Beruflich Dokumente
Kultur Dokumente
cr
(N/mm
2
)
hor
(mm/m)
UR.1 0.024 0.182 --- --- --- --- 6.68 0.370
UR.2 0.024 0.083 0.218 2.468 3.646 --- 9.58 0.394
1S.AS.1
0.046
(+91.7%)
0.149
(+12.5%)
0.634
(+191%)
--- --- ---
10.36
(+27.4%)
1.120
(+193%)
1S.A.1
0.023
(-4.2%)
0.086
(-35.1%)
0.436
(100.0%)
--- --- ---
7.47
(-8.1%)
0.174
(-54.5%)
2S.B.1
0.026
(+8.3%)
0.111
(-16.2%)
0.258
(+18.3%)
0.405
(-83.6%)
0.567
(-84.4%)
1.785
11.56
(+42.2%)
0.567
(+48.4%)
2S.A.1
0.016
(-33.3%)
0.132
(-0.4%)
0.45
(106.4%)
1.195
(-51.6%)
--- ---
3.85
(-52.6%)
0.042
(-89.0%)
2S.A.2
0.044
(+83.3%)
0.190
(+43.4%)
0.433
(+98.6%)
0.754
(-69.4%)
1.077
(-70.5%)
1.815
9.85
(+21.2)
0.576
(+50.8)
1S.A.1 Side C 2S.A.2
Crack
pattern
before
test
Side A
Crack
pattern
after test
Spalling
after test
2S.B.1 Side A UR.2 Side A 1S.AS.1 Side A 2S.A.1 Side A
CFRP
sheet
Fig. 3.1. Crack patterns of some strengthened masonry panels after monotonic compression test.
International RILEM Conference on Material Science MATSCI, Aachen 2010 Vol. II, HetMat 149
This effect was confirmed by panel 2S.A.2, where the additional CFRP sheets postponed the
corner failure mechanism and the CFRP strips could bridge the vertical cracks on sides A and
C. This latter result can suggest that L-shaped strips for corner overlapping might be useful
[Val05b, Gar08].
4 NUMERICAL ANALYSIS
Numerical models were implemented to investigate the compressive behaviour of
symmetrically strengthened masonry panels with different strengthening configurations for
the bed joints reinforcement technique and different materials as reinforcements. Three-
dimensional numerical models depicting only one eighths of the whole panels, thanks to the
exploitation of the symmetry, were implemented using a micro modelling approach [Gar08]
(Fig. 4.1). Fig. 1.1 & Fig. 2.1 show the geometric dimensions of the panels and the
reinforcement configurations. Six basic models with different reinforcement patterns and one
unreinforced, reproducing some of the actual reinforcement patters (Table 2.1) of the masonry
panels were implemented (Table 4.1). In particular, the model 2S.A_LS1_SS1 corresponds to
masonry panels 2S.A.1 and 2S.A.2, while the model 2S.B_LS2_SS1 corresponds to the
specimen 2S.B.1. On the basis of the previous experimental campaigns [Gar08], only the
insertion of CFRP strips each 1, 2 or 3 bed joints was considered for the construction of the
numerical models. Furthermore, two typologies of reinforcing bars have been modelled: a
high modulus CFRP and a steel strip. The same geometrical dimensions and surface typology
were used for all the three reinforcing bars.
a) b) c)
Fig. 4.1. Implemented 3D numerical model of symmetrically strengthened masonry panels: (a)
modelled portion of the masonry panel, (b) outside view and (c) inside view of the
one eighth portion.
Table 4.1. Matrix of basic numerical models reinforced with CFRP strips.
Model Reinforcement in long sides A&C Reinforcement in short sides B&D
UR_FEM --- ---
2S.A_LS1_SS0 1 CFRP strip each bed joint ---
2S.A_LS1_SS1 1 CFRP strip each bed joint 1 CFRP strip each bed joint
2S.B_LS2_SS0 1 CFRP strip every 2 bed joint ---
2S.B_LS2_SS1 1 CFRP strip every 2 bed joint 1 CFRP strip each bed joint
2S.C_LS3_SS0 1 CFRP strip every 3 bed joint ---
2S.C_LS3_SS1 1 CFRP strip every 3 bed joint 1 CFRP strip each bed joint
150 GARBIN, GUIDI, VALLUZZI, MODENA: Numerical Modelling of Compressive Behaviour of Brick Masonry
In total, nineteen numerical models have been solved; one for the unreinforced masonry panel
and eighteen for the different strengthening configurations (6 with CFRP strips, 6 with CFRP
high modulus carbon strips and 6 with steel strips).
