Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
BY
Supervisor:
Prof. Kypros Pilakoutas
The author certifies that all the material within the thesis titled Seismic Response and
Rehabilitation of Historic Masonry Buildings is his work except where it its clearly
referenced to others.
ii
Abstract
Historic masonry buildings in seismic areas are very vulnerable because they have
not been explicitly designed to withstand seismic forces. The assessment of their seismic
behaviour is very important to planning the correct rehabilitation strategies for the
improvement of their resistance during seismic events. This dissertation presents a procedure
for assessing the safety of historic masonry buildings under seismic vibrations based on
measurements of their natural frequencies, numerical simulations and failure mechanisms in
order to recommend the best rehabilitation technique.
As example a case study of the seismic behaviour of the Santo Domingo Church,
located in México, is included. For this purpose, a model of the historic masonry building
was created using the finite element basis.
The geometrical configuration of the church is as follows: the plan view inscribes a
cross (67.50m x 43.70m) with walls of 2.0m to 3.0m in depth, and vaulted roof; in the
connection a big cupola communicates with adjacent modules. In the first section of the
building two towers take part over the main facade, with 27.46m in high. The numerical
solutions obtained from the distinct element analysis are validated by comparing the results
with those obtained from the existing works (Mistler, Butenweg, & Meskouris, 2006; &
Meli, 1998) and by measurement of its natural frequencies and failure comparison with the
real structure.
The two seismic analyses performed on the structure reveals a brittle failure in wall
to wall connections of the main facade with the vaulted body, a possible cracking in middle
span of the vaulted roof and collapse of the towers by excessive tensional stresses. These
results are related to the elastic linear behaviour of the structure and should be considered
even more critical because nonlinear issues as crack, sliding and disconnections were no
measured in the structure. It is suggested the use of GFRP reinforcement in both intrados and
extrados faces of the vaulted area and the use of CFRP reinforcement bars as connections
between the towers and its base.
The conclusion of this assessment is that even when the finite element method
presents a revolutionary tool for the assessment of any structure, needs the application of
interdisciplinary tests to be confirmed as representative. Never must be considered as a
duplicate of the real behaviour and some engineering judgement is always necessary for the
correct interpretations of the results.
iii
Acknowledgements
I would like to thanks to my supervisor Prof. Kypros Pilakoutas for his guidance and
supportive analysis.
Particularly I really want to thank to my parents for all their support during all this
time without which I would not be able to study this degree.
iv
Table of Contents
Abstract .................................................................................................................. iv
1 INTRODUCTION ......................................................................................... 14
v
2.6 Assessment methods of historic masonry buildings .............................................. 36
2.6.1 Introduction ................................................................................................... 36
2.6.2 Scopes and limitations ................................................................................... 37
2.6.3 Linear static analyses ..................................................................................... 38
2.6.4 Nonlinear static analyses ................................................................................ 38
2.6.5 Dynamic analyses .......................................................................................... 39
2.6.6 Numerical modelling ..................................................................................... 39
2.6.6.1 Finite element method .............................................................................. 40
2.7 Seismic safety ...................................................................................................... 41
2.8 Rehabilitation methods ........................................................................................ 42
2.8.1 Repointing ..................................................................................................... 43
2.8.2 Grouting ........................................................................................................ 44
2.8.3 Pining ............................................................................................................ 44
2.8.4 Stitching ........................................................................................................ 45
2.8.5 Overall strengthening of old masonry buildings .............................................. 45
2.8.6 Fibre Reinforced Polymer (FRP) .................................................................... 45
2.8.6.1 FRP Strengthening of vaults and arches ................................................... 47
vi
3.5.3.3.3 Concluding remark ........................................................................... 66
3.5.3.4 Time history analysis ................................................................................ 67
3.5.3.4.1 Global model .................................................................................... 67
3.5.3.4.2 Concluding remark ........................................................................... 70
3.5.4 Strengthening and repair recommendations .................................................... 71
4 CONCLUSIONS ........................................................................................... 72
5 REFERENCES .............................................................................................. 74
vii
List of Figures
Figure 2- 1 Flowchart with the methodology for structural interventions proposed by
ICOMOS. ..........................................................................................................................16
Figure 2- 4 Adobe bricks drying in the sun (Isla del Sol, Titicaka Lake, Bolivia) ................19
Figure 2- 6 Load condition in a column (Section of St. John's Chapel showing Vaults,
Adapted from Bond, Gothic Architecture, p283).................................................................23
Figure 2- 8 View of a barrel vault from above showing forces developed by the static system.
..........................................................................................................................................25
Figure 2- 12 Failure modes of masonry panels. (M. Mistler & C. Butenweg & K. Meskouris,
2006) .................................................................................................................................30
Figure 2- 15 Different FEM approaches for modelling masonry. (M. Mistler & C. Butenweg
& K. Meskouris, 2006).......................................................................................................40
Figure 2- 16 Impact on masonry of using (a)Repointing with soft mortar. (b)Repointing with
hard mortar. (Pearson, 2007) ..............................................................................................42
Figure 2- 17 Masonry strengthening solutions with FRP composites (Oprisan at. al. 2004) .45
viii
Figure 2- 19 Summary for the full length CFRP reinforcement at the intrados and extrados.
