-
mental evidence, i.e. to the depth at which the dowel appears
deformed during the inspection of the specimen after the mono- reinforcement steel and dowels steel) are assumed equal to the
tonic test (Fig. 22). A good marching is also found in terms of con- average properties of the tested specimen (Sections 3.1 and 3.2).
crete strains: at the failure step, corresponding to the maximum All the connections are implemented by the ABAQUS software,
strength of the connection, the strains recorded by the installed according to the proposed numerical model (Section 3), and hori-
instrumentation (ec,exp) are very close to the values estimated by zontal loads are applied both against the column concrete core
the proposed numerical model (ec,num) (Tables 3 and 4) (see and against the column concrete cover.
Fig. 23). The results are presented in terms of strength at the failure
mechanism of the connection, that corresponds in all the cases to
the last step of the nonlinear analyses. At this step the stress values
5. Shear strength of the dowel connection of the materials are equal or very close to their limit values, as
shown in the following.
An extensive parametric study is carried out in order to investi-
gate the main parameters which influence the shear strength of the 5.2. Evaluation of the dowel connection shear strength
dowel connection. The results of this investigation are used to
compare the numerical results of the FEM model with the formulae In this Section the reference formulae, which provide the shear
of the technical literature in terms of connection strength and fail- strength of the dowel connection, are reported. They are compared
ure modes. with the results of the numerical analyses, taking into account the
achieved failure mechanism. In all the considered relationships the
5.1. Parametric study mean values of the materials strength are used.
The CNR 10025/98 [23] provides a formula to evaluate the shear
The case studies are defined considering three variables: the strength of the dowel connection. According to this code, if the
dowel diameter, the size of the frontal concrete cover and the size eccentricity (e) of the shear force (Fig. 24) is less than half of the
of the lateral concrete cover (Fig. 8). The connection layout as well dowel diameter (db), the connection shear strength is equal to:
as the mechanical properties of the materials (concrete, qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
V Rd ¼ c db f yd f cd ð1Þ
Fig. 22. (a) Steel dowel and (b) measurement of the plastic hinge depth in the steel dowel after the monotonic test.
Table 3 ratio between the concrete cover in the load direction (frontal
Comparison between numerical (ec,num) and experimental (ec,exp) strains of the cover cF) and in the orthogonal direction (lateral cover cL), a bottom
column frontal and lateral cover (Fig. 11).
splitting (i.e. the failure of the frontal cover) or a side splitting
SV1 SV2 occurs. For low values of cL/cF, a side splitting occurs and the con-
ec,num ec,exp ec,num ec,exp nection shear strength is equal to:
0.0033% 0.0038% 0.0099% 0.0082% V Rd ¼ 2 db bct f ct ð3Þ
where bct is the net width of the concrete section, evaluated as the
section width (normal to the load) minus the diameter of the dow-
Table 4
Comparison between numerical (ec,num) and experimental (ec,exp) strains of the els, and fct is the concrete tensile strength.
column top surface (Fig. 23). If the strength of the connection is related to the concrete fail-
ure rather than to the dowel crisis (the concrete cover is lower than
S1 (%) S2 (%) S3 (%) S4 (%)
6–7 times the dowel diameter), for low values of cF/cL a bottom
ec,num 0.016 0.003 0.012 0.073
splitting occurs and the connection shear strength is equal to:
ec,exp 0.011 0.0008 0.010 0.097
c
V Rd ¼ 5:0 db c f ct ð4Þ
0:66 c þ db
COLUMN where c is the frontal concrete cover.
Different formulae have been proposed for cyclic loads and
eccentric force by Vintzeleou and Tassios [12], but they are not dis-
cussed in this paper because they are not comparable with the per-
formed experimental test.
FV
8 cm 8 cm
5.3. Influence of the dowel diameter
6 cm
S1
6 cm
connection shear strength of the dowel diameter, ranging from
S4 14 mm to 24 mm. In these case studies constant values of lateral
(cL = 100 mm) and frontal cover (cF = 130 mm) are assumed. The
S3
Table 5 Table 6
Damage percentage at the connection failure for each collapse mechanism, when the Damage percentage at the connection failure for each collapse mechanism, when the
force is applied against the concrete core. force is applied against the concrete cover.
