OUT-OF-PLANE IN SITU CYCLIC TESTING OF UNREINFORCED
STONE MASONRY WALLS WITH DISTRIBUTED LOADS
Aníbal Costa1; Alexandre A. Costa2; António Arêde3; Fábio Garcia4;
Tiago Ferreira5; Humberto Varum6
1
PhD, Full Professor, University of Aveiro, Department of Civil Engineering, agc@ua.pt
2
PhD student, Faculty of Engineering, University of Porto, Department of Civil Engineering, aacosta@fe.up.pt
3
PhD, Professor, Faculty of Engineering, University of Porto, Department of Civil Engineering, aarede@fe.up.pt
4
MsC, Researcher, University of Aveiro, Department of Civil Engineering, fabio.garcia@ua.pt
5
PhD student, University of Aveiro, Department of Civil Engineering, tmferreira@ua.pt
6
PhD, Professor, University of Aveiro, Department of Civil Engineering, hvarum@ua.pt
The present paper reports an in situ experimental test campaign carried out on existing buildings, in order to investigate the seismic behaviour of traditional masonry walls subject to out-of-plane loads. For the testing proposes, an experimental test setup based on a self-equilibrated scheme was developed and optimized to be applied in situ in two specimens on original and strengthened conditions. The obtained results are presented and carefully discussed namely from the reinforcement solutions’ efficiency point-of-view, as well as compared to previous experimental data obtained for the same type of masonry walls.
Additionally, a simplified linearized displacement-based procedure was adapted in order to characterize the nonlinear force-displacement relationship for unreinforced traditional masonry walls and to analytically predict the experimental test results. The confrontation between the experimental and the analytical results are presented and discussed.
Keywords: in situ test, cyclic tests, masonry walls, out-of-plane, strengthening, displacement-based
INTRODUCTION
The recent earthquakes around the world, including the Azores archipelago (Portugal), have led to the abandon of a considerable number of damaged masonry buildings, calling into question the preservation of this built heritage. This fact has naturally aroused the concern of the scientific community, leading to the development of several studies focused on the seismic response of these structures and on its structural rehabilitation and strengthening.
The tests presented in this work are aimed at characterizing the behaviour of typical stone masonry elements under out-of-plane horizontal loads. The difficulties felt on the characterization of rubble stone masonries have motivated a continuous work not only on the improvement of knowledge related to the techniques but also on the reliability of the results. Under this work an in situ test setup to perform cyclic tests on masonry walls was been developed and implemented, consisting on the application of surface loads to simulate the mass of the masonry element when subject to lateral accelerations. The technique presented herein is different from the previous proposes described in the literature by (Costa et al., 2011) where specimens with similar characteristics were tested.
IN SITU TEST CAMPAIGN
General overview
The experimental test campaign presented in this work follows a previous in situ tests campaign performed by the authors (Costa et al., 2011). The main goal of this new experimental test campaign is to develop and apply a test setup based on previous proposals by other authors regarding the out-of-plane behaviour of masonry walls (Griffith et al., 2007).
The tests were carried out on typical masonry houses from Azores, constituted by double leaf stone masonry with poor infill (also known as “sacco” masonry). Although these buildings normally present only 1, 2 or 3 floors at maximum, they are seismically very
vulnerable, as proved by the earthquake that hit the archipelago in 9th July 1998.
Description of the testing scheme
Taking into account the difficulties inherent to in situ tests, three fundamental orientation lines were defined: (i) the test setup should be self-equilibrated; (ii) high level of load capacity to perform experiments on strengthened specimens; (iii) ablity to perform out-of-plane tests. Concerning its versatility and straightforward implementation, the testing system was designed to work with simple and light components (less than 30 kg), avoiding exterior reaction elements (self-equilibrated system).
The reaction frame defined for the experimental test was composed by tubular steel elements (ϕ=60 mm) connecting this reaction frame of the airbags to a reaction wall (part of the existing construction). Besides the metallic elements, there are two reaction surfaces to the airbags, constituted by wood elements and marine plywood plates (see Figure 1).