(Basilio, Oliveira, & Lourenco, 2004). ...............................................................................48
Figure 3- 1 The front entry of the Church of Santo Domingo de Guzman in San Cristobal de
Las Casas. (Agguizar) ........................................................................................................50
Figure 3- 10 (a)Joints location into the numerical model. (b)Sections distribution. ..............59
Figure 3- 11 Principal maximum (a) and minimum (b) stresses under dead load. ................59
Figure 3- 12 (a)Principal compressive stresses under dead load. (b)Principal stresses of main
arches under dead load. ......................................................................................................60
Figure 3- 13 (a) First mode of vibration (T=0.28sec). (b) Second mode of vibration
(T=0.23sec). .......................................................................................................................61
Figure 3- 17 (a)Shear stress diagram and deformed shape. (b)Stress diagram under
earthquake loading in principal direction. ...........................................................................64
ix
Figure 3- 20 (a) 1995 earthquake accelerograms in N-S direction (b) 1995 Earthquake
accelerograms in E-W direction..........................................................................................66
Figure 3- 25 (a) Maximum envelop of principal stresses in Y-direction. (b) Deformed shape
under recorded accelerations [95 Earthquake]. ....................................................................69
Figure D- 1 Displacement in X-direction of joint 588 (orange) and 579 (blue) VS time.......89
Figure D- 2 Displacement in X-direction of joint 591 (green) and 578 (blue) VS time. .......90
Figure D- 3 Displacement in Y-direction of joint 588 (orange) and 579 (blue) VS time.......91
Figure D- 4 Displacement in Y-direction of joint 591 (green) and 578 (blue) VS time. .......92
Figure D- 5 Response spectrum curve for joint 1365 matching with the third mode of the
structure at 0.23sec Period vs. Pseudo spectral acceleration in Y-direction. .........................93
Figure D- 6 Response spectrum curve for joint 591 matching with the first mode of the
structure at 0.3sec Period vs. Pseudo spectral acceleration in Y-direction. ...........................94
x
List of Tables
Table C- 2 Modal periods and frequencies for each vibration mode. ...................................88
xi
Without words,
Hermann Hesse
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 1 3
1 INTRODUCTION
1.1 Foreword
During the last decades, great advances in the application of numerical methods
relevant to the assessment of the structural behaviour of monuments and historic buildings
have been developed around the world; e.g. the finite element method (FEM). Recent facts
of two Peruvian churches collapsing by the 8.0 magnitude earthquake that hit the Peru coast
(BBC-NEWS, 2007); killing more than 500 people, confirm the idea that the applicability of
past seismic response calculation methods on historic masonry buildings and reliability of
existent retrofitting techniques need to be improved. As consequence advance analysis
simulation need to be implemented as additional tool that can lead us to increase the
confidence about the limits and capabilities of the strengthening work.
In order to do this, a good definition of the problems into the structure is required in
order to select the appropriate restoration and strengthening work. For the computational
evaluation of the structure, the development of a numerical model is needed. During this
process some difficulties need to be confronted including determining of material properties,
geometry, boundary conditions, loading characteristics and verification of the model.
Because masonry materials present a nonlinear behaviour and due to the lack of information
on their characteristics, a simplification by considering continuous material properties into
the elements can be suggested for the calculations.
Since this thesis intends to demonstrate the typical failure mechanisms and improved
repair techniques of masonry buildings under seismic actions, in particular historical
structures, a brief introduction to some materials, elements and historical facts is given first.
Following that, a numerical model for the a structure is developed, and at the end, a
computational analysis is used to determine the seismic behaviour of the Santo Domingo
Cathedral in Mexico and presented as a practical application of seismic assessment.
13
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 1 4
The overall aim of this dissertation is the seismic assessment of historic masonry
buildings through finite element modelling with a commercial software (SAP2000®) and the
proposal of strengthening solutions by measuring the common failure mechanisms (local and
global cases). In this exercise, the engineer will be aided by the knowledge of the magnitude
of principal stresses, maximum displacements and deformed shapes, obtained by a linear
elastic finite element analysis under two sources of excitation: a response spectrum
according to the Mexican code (Mexican-Norms, 2007) and time history accelerations
recorded during the 21 of October of 1995 seismic event in the south of Mexico.
· Performing both static and dynamic linear analysis using CAE software
(SAP2000®) to determine potential stress concentrations and failure mechanisms.
Comparison with previous failures in the structure.
14
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 1 5
Chapter 2.2 The purpose of this chapter is to explain the structural characteristics of
composite materials employed in historic masonry buildings. The text centralizes in stone
and brick masonry, that were the most common materials employed during the last centuries.
Some mechanical properties of materials used in historic buildings are indicated in this
section, but according to experts in the matter, extreme provision should be taken in the
employment of these values.
Chapter 2.3 A short explanation is given in this chapter of masonry types used for
construction of monuments, cathedrals and other important buildings; mainly built with
composite masonry in different configurations.
Chapter 2.4 The purpose of this chapter is to describe the common elements in
historic masonry buildings that satisfied with the basic structural requirements of the epoch;
to explain its behaviour against external loads and failure modes as single components.
Chapter 2.5 The purpose of this chapter is to explain the typical failure mechanism
developed in masonry structures.
Chapter 2.6 This chapter discusses and present the new calculation techniques. The
improvement achieved by processors capable of compute thousands of calculations in
seconds and its impact on numerical methods is commented.
Chapter 2.8 Retrofitting methods are discussed in this chapter; its advantages and
limitations according to the type of failure presented in Chapter 2.5.
Chapter 3 Case study. The first part, contain a detailed description of the structure
and material properties chosen for structural analysis. The second part comments about the
way the structure is modelled, analytical results and strengthening recommendations.
Chapter 4 Present conclusions of the overall paper and gives recommendations for
future work.
15
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 1 6
2 LITERATURE REVIEW
2.1 Introduction
Polemic has exited during the last two centuries in establishing criteria for
rehabilitation of historic buildings and monuments. Without coming to any general
consensus, the letter of Venice (see Appendix A) was written in May of 1964 as a result of
deliberations of many specialists and technicians in the restoration of historic monuments.
During the congress many issues for the preservation of historic structures were discussed.
The letter focuses on the maintenance of harmony between the structure and new
rehabilitation work. According to the paper such interventions must follow the next basic
principles: material compatibility, conservation of overall lay-out or decoration and mass
colour relationship, avoiding of removing any part, or additions to the building. At the end
the text precise documentation of all rehabilitation works by means of critical reports
(including drawings and photographs) and recommends its publication.
Figure 2- 1 Flowchart with the methodology for structural interventions proposed by ICOMOS.
16
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 1 7
For a historic building anyone that is the material, shape and construction method,
the structural behaviour is always dominated by the same basic principles of structural
mechanics applied to any present building; as a result it is common to guess that its stability
can be studied under the same principles. However, specialists in the matter consider, due to
different motifs derived from negative experiences in the past (Binda, Penazzi, & Saisi,
2003), to not limit the structural engineering only to calculus if not make use of different
implements as supervision, monitoring and constant evaluation of the general behaviour to
correct and establish the real conditions of the structure to be studied.
2.2.1 Introduction
Masonry is well known as the oldest material used for construction. The ease to
initial assembling, structural properties, strength against majority of external conditions;
17
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 1 8
made masonry the preferred construction method during ancient times. The Greek structures
and ancient cathedrals around the world are proving of the durability of this kind of
structures.
2.2.2 Stone
The natural stone is the most common component in historic masonry buildings and
is considered as a structural material by its own. Many classical monuments are made of
stones which lie one on top of each other without any application of mortar. One clear
example is the Temple of Olympian Zeus (Figure 2-2).
In old times, the type of stone to be used during construction was selected based on
its durability and workability. The durability is considered as the most important property of
stone, even when both characteristics are connected by the density of the material. Another
important property is the strength, but in those times it wasn’t as important as the ability to
resist the local weather conditions (Meli, R., 1998).