db Failure mode II (side) Failure mode II Failure mode I db Failure mode II (side) Failure mode II Failure mode I
(mm) (bottom) (mm) (bottom)
14 14
0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1
16 16
0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1
18 18
0 0.4 0.6 1 0 0.4 0.6 0.8 1 0 0.4 0.6 1 0 0.4 0.6 1 0 0.4 0.6 1 0 0.4 0.6 1
0.8 0.8 0.8 0.8 0.8
20 20
0 0.4 0.6 1 0 0.4 0.6 0.8 1 0 0.4 0.6 1 0 0.4 0.6 1 0 0.4 0.6 1 0 0.4 0.6 1
0.8 0.8 0.8 0.8 0.8
22 22
0 0.4 0.6 1 0 0.4 0.6 0.8 1 0 0.4 0.6 1 0 0.4 0.6 1 0 0.4 0.6 1 0 0.4 0.6 1
0.8 0.8 0.8 0.8 0.8
24 24
0 0.4 0.6 1 0 0.4 0.6 0.8 1 0 0.4 0.6 1 0 0.4 0.6 1 0 0.6 1 0 0.6 1
0.8 0.8 0.8 0.4 0.8 0.4 0.8
low tensile stresses for small diameters and low compressive stres-
ses for higher diameters. In Table 5 the black arrow represents the Since in all the analyzed cases the ratio between the lateral and
percentage of the concrete tensile or compressive stress with the frontal cover is constant and lower than one, the connection
respect to the corresponding strength. Failure mode I only occurs failure is always caused by the lateral cover splitting. For this load-
for the connection with the smallest diameter (M14). In all the ing condition, even though the trend of the force–displacement
other cases, the failure occurs because of the concrete splitting, curves is more regular, the strength on average decreases of the
according to the rule for which the concrete failure occurs if the 15% with respect to the case of load applied against the concrete
cover is smaller than 6–7 times the dowel diameter. As the diam- core.
eter increases, the stresses in the dowels and in the surrounding
concrete decrease until the 45% of their limit values. The force–dis-
placement numerical curves of the six analyzed cases are reported 5.3.2. Comparison with the literature formulae
in Fig. 25. The first evidence is the increasing of strength and stiff- The strength of each case study is compared (Fig. 27) with the
ness with the dowel diameter. The major failure displacements are above presented literature formulae (see Section 5.2). The consid-
recorded when the connection failure is also characterized by high ered formulae are the CNR one (Eq. (1), black line) and the Vintzel-
stresses in the dowels (for the three smallest diameters), while in eou and Tassios relationships (Eqs. (2)–(4), green, magenta and
the other cases the failure displacements are lower and remain cyan line, respectively). The failure modes are identified by differ-
almost constant (the decrement with respect to the maximum dis- ent marker shapes: the diamond marker is used for the failure
placement is around the 30%). mode I, the circle marker is used for failure mode II with side split-
If the force is applied against the concrete cover, the frontal ting mechanism and the square marker is used for failure mode II
cover always exhibits tensile stresses. In Table 6 the percentages with bottom splitting mechanism. The literature formulae provide
of damage are reported for all the three mechanisms. The black different values with respect to the numerical results; however
arrow in the column of the failure mode II due to the bottom split- these differences can be justified.
ting indicates the percentage of the tensile stress in the column For failure mode I the formulae are based on experimental tests
frontal cover with respect to the tensile strength of the concrete. in which the tested connections were different with respect to the
investigated ones. For example, the coefficient k = 1.3 is based on
tests performed on connection systems with smooth steel bars as
180 reinforcement.
In the case of splitting failure, the differences can be understood
160 discussing the expression (3). According to Vintzeleou and Tassios
[12], in a horizontal section I-I (Fig. 28), the compressive stresses
140
on the concrete (rcc) are equilibrated by the tensile stresses (rct).
Shear strength [kN]
120 When these tensile stresses achieve the tensile strength of the con-
crete, a longitudinal crack opens and the failure starts.
100
According to Hetenyi [24] and considering the bar as a beam on
80 an elastic foundation, the compressive stress (rcc) distribution
M24 along the bar can be computed as:
60 M22
M20
40
M18 db rc cðxÞ
" #
20 M16 2bD
M14 ¼ 2 2
½sin bL cosh bx cos bðL xÞ sinh bL cos bx cosh bðL xÞ
sinh bL sin bL
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 ð5Þ
Displacement [mm]
In the formula (5) b is the foundation modulus, x is the distance
Fig. 25. Shear force against the concrete core: force–displacement curves varying from the concrete external surface, L is the dowel length embedded
the dowels diameter. in the concrete and D is the shear load. According to the Vintzeleou
G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284 281
and Tassios [12] assumptions, the compressive stresses result concrete (Table 7). For this loading condition, the frontal cover
equal to zero at x = 2.5 db, assuming bL = 5 (usual concrete quality) does not significantly influence the connection strength, that varies
and L = 8 db. The total compressive force is equal to: of the 12% (Fig. 33). This happens because, if the horizontal force
Z 2:5db acts against the concrete core, the possible failure mechanisms
F cc ¼ db rcc ðxÞdx 1:22D ð6Þ are only the failure mode I and the failure mode II due to the con-
0 crete side splitting; this is confirmed by the Table 7, where the
Then the strength Dcr can be obtained by the equilibrium at frontal cover always shows very low compressive stresses. Further-
x = 2.5 db: more, in the analyzed case, the ratio between the lateral cover and
the dowel diameter is constant and much lower than 6.