The numerical models were implemented in the finite elements code DIANA 9.2. All the
numerical models were built using 20 nodes isoparametric finite elements. Contact surfaces
between reinforcements and repointing mortar were modelled with 8 nodes 3D interface
elements. To simplify the model, interface elements were not considered between bricks and
all types of mortar joints. Bricks and mortars were considered as isotropic materials, whereas
CFRP strips were modelled as orthotropic material. An isotropic total strain rotating crack,
with a linear softening in tension and a multilinear stress strain diagram in compression,
failure criteria were used for bricks and mortars (Fig. 4.2). The reinforcement was considered
linear elastic. Interface elements were provided with the bond stress slip law experimentally
identified in [Gar08] for all the types of bars. The mechanical parameters of bricks, mortars,
CFRP thin strip and bond behaviour were defined according to experimental results, and
properly adapted when needed, to take into account three dimensional stress states. The
missing orthotropic mechanical parameters used in modelling the CFRP strip have been
defined according to the available values in literature [Gar08]. The values of fracture energy
in compression and in traction of bricks and mortars were defined by means of extensive
literature research for quasibrittle materials, such as concrete and masonry [Gar08]. The
relevant values are summarized into Table 4.2, where E is the modulus of elasticity, the
Poissons coefficient, f
t
is the tensile strength, G
ft
is the tensile fracture energy, f
c
is the
compressive strength, G
fc
is the compressive fracture energy and h is the crack bandwidth. For
the high modulus CFRP and a steel strip a modulus of elasticity of 200 and 210 GPa was
used, respectively.
0
5
10
15
20
25
30
35
40
0,00 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08 0,09 0,10
(-)
(N/mm
2
)
Brick sM
Mortar AS
Mortar AA
0,0
0,5
1,0
1,5
2,0
2,5
3,0
0,0000 0,0002 0,0004 0,0006 0,0008 0,0010 0,0012 0,0014 0,0016
(-)
(N/mm
2
)
Brick sM
Mortar AS
Mortar AA - bed joint
Mortar AA - head joint
Fig. 4.2. Constitutive material laws in compression (left) and tension (right).
Table 4.2. Main material properties used in FEM models.
Mechanical
property
Brick Mortar AS
Mortar AA
bed joints
Mortar AA
head joints
CFRP Strip
(longt. dir.)
E (N/mm
2
) 16800 9000 6400 6400 80000
0.25 0.20 0.20 0.20 0.10
f
t
(N/mm
2
) 2.68 1.76 1.54 1.34 ---
G
ft
(N/mm) 0.045 0.025 0.020 0.015 ---
f
c
(N/mm
2
) 35.70 18.24 11.88 11.88 ---
G
fc
(N/mm) 36.00 17.00 11.00 11.00 ---
h (mm) 41.30 18.40 28.70 27.00 ---
International RILEM Conference on Material Science MATSCI, Aachen 2010 Vol. II, HetMat 151
The finite element models were able to reasonably capture the experimental compressive
behaviours (Fig. 4.3). However, the numerical models, constructed with 3D solid elements
bonded in a continuum, were not able to follow the behaviour of the masonry panels under
highly damaged stages. Nevertheless, the models were able to confirm the effectiveness of the
intervention, by showing and quantifying the reduction of the tensile stresses in the bricks due
to the presence of CFRP strips. Furthermore, in Fig. 4.3, it can be noticed as the numerical
curve of the model 2S.C_LS3_SS1 basically reached the same ultimate vertical and horizontal
deformations of the unstrengthened model UR_FEM, but with a slight increment of the
ultimate compressive stress (+9%). The other numerical models reached the same increment
for the ultimate stress, but through a more brittle compressive behaviour that is testify by
minor vertical and horizontal deformations. Moreover, the numerical results showed that the
higher the reinforcement ratio the lower the tensile stress and damage level. In particular, the
numerical results pointed out that a reduction of the tensile stress on the long sides up to 47%
can be achieved with the insertion of CFRP strips each bed joint, whereas the others two
strengthening configuration with CFRP strips each 2 or 3 joints are basically equivalent with a
minimum reduction of 1820% and 1617%, respectively (Table 4.3).