Figure 1: In situ implementation for tests
The test setup developed makes use of three airbags on each sides of the wall, one compressor, a series of pipes (ϕ=8 mm and ϕ=14 mm) connecting the airbags, pressure-control valves, displacement and pressure transducers properly connected to a portable data acquisition system. As already noted, the application of the distributed loads on this kind of test is not new have already been used in the past by other authors. However, the way how the test setup was implemented in experimental campaign, associated with an original structural configuration of the reaction system which allows bigger displacements, could constitute a step forward in this kind of tests, as well as its cyclic loads capability.
Calibration of the pressure cell for the airbag test
The displacement of the wall relatively to the reaction structure is a key factor to estimating values of force during the airbag test. During the test, with the gradual inflating of the airbag, this displacement is responsible for the reduction of the contact area between the wall and the airbag. This phenomenon occurs because the inflated airbag does not assume a perfect rectangular parallelepiped shape presenting curved faces, leading to deficient contact areas near the boundaries of the airbag body (see Figure 2).
(a) Underformed (b) Deformed
Figure 2: Variation of the contact area for extreme displacement situations
In order to calibrate the contact area between the airbag and the wall, a laboratory full-scale test was performed, through which was possible to estimate a correction factor. Therefore, for a certain value of displacement, the correction factor is given by the ratio between the value measured by a load cell and the value measured by the pressure cell.
In this test, the calibration was performed only for a single airbag. For the cases of tests with multiple airbags (see Figure 3 b)) the calibration process follows exactly the same procedure, resulting in a correction factor for each single airbag. In this case the control displacement (d) is determined in relation to the centre of gravity of each airbag.
(a) Correction factor (b) Control displacement, d
Figure 3: Correction model of the contact area
Airbag Wall Reaction Structure y = 557.09x4 - 2062.1x3 + 2771.5x2 - 1640.2x + 397.58 R² = 0.9882 20 22 24 26 28 30 32 34 0.78 0.83 0.88 0.93 0.98 1.03 1.08 Dista n ce w all -re ac ti o n stru ctu re (m m )
Buildings and elements tested
In order to assess the out-of-plane seismic behaviour of traditional masonry walls, two cyclic tests were carried out in two different 1-storey buildings, named “Casa do Salão” (S) and “Casa Nova” (CN). However both construction structures were similar concerning both in-plane wall distribution, height and construction typology, S walls were retrofitted with a strengthening technique widely used after the 1998 Azores earthquake. In both cases, the monitoring system adopted was similar. Pictures, schematic layouts and monitoring schemes are presented in Figure 4 (CN) and Figure 5 (S).
(a) (b)
(c)
Figure 4: Building CN: (a) main façade; (b) Plan; (c) main façade and monitored points; and (d) cross section A1-A1
(a) (b) 1.78 0.96 1.43 1.01 1.82 0.93 1.57 0.93 1.57 0.95 4.09 1.00 2.50 2.50 4.40 0.80 Tested pier 2.20 CN Airbag Demolished area 9 12 14 10 11 13 15 0.91 0.90 0.33 0.47 A1 A1 1 6 4 3 2 1.72 0.16 0.17 0.20 Section A1-A1 0.80 1.00 0.80 0.98 0.96 1.43 1.01 1.82 0.93 1.57 0.93 1.38 2.11 0.83 1.90 0.94 1.85 0.84 0.66 3.43 1.03 1.14 1.04 0.23 0.15 2.64 2.61 0.00 3.08 1.00 0.62 0.84 2.01 2.59 0.84 1.32 2.91 Stone masonry 2.27 0.87 Pier Brick masonry (d))
(c)
Figure 5: Building S: (a) main façade; (b) Plan; and (c) main façade and monitored points; and (d) cross section A2-A2
STRENGTHENIING SOLUTIONS
In one of the tested masonry wall, strengthening techniques were applied and tested. A reinforced concrete jacking was applied in Salão house (S), in the same line as used in the reconstruction process after the 1998 Azores earthquake, and it is also recommended in Eurocode 8-Part 3-Annex C.5.1.7 (CEN, 2005) and explained briefly in (Costa, 2002) for Azorean buildings (see Figure 6). Moreover, a reinforced concrete foundation on each side of the wall was introduced, as presented in Figure 6 c) and d).