18
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 1 9
2.2.2 Mud
The mud started to be used as rolled ground, or as covering or landfill. Its main
weakness is the degradation by environmental agents and cracking by drying of its
components. Few constructions of this kind are still remaining in America, were the use of
mud was relatively frequent in some Spanish constructions.
Construction based on adobe data from about 3000 years ago, and is still being used
in rural areas of development countries (Meli & Sanchez-Ramirez, 1993).
The mechanical properties of adobe, has big variations and follows the
characteristics of the ground and elaboration. Its strength in compression varies from 5 to 20
kg /cm2 and between 0.25 and 1 kg /cm2 in tension (Meli, R., 1998).
Figure 2- 4 Adobe bricks drying in the sun (Isla del Sol, Titicaka Lake, Bolivia)
19
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 0
2.2.3 Mortar
The necessity of a new material that could fill the holes between stones and provide
with some adherence and continuity gave place to the appearing of mortars. The first element
used for this purposes was mud. In Mesopotamia was used since the third century B.C.;
adding asphalt as stabilizer of the mud for joints and landfills (Meli, R., 1998). Later, sand
and lime mortar will allow the construction of more resistant and durable elements until the
widespread and acceptance of Portland cement.
In our days, the historic masonries structures are generally made of lime mortar. It is
very important to not confuse the non-hydraulic with the hydraulic mortar; the last
demonstrate a different behaviour as construction material and is not valid to repair historic
monuments. Only the non-hydraulic lime mortar is capable of preserve its properties during
centuries; this because the lime requires many years to acquire certain strength during the
carbonization process, maintaining a constant porosity in the element and allowing lime to
have certain impermeability in the compound.
The lime and sand mortars have a typical strength around 50kg/cm2 and 200kg/cm2
(Meli, 1998). Even when there are more durable than the adobe mortars, are also affected by
external agents as humidity and salts.
In distinct eras and cultures, different additives where employed in mortars. Many of
these new mortars contained many types of different aggregates as chopped straw, reed,
manilla hemp, jute, sisal, and even sawdust (Bennett, 2002). These mortars were expensive
and employed only in very important monuments.
Decades later, Romans will achieve a significant advance when they added
puzzolana to lime mortars. The last caused lime to increase its carbonization process without
need of a long time period, giving place to a much faster hardening and important increase in
the strength and durability. This discovering will facilitate the development of the first
concrete.
The Portland cement was introduced in the XIX century. The cement mortar,
reached strengths a lot much superior to lime mortars (around 50-200 kg/cm2), had a very
fast hardening and high elasticity modulus (Meli, R., 1998).
The main characteristic of mortars such as concretes, bricks and stones is that have
a good capacity to absorb compression stresses, but its tensile strength is typically low,
generally around 10% of the compression strength.
20
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 1
Masonry is without doubt the must important and developed building method of
early civilizations.
Masonry is the combination of stones or bricks with mortar, in which the mortar
plays multiple functions as to fill the holes between stones and bring better uniformity in the
transfer of loads into the element.
In the beginning the structures were erected without any grout, but with time soils
like clay and lime became to be widely used. For example, after the cyclopean masonry the
Greeks developed the dressed stone masonry, which became the standard way for
constructing monumental buildings in both Greek and Roman civilizations. Stones were
connected without mortar, formed of entire stone blocks one above other and with joints as
small as possible.
Brick masonry started to be employed in Greece since the 7 th century A.D. as the
common construction form using opus caementitium (Roman concrete) (Mistler, Butenweg,
& Meskouris, 2006).
“The Romans have often combined the new technique of opus caementitium with
conventional masonry construction sacrificing some aesthetics along the line. To counteract
this, an outer layer from natural stones was used. Depending in the material used for the
outer layer of the wall, it was called opus incertum (dressed stones) (Figure 2-4(a)) or opus
implectum (quarry stones) (Figure 2-4(b)). Also, special shapes called opus mixtum were
developed (Figure 2-4(c))” (Mistler, Butenweg, & Meskouris, 2006).
21
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 2
Europe (Mistler, Butenweg, & Meskouris, 2006). The outer layer consisted of quarry-stones,
dressed stones or brick and the inner part was formed of loose stonewall masonry.
All these layers were needed because in past centuries the thickness of the walls was
considered as the most important factor in the resistance of the structure as a whole; it is
clear in remaining structures that the dimensions often depended on the height of the
structure.
According to Williams (1986) a thicker mortar results into a weak masonry in terms
of strength. The old constructers having this in mind started to make use of a bigger
relationship of mortar in the final compound.
2.4.1 Column
This element fulfils with the basic functions of supporting and transferring loads to
the ground and is considered as the simplest one according to its structural behaviour,
because is generally subjected to compression stresses and all loads are solved axially.
However, sometimes we must take into account the possible eccentricity in the load
application.
The more primitive columns were made of wood trunks buried into the ground until
reach some good support in the base. These forms evolved to columns formed by blocks and
in some cases masonry. Their localization into the structure caused that architects considered
columns as an opportunity to develop some aesthetics with the style of the epoch.
The load capacity of the column depends on the strength of the constitutive
materials, but is also affected by certain factors that considerably reduce its capacity. It is
argued (Meli, 1998) that added to axial loads, the arches sustaining the roof produces lateral
movements adding an extra load to the columns. Only if the loads at both sides of the
columns are well balanced, the horizontal components of the forces balance each other. As
result, in general, there is always some eccentricity in the resultant, and the columns are
constantly under flexion-compression (Figure 2-5).
22
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 3
Figure 2- 6 Load condition in a column (Section of St. John's Chapel showing Vaults, Adapted from Bond,
Gothic Architecture, p283)
Columns only shows signs of damage when are close to collapse conditions, its type
of failure in historic structures is brittle. The main symptom of a column reaching its load
capacity is the developing of vertical cracks that indicates lateral expansion of the material
and result in the detachment of the constituents. There are some cases were columns clearly
indicate several damage but still under those situations have stand for a long time. Many
experts attribute it to load transmission between adjacent elements of support.
The main problematic with these elements is that they do not absorb lateral forces or
moments.
2.4.2 Wall
Walls are like columns, elements that usually support the weight of the building and
comply with different functions as absorbing lateral forces from wind, earthquakes and those
from static systems (arches, vaults).
The failure modes depend on the type of load is acting on the structure and to the
internal composition of the masonry. The cracking configuration is one of the best signs of
which phenomenon is affecting the structure, and is always merit of a careful evaluation
(Lourenco, 2004).
23
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 4
The typical failure modes of this element will be discussed later in Chapter 2.5.