F cc ¼ F ct ! 1:22Dcr ¼ f ct bct 2:5 db ð7Þ
The comparison between the numerical results obtained by the
ABAQUS model and those obtained by the expressions (5)–(7) is 350
reported in the following in the case of dowel diameter equal to numerical model (force against core)
M20. Fig. 29 shows the compressive stress distribution, obtained 300
numerical model (force against cover)
CNR 10025/84
by the ABAQUS model, while Fig. 30 shows the same distribution V&T failure I
evaluated by the expression (5) and according to the investigated V&T failure II-side
180
160
140
Shear strength [kN]
120
100
80
M24
60 M22
M20
40
M18
M16
20
M14
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Displacement [mm]
Fig. 26. Shear force against the concrete cover: force–displacement curves varying Fig. 29. Stresses in the concrete near the dowel at the failure step by the ABAQUS
the dowels diameter. model. Stresses are in N/mm2.
282 G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284
Table 7
0.06 Damage percentage at the connection failure for each collapse mechanism, when the
force is applied against the concrete core.
d σ (D=1) [N/mm]
200
175
150
100
75
CF=50 mm
50 CF=100 mm
CF=150 mm
25 CF=200 mm
CF=300 mm
0
0 0.2 0.4 0.6 0.8
Fig. 31. Stresses in the concrete near the dowel at the failure step by the ABAQUS
Displacement [mm]
model. Stresses are in N/mm2.
Fig. 33. Shear force against the concrete core: force–displacement curves varying
the frontal cover.
Table 8
Damage percentage at the connection failure for each collapse mechanism, when the
force is applied against the concrete cover.
Fig. 32. Tangential stresses distribution in the concrete near the dowel at the strength is very low (Fig. 34). As expected, the tensile stresses in
failure step by the ABAQUS model. Stresses are in N/mm2. the frontal cover decrease at the increasing of the frontal cover
(Table 8) and, for values of cF larger than 100 mm (cL/cF lower than
1), the connection strength reaches an almost constant value
If the horizontal force acts against the concrete cover, the con- (Fig. 34), which is related to the side splitting.
nection failure always occurs when the concrete reaches the ten- In Fig. 35 the results of the numerical analyses obtained varying
sile strength in the lateral cover (Table 8), unless in the case the the frontal cover are compared to the results of the formulae (1)–
frontal cover cF is equal to 50 mm (cL/cF = 2). In this case the bottom (4). Even though for the most of the cases the failure is related to
and the side splitting contemporaneously occur and the connection the side splitting, the formula (3) provides too low strength values,
G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284 283
200 250
225
175
200
150
175
125 150
100 125
100
75
CF=50 mm 75
50 CF=100 mm
CF=150 mm 50 CL=50 mm
CF=200 mm CL=150 mm
25 25
CF=300 mm CL=200 mm
0 0
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 1 1.2
Displacement [mm] Displacement [mm]
Fig. 34. Shear force against the concrete cover: force–displacement curves varying Fig. 36. Shear force against the concrete core: force–displacement curves varying
the frontal cover. the lateral cover.
600
550 numerical model (force against core) Table 10
numerical model (force against cover) Damage percentage at the connection failure for each collapse mechanism, when the
500 CNR 10025/84 force is applied against the concrete cover.
450 V&T failure I
cL Failure mode II (side) Failure mode II Failure mode I
Shear strength [kN]
200 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1
150 200
100 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1
50
0 180
50 100 150 200 250 300
Frontal cover [mm] 160
Fig. 35. Comparison between the shear strength obtained by the numerical model 140
and the shear strength obtained by formulations available in technical literature, for
Shear strength [kN]
100
Table 9
Damage percentage at the connection failure for each collapse mechanism, when the 80
force is applied against the concrete core.