(a) (b)
(c) (d)
Figure 6: Reinforced connected plaster operation: (a) schematic representation (b) application; and strengthening at foundation level: (c) schematic representation (d) application (adapted
from (Costa et al., 2011))
A2 Section A2-A2 1.83 0.84 1.85 0.94 1.90 0.83 1.45 0.66 0.56 1.10 0.84 1.10 0.68 AIRBAGS (interior ) 0.23 0.15 0.56 1.87 Interior transducers # 1.96 0.56 8 12 14 11 13 15 10 0.94 0.94 0.28 0.25 1 6 4 3 2 0.10 1.67 0.10 0.10 0.80 (d) Metallic mesh
Connectors with max. Spacing of 1.50 m Layer of plaster and joints fill Stone masonry
Mortar
VERTICAL CROSS SECTION
6Ø12 Stirrup Ø6//0.15
Removal of small stones at the foundation allowing the penetration
of concrete 0 .5 0 TERRAIN LEVEL anchorage Steel mesh Connector to steel mesh 1Ø6//0.50 R.C. 0.30 x 0.20 m2 FOUNDATION (SUPPOSED AT -1.50 m)
DISCUSSION OF OBTAINED RESULTS
The following topics report on the out-of-plane test results carried out during the experimental campaign presented, stressing the improvement of the seismic resistance by the use of strengthening techniques. The graphical analysis of the results of “Tension vs. Displacement” and the “Displacement Profiles”, allows the characterization of mechanical properties of the masonry wall and its response to horizontal loads.
Figure 7 shows the hysteretic cycles of force vs. displacement performed on CN house (CN03 panel) as well as tension vs. drift results. From these results it is possible to highlight that the tested wall has a high displacement capacity, presenting approximately 180 mm of maximum displacement. It is important to note that the test ended due to setup limitations in terms of maximum displacement. Concerning the drift, a value of 7.54% was obtained which evidences high displacement ductility of the wall. The value of maximum surface stress obtained was approximately 6.2 kPa. It is possible to stress that on the two last F-D cycles there was a strength increase of the wall due to a undesirable contact between the wall and the reaction structure.
Figure 7: Force vs. displacement and tension vs. drift results (CN03)
The lateral displacements profile shown in Figure 8 reveals that the walls presented two distinct displacement regimes: a linear displacement between 0 m and 1.75 m height; and a non-linear displacement on the top of the wall (between 1.75 m and 2.41 m height). This phenomenon may be related to some local fragilization caused by the demolition of the lintels (previously to the experiments).
Figure 8: Vertical displacement profile and curve of dissipated energy vs. time (CN03)
-10 -8 -5 -3 0 3 5 8 10 -10 -5 0 5 10 -25 -15 -5 5 15 25 -200 -150 -100 -50 0 50 100 150 200 Drift (%) S tres s (k P a ) F o rc e (k N) Displacement (mm) 0 0.5 1 1.5 2 2.5 3 -200 -150 -100 -50 0 50 100 150 200 He ig h t (m ) Displacement (mm)
until 50% Fmax until 50% Fmax
75% Fmax 75% Fmax Fmax Fmax 0 5 10 15 20 25 En erg y Diss ip ated (k Nm ) Step
From the analysis of Figure 8 it is possible to observe that, after reaching the nonlinear regime, the energy dissipation increased linearly, reflecting the continuous propagation of damage.
The second test presented in this paper was carried out in a stone masonry wall retrofitted with the strengthening technique previously presented (building S). Due to the reduced space available outside the building, the implementation of part of the reaction structure was impossible, compromising the application of a bidirectional test (see Figure 9). Nevertheless, this fact does not preclude the assessment of the out-of-plane behaviour of the wall because
normally its out-of-plane collapse is internally restrained. Figures present the obtained results.