2.4.3 Arch
The arch was first developed in the Indus Valley civilization close to 2500 BC and
subsequently in Mesopotamia, Egypt, Assyria, Etruria, and later refined in Ancient Rome
(Meli, 1998).
While for support elements the solution was direct and simple, for roofs and stories
the solutions became more difficult to formulate. Beams and wood were used at the
beginning to cover short spans. It is common to assume that these solutions evolved from
roofs with shapes of inverted V to arches.
The structural behaviour of an arch is significant. The vertical loads applied are
transmitted to the supports following a pat that depends on the distribution of the external
loads and geometry of the arch. This configuration eliminates tensile stresses in spanning an
open space and all forces are resolved into compressive stresses (Figure 2-6).
a) b)
Figure 2- 7 (a)Schematic illustration of an arch. 1.Keystone, 2.Voussoir, 3.Back, 4.Impost, 5.Intrados, 6.Rise,
7.Bay, 8.Abutment. (Messer Woland), (b)Arch supported by a leaning buttress. (Block, Ciblac, & Ochsendorf,
2006)
By using the arch pattern significant spans could be achieved; looking in this way;
the design of the arches became more in a geometry problem than a stresses one. The
24
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 5
geometry is chosen based on two goals: to minimize the eccentricities between the pressure
line of the arch; and to minimize the lateral forces in the supports.
It is very difficult to find arches that collapsed only by effect of static loads, there is
always a movement in the supports due to external conditions as foundation settlements, or
seismic movements. The opening in these supports develops a cracking pat that begins in the
lower part of the middle span, and later cracking start to appear in the upper part of the
imposts. A very big crack is required into the structure to create a mechanism that fails by
collapse (Meli, 1998). This topic will be discussed more deeply later in the next.
2.4.4 Vault
The natural extension of the arch to form a complete roof is the barrel vault, which
can be considered as a series of arches. Its behaviour, failure modes and stress analyses can
be studied by considering a simple section of vault unity wide.
As in the arches, the critical aspect is in the rigidity of the supports. In this case is
needed to avoid any lateral movement in the whole vault or cracking that can induce sliding
in the supports (Figure 2-7).
Figure 2- 8 View of a barrel vault from above showing forces developed by the static system.
25
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 6
2.4.5 Dome
Same as arches, the development of the domes, passed through the false dome, in
which circular rings made of stone are built enclosing progressively the span, to double
domes with inner and outer shells (Meli & Sanchez-Ramirez, 1993).
Domes by its geometry sit directly on a circular base; however, when this is not
possible, vault sections are added in the corners providing the transition between dome and
the square base on which it is set. These structures that transfer the weight of the dome are
called pendentives (Figure 2-8).
For its construction, the dome requires the use of stones of complex geometry. The
last triggered that most domes started to be built with mud bricks, lime mortars and
puzzolana additives. The structural performance of the domes is related to these elements
26
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 7
that act as shells: transferring all dead loads produced by its own weight, generating friction
and compression stresses in the two main directions and distributing the loads from the top to
the supports (Meli, 1998). Tangential stresses in compression are also produced in the top
and tensional in the bottom, where cracking and separation can be produced. The big
majority of damages are due to tangential stresses that are in compression in the top part but
that take some tension values when reach the base. Due to the low resistance in tension of
masonry, this stresses origin a cracking across the meridians virtually separating the dome
into a series of stripes.
The bulging problem is not critical in masonry domes, the minimum depth
commonly used for its construction is sufficient to avoid this mode of failure. As stated
before, the radial stresses of compression increase progressively from the top to the supports;
the preceding, guided engineers around the globe to solve it by decreasing the depth in the
crest part of the structure; in many historic constructions, nervures where used to reduce the
weight of the element.
2.4.1 Foundations
The necessity for transmitting loads from support elements to the ground with
generally lower strength than the composite materials of the elements in the superstructure,
gave place to the requirement of a substructure (a transition structure between the ground
and the superstructure). This gave place to the appearing of foundations. This improved from
the employment of single support elements to form raft or strip foundations.
Foundations in historic masonry structures are one of the most important elements to
be considered for rehabilitation work. Many of the cracking in masonry walls of the structure
is due to consolidation of the ground. When this kind of problem is detected, one common
method to strengthening is by underpinning the entire foundation.
The mechanical properties of masonry vary from section to section and depend on
the quality of stones and mortar and the interaction between these materials, which lies in
general on its internal distribution.
27
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 8
To verify the mechanical behaviour of a masonry wall, Berto, Saetta, Scotta, &
Vitaliani (2005) carried out a computational simulation by finite element model analysis
using an isotropic 3-dimensional micro-model based on the hypothesis of strain equivalence.
Two cases were considered and studied. Figure 2-9 shows the deformed mesh at failure of
the first case study; mortar weaker than units. Figure 2-10 represent the failure mode of a
masonry prism subjected to uniaxial compression load with mortar stronger than brick units.
Figure 2- 10 Case 1. Contours of damage variables at failure. (a)Tensile damage. (b)Compressive damage.
(Berto, 2005)
28
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 2 9
Figure 2- 11 Case 2. Contours of damage variables at failure. (a)Tensile damage. (b)Compressive damage.
(Berto, 2005)
The following table shows typical values of materials used for the construction of
Spanish historic masonry buildings.
29
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 0
stones
The main difficulty to determine the structure response and failure mechanisms
comes from the variability of the materials, mechanical properties and building techniques
applied during construction.
According to Mistler, Butenweg, & Meskouris, (2006) a masonry wall can fail in
different ways.
· Shear failure. Characterized by cracks in the mortar and where the case of a strong
mortar – low strength brick is presented, the cracks simply bisect the bricks (Figure
2-11(a)).
· Friction failure. This type of failure is presented when a strong lateral load is applied
to the structure in combination to the low vertical loads of the structure (Figure 2-
11(b)).
· Bending failure. This occurs when the sections are slender in relation with the
vertical loads (Figure 2-11(c)).
Figure 2- 12 Failure modes of masonry panels. (M. Mistler & C. Butenweg & K. Meskouris, 2006)
“The majority of these problems may occur due to decay of masonry materials and
subsequent weakening of the structures, the weak tensile and shear strength of unreinforced
masonry and the inadequate interconnection of the masonry elements” (Hassapis, 2000).
30
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 1
In this case four new failure mechanisms are likely to occur (Basilio, Oliveira, &
Lourenco, 2004).
Moreover, in addition to the usual stresses parallel to the fibbers, the curved shape of
arches originates stresses with a component normal to the fibbers, which may lead to the
detachment of the reinforcement from masonry (Oliveira, Basilio, & Lourenco, 2006).