0
2
4
6
8
10
12
14
16
-3 -2 -1 0 1 2 3 4
(mm/m)
(N/mm
2
)
UR.1
UR.2
2S.B.1
2S.A.1
2S.A.2
UR Average
UR_FEM
2S.C_Ls3_SS0
2S.C_LS3_SS1
2S.B_LS2_SS0
2S.B_LS2_SS1
2S.A_LS1_SS0
2S.A_LS1_SS1
Fig. 4.3. Experimental and numerical average compressive stress versus vertical and horizontal
deformations, of masonry panel reinforced with CFRP strips having E=80000 N/mm
2
.
Table 4.3. Value of the average maximum principal stress (
1
) recorded inside bricks of the
numerical models at the peak compressive load and comparison between the
unstrengthened and strengthened models.
Average
1
on short sides Average
1
on long sides Numerical models
with CFRP strip,
E=80000 N/mm
2
Value (N/mm
2
) Variation (%) Value (N/mm
2
) Variation (%)
UR_FEM 2,40 - 1,78 -
2S.A_LS1_SS0 2,40 0,0 0,98 -44,9
2S.A_LS1_SS1 1,68 -30,0 0,93 -47,8
2S.B_LS2_SS0 2,40 0,0 1,41 -20,8
2S.B_LS2_SS1 1,18 -50,8 1,45 -18,5
2S.C_LS3_SS0 2,40 0,0 1,49 -16,3
2S.C_LS3_SS1 1,21 -49,6 1,48 -16,9
152 GARBIN, GUIDI, VALLUZZI, MODENA: Numerical Modelling of Compressive Behaviour of Brick Masonry
The results regarding the reduction of the tensile stress along the short sides are also reported
in Table 4.3. They indicate that the insertion of reinforcement also through the thickness of
masonry walls can produce beneficial effects. This transversal reinforcement should be
overlapped with that on the long sides to avoid stress and strain concentrations at the corners
of the masonry walls as mentioned in sections 2 and 3. The relevant principal deformations
and stress reported in Fig. 4.4 show the effects on tensile stress and crack strain due to
different reinforcement patterns on the long sides of the panels.
Numerical model
Principal strain
1
(mm/m)
Normal crack strain
k
(mm/m)
Principal stress
1
(N/mm
2
)
UR_FEM
2S.A_LS1_SS0
(CFRP strips inserted
each bed joint)
2S.B_LS2_SS0
(CFRP strips inserted
every 2 bed joint)
2S.C_LS3_SS0
(CFRP strips inserted
every 3 bed joint)
Fig. 4.4. Principal strain (
1
), normal crack strain (
k
) and principal stress (
1
in N/mm
2
) at the
maximum compressive load for numerical models strengthened only on the long sides
with CFRP strips having E=80000 N/mm
2
.
International RILEM Conference on Material Science MATSCI, Aachen 2010 Vol. II, HetMat 153
In particular, in Fig. 4.4 it can be noticed as the higher the reinforcement ratio the lower the
cracking damage. A similar trend can be observed for the principal tensile stresses inside the
bricks that is consistent with the result presented in Table 4.3.
Finally, a parametric analysis on the influence of the longitudinal elastic modulus of the
reinforcement was performed. The main results obtained from the finite element numerical
analysis showed that very little or negligible further improvements of the compressive
behaviour of the masonry panels occurred when a reinforcement bar with a higher elastic
modulus than the original one was inserted in the bed joint. Furthermore, no considerable
alteration of the overall stress and strain patterns has been noticed. They are practically as
those presented in Fig. 4.3 for the original CFRP thin strip (E=80000 N/mm
2
). In addition,
minor variations of the longitudinal tensile stresses of the different bars have been noticed,
see Fig. 4.5.
(a) (b) (c)
Fig. 4.5. Numerical model 2S.C_LS3_SS0. Comparison of tensile stress (
1
in N/mm
2
) inside the
different typologies of bars at the peak compressive load: CFRP strip E=80000 N/mm
2
(a), CFRP strip E=200000 N/mm
2
(b) and Steel strip E=210000 N/mm
2
(c).
5 CONCLUSIONS
The experimental and numerical study on the compressive behaviour of masonry panels
strengthened with the bed joints reinforcement technique using CFRP thin strips embedded in
hydraulic lime-based mortar has been presented. The experimental results on monotonic
compression tests demonstrated that the insertion of small amounts of reinforcement can be
an effective technique for counteracting the masonry dilatation due to cracking. Furthermore,
this technique does not influence significantly the ultimate load and the modulus of elasticity
of the original masonry. Experimental results showed that the insertion of CFRP strips every
two bed joints, resulted particularly effective, if compared with the each joint one. This might
be related with a lower damage introduced during the application of the technique when
grooving and pointing bed joints. An interesting strength increment (+14%), together with a
remarkable reduction of the lateral dilatation up to 80%, were obtained. Numerical results
confirmed the effectiveness of the intervention, by showing and quantifying the reduction of
the tensile stresses, up to 47%, in the bricks due to the presence of CFRP strips inserted at
each bed joint, and a still good 18% for the insertion every 2 or 3 bed joints.