Figure 9: S01R2 test: reaction structure and tested wall
Figure 10: Force vs. displacement curve obtained from the S01R2 test
Figure 11: Vertical displacement profile and curve of dissipated energy vs. time (S01R2 test)
0 1 2 3 4 5 6 7 8 9 10 0 5 10 15 20 25 0 15 30 45 60 75 0 25 50 75 100 125 150 175 200 225 250 Drift (%) S tres s (k P a) F o rc e (k N) Displacement (mm) 0 0.5 1 1.5 2 2.5 3 0 50 100 150 200 250 He ig h t (m ) Displacement (mm) 25% Fmax 50% Fmax 75% Fmax Fmax 0 5 10 15 20 25 30 35 En erg y Diss ip ated (k N.m ) Step
In what concerns to the “CN03” test, a higher level of horizontal top displacement (210mm) was measured. From the comparison between both tests (CN03 and S01R02) it is important to highlight the maximum resistance of the retrofitted wall, about 3.7 times higher than the non-retrofitted wall, for similar geometry, material and support conditions. It also important to note that for the S01R2 test, the wall presented exponential energy dissipation in the second half of the test (about 2/3 of the total energy). In the force vs. displacement curve presented in Figure 10, the existence of a linear elastic plateau until 4 mm displacement is evident, followed by an increasing plastic behaviour (210 mm maximum displacement). This plastic behaviour arises due to the high base rotation capacity of the wall conferred by the strengthening solution (rigid-body behaviour).
APPLICATION OF A SIMPLIFIED DISPLACEMENT-BASED PROCEDURE
The simplified displacement-based procedure applied in this work is based on the tri-linear F-Δ relationship originally described by Lam et al. (2003). This original formulation was based on statics assuming a rigid body behaviour of the wall and neglecting the effects of its deformation and degradation. Based on this F-Δ relationship, a tri-linear model can be constructed for a one-way out-of-plane URM wall, knowing its mass, boundary conditions, overburden and dimensions.
To construct the tri-linear model, two ratios (Δ1/ Δf and Δ2/ Δf ratios) are used in
conjunction with the bi-linear rigid body model of the wall. The displacement values Δ1 and
Δ2 control respectively the initial stiffness reduction and the strength reduction and Δf
represents the maximum stable displacement. In summary, Δ1, Δ2,Δf and F0 are the only input
parameters needed to define the tri-linear F- Δ relationship forming the macro SDOF model.
Values for Δf and F0 are first determined to construct the bi-linear spine based on the wall
dimensions, boundary conditions and overburden loading conditions. The final tri-linear
relationship is then defined according to representative values of Δ1 and Δ2 which account for
the real non-linear behaviour of the wall (Lam et al., 2003).
CONSTRUCTION OF THE BI-LINEAR RIGID BODY MODEL
Considering rigid-body behaviour of a parapet wall, it is possible to describe its behaviour using basic principles of static equilibrium. Considering the overturning equilibrium of the wall about the pivot point O located at the base of the wall is possible to obtain the force of
incipient rocking F0: x Mgt F 8 3 0 (1)
where M is the total mass of the wall, g, the gravity acceleration, t, the wall thickness and x, the height of the distributed load resultant. For the cases which the wall slenderness ratio is low, the value of the incipient rocking force should be incremented due to the development of a compression strut located at the base of the wall. The value of this compression force can be calculated using Equation (2):
RFmaxAvtan
(2)where Fmax is the maximum force reached during the experimental test, Av, the horizontal
Betti & Vignoli, (2008)). Figure 12 presents the tri-linear model constructed based on the
results obtained from the experimental test. The displacement value Δ1 was obtained directly
from the experimental response envelope and Δ2 from the intersection point between the
bi-linear model spine and the ultimate resistance of an idealised envelope Hu. According to
Tomaževič (1999), this value can be assumed as 90% of the maximum resistance, Hmax.