31
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 2
Sliding between the fibbers and its support is usually neglected since shear stresses
at the FRP-masonry interface are of minor magnitude (Valluzzi & Modena, 2006). Also FRP
tensile failure is not likely to occur due to its high tensile strength.
One big disadvantage of historic masonry structures consist in the very low tensile
strength of its constituents triggering in an absence of capacity to transfer tensional forces or
bending moments between adjacent elements. Moreover, the inertial forces are not correctly
transferred to the ideal elements to resist lateral forces.
The shear and tensile forces developed during earthquakes usually result in cracks at
the main body and disconnections of walls in the intersections.
Cracks can develop diagonally and run either partially or completely through the
masonry piers between window openings due to tensile stresses, or horizontally in masonry
piers between window openings due to alternating bending moments or diagonally above the
wall opening due to shearing (Hassapis, 2000).
In addition, for the case of isolated buildings four main mechanisms were identified
for non repaired structures (Penazzi, Valluzi, Saisi, Binda, & Modena, 2001):
1. Out of plane of loadbearing walls with local or total collapse of the facades or of
the corners, or large deformation of the walls. This mechanism is due to the lack of
connection between orthogonal walls (Fig. 2-13(1)) and between walls and floors or roofs
and to the presence of large openings (Fig. 2-13(2)).
2. Out of plane mechanisms with local or large failures of the upper part of the walls
and collapses of parapets, cornices and spandrels. This occurs due to the thrust of the roof
and absence of connection between the roof and the masonry (Fig. 2-13(3)).
3. Wall disconnection and leaf separation with local or global failures. The presence
of in-homogeneities in the wall, the lack of connection between the leaves of multiple leaf
walls (Fig. 2-7(4)), the filling of openings without good connection between the old and the
new parts or the use of different types of materials can be the causes of such mechanism.
4. In plane mechanisms due to shear stresses with diagonal cracks of piers and walls
at the different floors. They are mainly due to: bad positioned openings, differential stiffness
of the walls between openings, presence of weak lintels (Fig. 2-13(5)).
32
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 3
Figure 2- 14 (1)Failure of a building corner, (2)Out of plane collapse of a bearing wall, (3)Partial collapse
due to the thrust of the roof and bad connection tie beam-wall, (4)Separation of the two leaves of a wall,
(5)Shear failure of a wall. (Penazzi, 2001)
The elements that present more risk to collapse in case of seismic events are: slender
towers, columns and isolated walls. According to Meli (1998), the failure mode in towers is
less critical than suspected. During the vibration of these elements a successive process of
opening and closing of cracks is presented by actions, bending and sliding of the joints and
horizontal cracking. Such process dissipates energy increasing the damping in the elements
and reducing any collapse threatens.
The addition of new buildings to old ones represents another source of problems
under this category. In particular, (Bidwell, 1977) argues that the following considerations
should take place when new extensions are added to already existing buildings:
· Whether and how to tie the new structure to the old one.
· The effect that this may have on the sub-soil conditions and the foundations.
· Whether they will trigger any direct or indirect stresses to the old structure.
33
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 4
“Differential settlements are common (due to the addition of new buildings)” (Hassapis,
2000).
Generally, many historic buildings do not present proper foundations; so, initial
settlements take place in the first years of its life (Cook & Pegram, 1993). However, such
building continues working firmly without compromising the safety of the structure;
normally such changes are likely to stop at an early stage, since structures usually adapt
themselves to their environment reaching to a new equilibrium of internal stresses, especially
when soft mortars were used during the construction of the structure (Feilden, 1994).
According to Feilden, (1994) common causes responsible for subsidence are the
following:
· Heavy structures built near old ones, which may change ground strains.
· Vibrations caused by heavy traffic passing from adjacent to an old structure road or
by pile driving operating near the structure.
34
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 5
· Earthquakes
“When unequal settlement of masonry takes place, the appearance of cracks at the walls
is a common result of the in-plane deformations of the walls” (Hassapis, 2000).
Cook & Hicks, (1992) argue that the allowed limit before a structure is deemed
unstable due to creep is dependent upon a variety of factors such as the flexibility of the
materials, their individual susceptibility to creep, the tolerance of the joints between the
different components and the interconnection of the structural elements, all of which are
usually more flexible in this type of structures.
There are many ways of crack to appear; due to moisture, thermal movements and
cracking.
2.6.1 Introduction
35
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 6
Even at present our knowledge about the mechanical behaviour of masonry is not
well known as for other materials and many calculation methods for capacity assessments
hardly ever consider the complex behaviour of masonry as a ‘composite’ material.
Another element that increases the use of this method nowadays is the computed
aided engineering (CAE) packages, capable of solve different types of structures under many
different solicitations. The technique of general application for this purpose is widely known
as the Finite Element Method (FEM), it consists in the division of the structure into sub-
elements for which equilibrium and deformation equations are already assigned; boundary
conditions are established in the joints intersecting to two or more elements (Zienkiewicz &
Taylor, 1967).
In the world there is still a preference for the intuitive a qualitative judgment of
historic monuments; however, for the engineer it is always important the support of
analytical models, as well as laboratory test of the material properties.
As discussed before, during the last decades there has been an important
improvement in the experimental and analytical methods for the studies of historic buildings,
also now we can count with powerful analytical tools that allow us to solve complex
structures with a reasonable computational cost. The weak point in these models is the
application of the procedures to determine the parameters and models that define the
response.
36
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 7
It is common to call static methods to those based on the hypothesis in which the
structural material has a linear behaviour, both in tension as compression; the interaction
between the internal deformations and applied stresses increase proportionally. This
hypothesis has allowed to obtain exact solutions for typical structural models, in which, the
equilibrium conditions are satisfied.
The main objection in the use of the elastic methods is that they don’t recognize the
non-linear behaviour of the masonry.
In summary, accurate results can not be obtained from a static analysis but can be
considered as representative of the structure and be used for the assessing.
Most of the non linear models consider that the properties against tension stresses
are the same than those of compression and that both are invariable against any load. These
limitations are surpassed when in the models the materials are linear in compression, but
have zero strength in tension. The solution of these models requires a nonlinear analysis
method, because the level of load is increased, the size of area under tension increase,
becoming necessary to modify the characteristics of the model during time (Berto, Saetta,
Scotta, & Vitaliani, 2005).
They are particularly useful in that they provide for (Petkovski, 2007):
37
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 8
• More realistic estimates of deformation demands for elements that must deform
inelastically in order to dissipate energy imparted to the structure by ground motions.