60
cL Failure mode II (side) Failure mode II Failure mode I
(mm) (bottom)
40 CL=50 mm
50
CL=150 mm
20
0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 0 0.4 0.6 0.8 1 CL=200 mm
150 0
0 0.2 0.4 0.6 0.8 1 1.2
0 0.4 0.6 1 0 0.4 0.6 0.8 1 0 0.4 0.6 1
0.8 0.8 Displacement [mm]
200
0 0.6 1 0 0.4 0.6 0.8 1 0 0.6 1 Fig. 37. Shear force against the concrete cover: force–displacement curves varying
0.4 0.8 0.4 0.8
the lateral cover.
as illustrated in Section 5.3.2. Formulae (1) and (2) provide results is still due to the side splitting, but the stress in the dowel becomes
closer to those provided by the numerical analyses, unless in the closer to the yielding one (third column in Table 9). The increase of
case of cF = 50 mm, when the failure involves the frontal cover. the lateral cover leads to a better confining effect, so that the con-
nection shows a higher shear strength (Fig. 36).
5.5. Influence of the lateral cover When the horizontal force acts against the concrete cover, the
connection failure is due to the concrete side splitting for the cases
A third set of numerical analyses is performed varying the lat- with cL = 50 mm and cL = 100 mm, while it is strongly conditioned
eral cover, assigning a constant dowel diameter (27 mm) and a by the bottom splitting failure for the largest value of the lateral
constant frontal cover (cF = 130 mm). cover (cL = 200 mm). Obviously, as the lateral cover increases, the
When the horizontal force acts against the concrete core, for the stress of the dowel at the failure also increases (Table 10). For this
lateral cover cL = 50 mm the connection failure is due to the side loading condition, the shear strength as well as the failure dis-
splitting (Table 9). For greater lateral covers, the connection failure placement are highly influenced by the lateral cover (Fig. 37).
284 G. Magliulo et al. / Construction and Building Materials 69 (2014) 271–284
550 numerical analyses, even though the FEM analyses in the most of
numerical model (force against core) the cases show a failure mode II (concrete splitting). The low reli-
500
numerical model (force against cover) ability of these formulas are justified considering that they are
450 CNR 10025/84
based on experimental tests in which the tested connections were
V&T failure I
400 V&T failure II-side different with respect to the investigated ones.
Shear strength [kN]
V&T failure II-bottom On the other side, the Vintzeleou and Tassios relationships con-
350 cerning the side and the bottom splitting provide too low strengths
300 with respect to the numerical ones. The large differences are dis-
cussed and justified considering the assumptions made by the the-
250
oretical formulae.
200
150
Acknowledgements
Fig. 38. Comparison between the shear strength obtained by the numerical model [1] D’Ambrisi A, Mezzi M. Probabilistic estimate of seismic response design values
and the shear strength obtained by formulations available in technical literature, for of RC frames. Earthq Eng Struct Dyn 2009;38(15):1709–27. http://dx.doi.org/
different values of lateral cover. 10.1002/Eqe.925.
[2] D’Ambrisi A, Mezzi M. A probabilistic approach for estimating the seismic
response of elasto-plastic SDOF systems. Earthq Eng Struct Dyn
2005;34(14):1737–53. http://dx.doi.org/10.1002/Eqe.509.
[3] Magliulo G, Ercolino M, Petrone C, Coppola O, Manfredi G. Emilia earthquake:
Fig. 38 shows the comparison between the numerical results the seismic performance of precast RC buildings. Earthq Spectra
obtained by the ABAQUS model and the results provided by the 2014;30(2):891–912. http://dx.doi.org/10.1193/091012EQS285M.
formulae (1)–(4), confirming the bad prediction of these relation- [4] Biondini F, Dal Lago B, Toniolo G. Role of wall panel connections on the seismic
performance of precast structures. B Earthq Eng 2013;11(4):1061–81. http://
ships in evaluating the connection strength. dx.doi.org/10.1007/s10518-012-9418-z.
[5] Magliulo G, Fabbrocino G, Manfredi G. Seismic assessment of existing precast
industrial buildings using static and dynamic nonlinear analyses. Eng Struct
6. Conclusions
2008;30(9):2580–8. http://dx.doi.org/10.1016/j.engstruct.2008.02.003.
[6] Magliulo G, Capozzi V, Fabbrocino G, Manfredi G. Neoprene-concrete friction
The presented paper investigates the response of dowel beam- relationships for seismic assessment of existing precast buildings. Eng Struct
to-column connections, typical of European precast industrial 2011;33(2):532–8. http://dx.doi.org/10.1016/j.engstruct.2010.11.011.