Therefore, the experimental and numerical data suggested that a strengthening configuration
with the insertion of 1 original CFRP thin strip every 3 mortar bed joints can be suitable for
onsite applications. In fact, it allows for a profitable reduction of the tensile stresses within
bricks and a good compressive behaviour. Moreover, it allows the minimum invasive and
more economic application configuration for the bed joints reinforcement technique,
performed with CFRP thin strips and hydraulic lime repointing mortar.
154 GARBIN, GUIDI, VALLUZZI, MODENA: Numerical Modelling of Compressive Behaviour of Brick Masonry
ACKKNOWLEDGMENT
Authors wish to acknowledge M.A.C. s.p.a. - DEGUSSA Group- of Treviso, Italy, for the
technical collaboration and the special support. Experimental tests on walls behaviour were
performed at the Laboratory of Material Testing of the Department of Structural and
Transportation Engineering of the University of Padova. Ph.D. D. Tinazzi and Eng. F.
Lucchin are acknowledged for their experimental activities regarding the CFRP bed joints
reinforcement technique.
REFERENCES
[Anz93] Anzani A., Binda L. e Mirabella G. (1993), Time dependent behaviour of masonry:
experimental results and numerical analysis, 3rd STREMA, Computational Mechanics
Publications, 415-422.
[Anz99] Anzani A., Binda L. e Mirabella G. (1999), The role of heavy persistent actions into the
behaviour of ancient masonry, Structural Faults + Repair 99, 8th International
Conference and Exhibition, London, England, July, 13-15, 1999 (CD-ROM).
[Bin92] Binda L., Gatti G., Mangano G., Poggi, C. Sacchi Landriani, G. (1992), The collapse of
the Civic Tower of Pavia: a survey of the materials and structure. Masonry International
6(1), 11-20.
[Anz00] Anzani A., Binda L., Mirabella Roberti G. (2000), The effect of heavy persistent actions
into the behaviour of ancient masonry, Materials and Structures, Rilem, Vol. 33, May
2000, 251-261.
[Sai04] Saisi A., Binda L., Valluzzi M.R., Modena C. (2004), Creep behaviour of brick masonry
panels strengthened by the bed joints reinforcement technique using CFRP thin strips,
4th Structural Analysis of Historical Constructions, Proc. intern. symp., 10-13 November
2004, Padova, Italy. Ed. Balkema, 837-846.
[Val05a] Valluzzi M.R., Binda L., Modena C. (2005), Mechanical behaviour of historic masonry
structures strengthened by bed joint structural repointing, Construction and Building
Materials, Vol. 19, 63-73.
[Mod02] Modena, C., Valluzzi, M.R., Tongini Folli, R. & Binda, L. (2002). Design choices and
intervention techniques for re-pairing and strengthening of the Monza cathedral bell-
tower, Construction and Building Materials, Special Issue, Elsevier Science Ltd, 16 (7),
385-395.
[Val05b] Valluzzi M.R., Tinazzi D., Modena C. (2005), Strengthening of masonry structures
under compressive loads by FRP strips: local-global mechanical behaviour, Science and
Engineering of Composites Materials, Vol. 12, 203-212.
[Bin01] Binda L., Modena C., Saisi A., Tongini Folli R., Valluzzi M.R. (2001) Bed joints
structural repointing of historic masonry structures, 9th Canadian Masonry Symposium
Spanning the centuries. Fredericton, New Brunswick, Canada, 4-6/06/2001, 12 pp. on
CD-Rom.
[Val03] Valluzzi M.R., Disar M., Modena C. (2003), Bed joints reinforcement of masonry
panels with CFRP bars, Composites in Constructions, Rende (CS), Italy, Sept. 16-19
2003, 427-432.
[Gar08] Garbin E. (2008), Characterization of bed joint reinforced brick masonry subjected to
compression, Ph.D. Dissertation, University of Trento.
[Tin03] Tinazzi D., Valluzzi M.R., Bianculli N., Lucchin F., Modena C., Gottardo R. (2003),
FRP strengthening and repairing of masonry under compressive load, Structural Faults
+ Repair 2003, 10th Inter. Conf. and Exhibition, London, UK, July, 1-3, 2003, 10 pp.
(CD-ROM).