• More realistic estimates of inter-story drifts that account for strength and stiffness
discontinuities that may develop during inelastic response.
The calculus of the effects over the structures produces external variations; it is
considered that the last has constant values in time (that proceed in static form). This is valid
for its own weight and for some solicitations as differential movements and contraction due
to changes in temperature.
Probably the must important tool for dynamic analyses can be the eigenfrequencies
and modal shapes. This allows evaluating the importance of the dynamic effects induced by
external agents.
To solve this, preference has been given to the employment of dynamic elastic
analyses, where the stiffness properties are modified manually. As a result, many
methodologies have been proposed, see (Ramos & Lourenco, 2005; Mistler, Butenweg, &
Meskouris, 2006 and Cardoso, Lopes, & Bento, 2005), where the modifications are guessed
from the result of the first analysis and results are compared according to the seismic
vulnerability of the structure.
For the structural modelling of masonry historic buildings the use of a three-
dimensional finite element model become necessary to obtain accurate results. This
38
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 3 9
numerical analysis is also found to be questionable due to many factors: the anisotropy and
decay of the constitutive materials, the fact that consists of two materials with different
properties and without any tensional strength. These factors trigger the developing of cracks
at any section affecting the global characteristics of the element. But, in spite of many
uncertainties, though, numerical models are considered as a very useful tool that used in
combination with experimental work can be very effective in the assessment of historic
masonry buildings.
The first attempt for the analytical treatment of monuments built of fitted stones
under dynamic excitation was presented by Housner (1963).
In actuality, to carry out a structural analysis for old masonry structures, many
engineers model the materials linked to linear behaviour without consider the whole complex
performance of masonry (as a composite material with anisotropic properties). This method
does not describe a realistic performance of the structure but, as stated before, gives an idea
about the main causes of failure and its more accessible and less cost-time consuming in
comparison with other methods, e.g. non-linear micro-models.
Lourenco (2005) point out that it is essential to verify the adequacy of the models
with the existing building; this can be carried out with different techniques, mainly flat-jack
testing or dynamic identification, but also a comparison with the damage survey (cracks,
displacements) is allowed.
The finite element method of analysis makes use of different finite elements: One-
dimensional (bar, frames), Two-dimensional (shell, plate, membrane) and Three-dimensional
(solid). For the simulation of the influence of composite materials in the structure, micro or
macro-analysis can be used, depending on the accuracy desired. During micro-analysis,
blocks, mortar and the interface are simulated separately simulating the contact nonlinearity,
while for macro-analysis, a homogeneous material represent the masonry behaviour. The
39
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 0
deformability in this type of models is considered by discretization of the mesh into finite
elements.
For the first previous mentioned approaching method, the accuracy found in the
results is very high, but is very expensive computationally speaking been suitable only for
small elements.
Figure 2- 15 Different FEM approaches for modelling masonry. (M. Mistler & C. Butenweg & K. Meskouris,
2006)
The historic masonry buildings are mainly very heavy and structurally stiff.
Therefore, high inertia forces that depend on the product of the mass and acceleration are
generated. As a result, the frequency of the vibration modes is typically between 1 and 4 Hz;
located in the interval where the dominant frequencies of dominant earthquakes are found
(Meli, 1998). Moreover, Cardoso, Lopes, & Bento, (2005) argued that a far distance
earthquake, with low frequency contents, is the one that induces higher accelerations to this
type of structures. This directs us to observe that the accelerations presented in this type of
buildings are very elevated and the damages are considerable.
40
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 1
Architectural walls and partitions can affect the stiffness of structural elements and
also introduce soft story and torsional conditions into otherwise regular buildings. Therefore,
many engineers having this in mind add extra lumped masses to simulate the effect of non-
structural elements in the total behaviour.
During any rehabilitation work special care most be taken for discontinuity in the
flow of forces, alterations in the existing structural systems and local differences in rigidity
could lead to shifts in load transfer that inevitably become into cracks.
41
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 2
When working on historic and, in particular, listed structures, repairs should ideally
be carried out using similar materials to the original. Not only are more appropriate to the
historic character of the architecture, but they usually work better than modern alternatives,
especially when used in conjunction with other traditional materials and construction
techniques (Bennett, 2002).
2.8.1 Repointing
According to (Ashurst, 1988c) pointing is the process of filling the outer part of the
joints between masonry units, where the bedding mortar has been deliberately left or raked
back from the surface or where the original mortar has weathered.
The selection of the appropriate mortar becomes a very important issue in the
effectiveness of this technique. A soft mortar than the units can absorb stresses developed
through movements of the adjacent masonry, while a hard mortar will produce cracks
between mortar and bricks or will cause spalling (Figure 2-15).
Figure 2- 16 Impact on masonry of using (a)Repointing with soft mortar. (b)Repointing with hard mortar.
(Pearson, 2007)
42
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 3
2.8.2 Grouting
“Is a term used for the introduction into a structure or into the ground of a material
in liquid form, which subsequently cures or sets into a durable solid or gel form” (Hassapis,
2000).
The degree of injection depends on the specific components of the old masonry
being repaired or strengthened and on its structure and moisture content, on the composition
of the suspension as well as on the procedure selected to prepare and carry out the injection,
including the applied pressure.
There are various methods of grouting, the most important of which are presented
next: hand grouting, pressed injection or pumped system, vacuum system and gravity
grouting.
2.8.3 Pinning
Pinning is more repair technique than a strengthening one. The process involves the
insertion of pins into holes left by missing stones and drilled holes. During the process it is
important to replace any missing pinnings or loose stones back into their original position
and to ensure that all the replaced stones make physical contact each other. It is also used for
replacing old masonry units or for redressing the faces of old units. Small fractures and voids
can be also repaired with this method. The insertion of the pins removes the resins or grouts,
which, consequently, flow into internal cracks and fissures (Coonie, 1992).
43
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 4
In doing so, this gives the wall back its full structural strength and original
appearance. Any missing pinning stones will generally be revealed by the remaining wide
joints.
2.8.4 Stitching
For seismic strengthening of historic masonry buildings, the solutions usually consist
in corrections that guide us to a better behaviour against any movement in the base.
44
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 5
· Fatigue characteristics
· Electromagnetic Neutrality
· Effective solution in aggressive environments
· Temperature resistance
· Low density
As a result, all this advantages makes FRR highly attractive and cost effective to be
used as a common material in strengthening works. Moreover, the use of FRP in special
applications in construction becomes cost-effective due to durability improvement, reduced
life-cycle maintenance cost and also savings from easier transportation and enhancement on
site-productivity (Triantafillou & Fardis, 1997).