[7] Magliulo G, Ercolino M, Manfredi G. Influence of cladding panels on the first
buildings, by means of a FEM model of this connection, validated period of one-story precast buildings. B Earthq Eng 2014; http://dx.doi.org/
by experimental test results. The reference test is a shear mono- 10.1007/s10518-014-9657-2.
tonic test on a dowel beam-to-column connection. The results of [8] Ercolino M. Inelastic seismic response and risk analysis of one-storey precast
buildings. Graduation Thesis 2010.
the test confirm the expected behavior of this kind of connection [9] McKenna F, Fenves GL. OpenSees Manual. Pacific earthquake engineering
under horizontal load, showing a brittle splitting failure in the con- research center; 2013 <http://opensees.berkeley.edu>.
crete lateral cover of the column. [10] Zoubek B, Isakovic T, Fahjan Y, Fischinger M. Cyclic failure analysis of the
beam-to-column dowel connections in precast industrial buildings. Eng Struct
The numerical model of the connection is presented in details 2013;52(0):179–91. http://dx.doi.org/10.1016/j.engstruct.2013.02.028.
and compared with the results of the monotonic test, showing a [11] Corp. DSS. Abaqus/CAE 6.10-1. Providence 2010.
good agreement in terms of maximum strength, failure mechanism [12] Vintzeleou EN, Tassios TP. Mathematical-models for dowel action under
monotonic and cyclic conditions. Mag Concr Res 1986;38(134):13–22.
and local stress. [13] Ngo D, Scordelis AC. Finite element analysis of reinforced concrete beams. J Am
An extensive parametric study is performed in order to discuss Concr Inst 1967;64:152–63.
the influence of some geometrical characteristics of the investi- [14] Rashid YR. Analysis of reinforced concrete pressure vessels. Nucl Eng Des
1968;7:334–44.
gated connection on its shear strength. Many case studies are
[15] Borst Rd, Remmers JJC, Needleman A, Abellan M-A. Discrete vs smeared crack
implemented by varying the diameter of the dowels as well as models for concrete fracture: bridging the gap. Int J Numer Anal Method
the lateral and frontal concrete covers of the column, and nonlinear Geomech 2004;28(7–8):583–607. http://dx.doi.org/10.1002/nag.37.
analyses are performed. The results of all the case studies show the [16] Vecchio FJ. Disturbed stress field model for reinforced concrete: formulation. J
Struct Eng-Asce 2000;126(9):1070–7. http://dx.doi.org/10.1061/(Asce)0733-
sensitivity of the model to the parameters variation in terms of 9445(2000) 126:9(1070.
strength and of failure mechanism. It is confirmed that, if the lat- [17] Systèmes D. Abaqus Scripting User’s Manual. Providence, RI, USA2008.
eral and the frontal covers are lower than 6–7 times the dowel [18] CEN. Eurocode 2: design of concrete structures – Part 1–1: General rules and
rules for buildings. Brussels, Belgium. 2004.
diameter, the failure involves the concrete splitting, both in the [19] Popovics S. A numerical approach to the complete stress–strain curve of
case of force acting against the concrete core and in the case of concrete. Cem Concr Res 1973;3(5):583–99. http://dx.doi.org/10.1016/0008-
force acting against the concrete cover. Furthermore, if the lateral 8846(73)90096-.
[20] Mander JB, Priestley MJN, Park R. Theoretical stress–strain model for confined
cover is equal or lower than the frontal cover, the side splitting concrete. J Struct Eng-Asce 1988;114(8):1804–26.
occurs; otherwise, only in the case of force acting against the con- [21] Guo YB, Liu CR. Mechanical properties ofhardened AISI 52100 Steel in hard
crete cover, the failure also involves the bottom splitting. machining processes. J Manuf Sci Eng 2002;124:1–9.
[22] Eligehausen R, Popov EP, Bertero VV. Local bond stress–slip relationships of
The results of the parametric study are finally compared with deformed bars under generalized excitation. University of California 1986. p.
some formulae proposed in technical literature: the CNR 10025/ 169.
84 one and the Vintzeleou and Tassios relationships. [23] CNR 10025/98. Istruzioni per il progetto, l’esecuzione ed il controllo delle
strutture prefabbricate in calcestruzzo (in Italian). 2000.
The CNR 10025/84 formula and the Vintzeleou and Tassios one,
[24] Hetenyi M. Beams on elastic foundation. Ann Arbor: University of Michigan
which takes into account the failure mechanism involving the steel Press; 1946.
dowel and the surrounding compressed concrete (failure mode I)
are very similar and generally are the closest to the results of the