Figure 12: Comparison between experimental force displacement curves and tri-linear force displacement model
According to this procedure is then possible to obtain new displacement values, Δ1 andΔ2,
calibrated for “sacco” stone masonry. As expected, for the tested URM wall (CN), the ratios
of Δ1/ Δf and Δ2/ Δf (see Figure 12) vary significantly from the values suggested by other
authors ((Doherty et al., 2002); (Derakhshan et al., 2009) and (Derakhshan & Ingham, 2008)). This confrontation is shown in Table 1.
Table 1: Different proposes for the tri-linear model parameters
Parameter
Reference Δ1/ Δf Δ2/ Δf
Doherty et al. (2002) 13% 40%
Derakhshan, Ingham, and Griffith (2009) 1% 25%
Derakhshan and Ingham (2008) 2% 60%
Tested URM wall (CN) 2% 49%
Note that Doherty et al. (2002) defined in his model three degradation stages which were the
criteria for determining the ratio of Δ1/ Δf and Δ2/ Δf. Considering this criteria, all the values
presented in Table 1 were obtained assuming a moderate state of degradation.
FINAL REMARKS
This work included a general overview of different topics, starting with the experimental out-of-plane behaviour characterization of unreinforced and strengthened stone masonry walls, and presentation of the new test setup with some comparisons of the obtained results. Nevertheless, the developed test setup has proved to be very effective for the characterization of masonry elements in their original in situ conditions. The bidirectional test showed to be limited to 200 mm displacement, which is indubitably a point that deserves particular attention for future developments of the test setup. Additionally, a simplified linearized
0 5 10 15 20 25 0 50 100 150 200 250 300 350 400 450 F o rc e (k N) Displacement (mm) Experimental envelope Trilinear model
displacement-based procedure was adapted and the obtained results were confronted with other values purposed by different authors.
REFERENCES
Betti, M., Vignoli, A. (2008). "Modelling and analysis of a Romanesque church under earthquake loading: Assessment of seismic resistance", Engineering Structures, 30(2), 352-367. doi:10.1016/j.engstruct.2007.03.027
CEN. (2005). Eurocode 8: Design of Structures for Earthquake Resistance. Part 3: General Rules, Seismic Actions and Rules for Buildings, prEN 1998-1. CEN: Brussels, Belgium. Costa, A. (2002). "Determination of mechanical properties of traditional masonry walls in
dwellings of Faial Island, Azores", Earthquake Engineering & Structural Dynamics, 31(7), 1361-1382. John Wiley & Sons, Ltd. doi:10.1002/eqe.167
Costa, A. A., Arêde, A., Costa, A., Oliveira, C. S. (2011)." In situ cyclic tests on existing stone masonry walls and strengthening solutions", Earthquake Engineering & Structural Dynamics, 40(4), 449-471. John Wiley & Sons, Ltd. doi:10.1002/eqe.1046
Derakhshan, H., Ingham, J. M. (2008)."Out-of-Plane testing of an unreinforced masonry wall subjected to one-way bending", Australian Earthquake Engineering Conference, AEES 2008.
Derakhshan, H., Ingham, J. M., Griffith, M.C. (2009). "Tri-linear force-displacement models representative of outof-plane unreinforced masonry wall behaviour", 11th Canadian Masonry Symposium, Toronto, Ontario, May 31-June 3. Toronto, Ontario.
Doherty, K., Griffith, M.C., Lam, N. T. K., Wilson, J. L. (2002). "Displacement-based analysis for out-of-plane bending of seismically loaded unreinforced masonry walls", Earthquake Engineering and Structural Dynamics, 31(4), 833-50.
Griffith, M C, Vaculik, J., Lam, N. T. K., Wilson, J., Lumantarna, E. (2007). "Cyclic testing of unreinforced masonry walls in two-way bending", Earthquake Engineering & Structural Dynamics, 36(6), 801-821. John Wiley & Sons, Ltd. doi:10.1002/eqe.654 Lam, N. T. K., Griffith, M.C., Wilson, J. L., Doherty, K. (2003). "Time-history analysis of
URM walls in out-of-plane flexure", Engineering structures, 25(6), 743–754. Elsevier. Tomaževič, M. (1999). Earthquake-resistant design of masonry buildings. Series on