Schwegler (1994) was the first person to propose and study the use of carbon
laminates as strengthening elements for masonry structures (Figure 2-16).
Figure 2- 17 Masonry strengthening solutions with FRP composites (Oprisan at. al. 2004)
Carbon, glass and aramid fibres exhibit a linear elastic behaviour under tensile
loading up to failure. Due to the absence of plastic flow at yield, FRP composites are
incapable of relieving stress concentrations.
45
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 6
Carbon fibres show high specific strength and stiffness, as the elastic modulus
increases, ultimate tensile strength and failure elongation decrease. Compressive strengths of
FRPs are much lower than the tensile strengths (Table 2-2).
One of the main conclusions from studies is that, for the sake of both economy and
effective mechanical response, unidirectional FRP reinforcement in the form of strips (that is
100 - 300 mm wide bands) is preferable than two-dimensional fabrics that cover the whole
surface.
These elements work by gravity transferring all the loads to walls and buttresses.
The instability is presented when there is an opening in the supports, modifying the centre of
pressure. In the Figure 2-17 is illustrated a typical repair of these elements. Basically consists
in the consolidation of masonry and reinforcement with layers of mortars and reinforcement
bars.
1
MAC S.p.a., Treviso, Italy.
2
GAVAZZI S.p.a., Lecco, Italy.
3
Produced in laboratory.
46
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 7
Landfill Arches
Buttresses
Longitudinal wall
Column Longitudinal
wall
a) b)
Figure 2- 18 (a)Reinforcement of vault with mortar. (b)Reinforcement with reinforced concrete.
Numerous experimental works were carried out showing that FRP technique is
effectively valid as an option to strengthen or repair masonry structures, in particular arches
ones, see (Valluzzi, Valdemarca, & Modena, 2001), (Lissel & Gayevoy, 2003) and
(Foraboshi, 2004) for further details. On the other hand, available experimental results show
that the strengthening of masonry arches with glass fibbers, which exhibit lower mechanical
properties than carbon ones, allow a better control of the collapse mechanisms and provide
higher strength and better global ductility characteristics (Valluzzi, Valdemarca, & Modena,
2001).
For both full length and partial length reinforcements, the optimal width is
approximately 14cms. Beyond this value only marginal increase in the load values is
observed.
Basilio, Oliveira, & Lourenco (2004) argued that the adoption of full length CFRP
reinforcement instead of partial length reinforcement at the intrados originates an important
increase in the peak loads. Consideration must be taken when simultaneous partial length
reinforcement option is used, since the location of plastic hinges might change causing
openin of the joints.
47
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 8
Figure 2- 19 Summary for the full length CFRP reinforcement at the intrados and extrados. (Basilio, Oliveira, &
Lourenco, 2004).
48
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 4 9
Many issues are relevant with respect to the strengthening of masonry arches with
composite materials. It is argued (Basilio, Oliveira, & Lourenco, 2004) that when the CFRP
reinforcement is placed at the intrados, it contributes to holding the bricks together. Also,
experimental research (Figure 2-19) has showed that for arches strengthened at the extrados,
sliding along the joints is the prevalent failure mechanism. A solution can be resolved by
increasing the amount of material near the abutments.
If we follow these experimentations, the fibres are able to maintain equilibrium until
total collapse of the specimens by adding continuous reinforcement at intrados. Another
important feature is the large deformation capacity exhibited prior to failure, providing to the
arches with important ductility behaviour.
49
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 0
3 CASE STUDY
The selected structure (Figure 3-1) is an old temple located in San Cristobal de las
Casas, Chiapas. In order of importance, the Santo Domingo Church was second after
Guatemala, but the first Dominican settlement in south of Mexico. The first stone was placed
by Francisco Marroquín, a Franciscan bishop of Guatemala, the 19 of January of 1547 and
the entire construction was completed in the year of 1551 (Sidney, 1984).
Figure 3- 1 The front entry of the Church of Santo Domingo de Guzman in San Cristobal de Las Casas.
(Agguizar)
50
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 1
Both Convent and Church have been under repair, modifications and even total
reconstruction during more than 400 years. Though the temple was built in the second half of
XVI century, very little from the first edification actually remains. The main body and the
adjacent modules probably come from the last third of XVII century, while the monumental
facade was built in the beginnings of the XVIII century (Sidney, 1984).
Taking advantage of the location of the Church (into a seismic active region) an
assessment of its seismic behaviour is carried out and presented in the following chapters.
a B C D E F G H
There is a certain variation in the depth of the walls due to the constantly repairs
carried out in the past. This works led to enhance the reinforcement, such as increasing the
51
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 2
dimensions of the masonry walls and addition of buttresses. The depth of the walls is as
follows: in the front wall, is around 3.10m, from which 90cms can be counted as
architectural detailing; the towers in the front are solid buttresses made of masonry with
1.30m in depth and 2.30 in width; the choir walls are 2.30m in depth; the walls supporting
the vault are thinner, approximately 2.00m; the longitudinal walls in the cross are 2.30m at
floor level, but reduced in the niches formed by arches, giving us a dimension of 1.55m to
1.60m in depth.
The height of the structure from the base to the towers is of 27.46m; from the base to
the vaulted roof, 16.31m. The main facade is clearly of baroque inspiration (Figure 3-3).
Masonry walls are conformed by a combination of local quarry stones and lime
mortar mixed with agave sap added as additive to increase the viscosity and gluing
properties of the mixture. The roof in the beginning was build with wood; after its collapse,
adobe bricks with lime mortar were used to form the vaulted roof.
52
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 3
The Material distribution of the Temple is shown in Figure 3-4 and Table 3-1; this
includes all the material values used in the numerical model. These values were based on
values proposed by (Meli, 1998).
Even when in the last decades important improvements have been achieved in
analysis techniques, the preparation of any analytical model confronts some difficulties. The
geometry is a lot more elaborated than for modern buildings and in many cases is very
difficult to distinguish between structural and non-structural (decorative) elements. There is
also an uncertainty about the materials employed for its construction; in consequence some
information related to the mechanical properties of the materials is not accurate.
The developing of the numerical model first started with the generation of a 3D
geometry model based on the drawings and information taken by previous data.
53
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 4
Due to the complexity of the structure geometry, it was divided into sub-models.
Each one was built using isoparametric three and four-node shell elements. The adjacent
temple surrounding the Church in one of the sides, which has positive influence on the
dynamic behaviour, was not considered leading to conservative results.
The basement conditions were found satisfactory and no failure in the foundations
has been reported during the life of the structure. Therefore, the boundary conditions of the
numerical model were modelled rigidly fixed at the bottom and did not include any soil
structure interaction effects.
Many experienced engineers in the area agree that for the use of a homogenized
model, the nonlinear behaviour of the material properties needs to be calibrated with the
modal characteristics of the structure and some testing must be done to obtain some
parameters (Binda, Penazzi, & Saisi, 2003; Lourenco, 2006; Meli, 1998), but in accordance
with Mistler, Butenweg, & Meskouris (2006) and due to absence of information, the
characteristic values of the constituents (which are known for a wide variety of brick and
mortar types) were used as input variables of the numerical model. In addition, different
depths were used in the vault, taken into account the recurrent variation from thin to thick as
the arch approach to the supports.
Shall be considered that the walls are in reality composed by rocks of different sizes
and in some areas the density of these varies in relationship with the mortar.
The adopted model validation technique was executed by means of deformed shapes
evaluation, comparison with the damage exhibited by the church in the reality and
correlation of resulting fundamental frequencies with similar works.
After completion of all sub-models, the overall structure was assembled and
exported to SAP2000® to execute multimodal, response spectrum and time history linear
analyses.
It must be clarified that even when this model is useful for a seismic assessment,
must not be considered as an exact representation of the structure. It is common that a
precise estimation of the stresses in this type of structures is difficult to obtain. Many issues,
as nonlinear material properties and ground conditions were not integrated into the numerical
simulation. In some cases, there were some uncertainties about the correct size of the
54
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 5
materials depth into each wall. So, results obtained with linear behaviour modelling can not
be supposed as exact, but instead can be considered qualitatively correct.
A model (Figure 3-5) built with 711 frame elements and 515 shell elements was
employed to study the stress distribution and failure mechanisms in the arches. Due to the
simplicity of the structure and for more confidence in the results; a refined mesh employing a
discretization of 0.5mx0.5m per element was produced.
a) b)
Figure 3- 5 (a)Finite Element model of a single arch, (b)Discretization of the model.
55
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 6
This model was employed to observe the stress distribution due to seismic actions; to
visualize the deformations in the towers and arch elements and the overall capacity to resist
lateral movements.
As it has been explained in the first part of this document, it is normal to expect a
good static behaviour from arches and vaulted structures. The best prove of its good load-
carrying behaviour is that are still there remaining with the time.
Another important aspect of the static system in terms of horizontal loads is the way
the forces are distributed in the connexion of the main copula with the perimeter walls and
pendentives.
The most typical failure in vaulted areas is due to load concentration at any point in
the spanning section. This can be minimized by adding landfill at the extremes in the
extrados face of the vault. This increases the amount area to be in compression in sections
56
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 7
near to the supports. In this way the pressure line is maintained in compression in this
section.
Many repairs were carried out through its existence. A new wall had to be rebuilt
due to its partial collapse during a seismic event in XIX; years later, the whole roof was
substituted by a masonry vault (Figure 3-9). In addition, minimal repair works have been
carried out during the last years. Restitution was made in the mezzanine and roof floors and
the rear facade was repaired by grouting the cracks (Figure 3-8).
An extra section was added to the building in module C during the XX century. This
section has helped to the structure to withstand several earthquakes through the years.
The last repair works carried out on the structure includes: Changing in the
fulfilment of the vault, by cement:sand:lime aggregate. Cleaning and painting of decorative
elements. Any structural interventions have being taken place in actuality.
The next figures show the areas that have been repaired during the existence of the
structure.
1 2 3 4 5 6 7 8
41.32
20.98
18.15
13.63
9.11
GROUTING OF CRACKS
57
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 8
a B C D E F G H
63.44
1.66
7
1.59
3.95
1.59
6
5.32
15.09
6.85
PROYECCION
5
1.72
1.39
2.50
1.81
4
6.06
5.40
2
CASA CURAL
RESTITUTION OF ROOF
8.29
MASONRY WALL SUBSTITUTION
1
3.02 3.55 1.50 3.32 1.53
12.93
As stated before, to calibrate the computations that define the capacity degradation
of the numerical model; the results obtained from the analytical calculations where compared
with the damage presented into the original structure. With this, an attempt was made to
model dynamic response in a simple way suitable for computational analyses, but with
satisfying final results on the behaviour at individual levels (Gavrilovic, Ginell, & Sendova,
2001).
The process to establish a final model was: geometry definition, drawing creation,
boundary definition, material definition, load cases definition, dynamic properties definition,
analysis process, connection checking, behaviour checking, result comparison, fixing of
tentative problems in the numerical model.
After the analytical model is established, the structure is divided into 3 modules in
order to simplify the interpretations of the results. The following figure illustrates such
distribution as well as joint locations later used in time history analysis.
58
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 5 9
Joint 591
Joint 588 Module A
Joint 591
Joint 578
Module B
Module C
a) b)
In the following, pictures with different stress contours of the structure are shown.
Negative values are treated as compressive stresses and positive values as tensile stresses.
All values in the figures are indicated in kg/cm2.
For assessing the stress distribution in a typical single arch, a four-node shell
element based model was employed.
a) b)
Figure 3- 11 Principal maximum (a) and minimum (b) stresses under dead load.
The Figure 3-11 shows the principal stress distribution in the structure. It can be
notice from the drawings that 4 plastic hinges are likely to develop. This stresses are not
59
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 0
bigger than 5kg/cm2 for compressive stresses but increases near to 13kg/cm2 in some small
individual areas. Very low tensile stresses are observed to take place.
In spite of having a heavy structure, the stresses due to vertical load are slightly
important comparing with the compression strength of masonry. The strength considered for
masonry is in the range of 15 kg/cm2. Probably for the masonry ordered in the core the
strength is even greater than considered.
For the gravity load case, the own weight of the structure produces compressive
stresses in most part of the vaulted area, whit low tensions in some zones. This tensional
stresses are known to be prompted to develop a failure mechanism in elements with the arch
configuration. The common zone were tensional stresses appear is in the bottom at middle
span of the arches; see (Figure 3-12).
It can be notice that the maximum compressions in most cases are lower than
3kg/cm2.
a) b)
Figure 3- 12 (a)Principal compressive stresses under dead load. (b)Principal stresses of main arches under
dead load.
60
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 1
The localized damage of the structure, the stability of the structure without any
differential movements detected, and the excellent conditions of the foundations, indicate
that must of the observed past damage is due to seismic events.
For the modal analysis, the resulted fundamental modes are lower than expected;
such difference can be attributed to the developing of cracks and foundation flexibility in the
real structure, which were not taken into account for this study case.
The Table 3-2 shows the natural periods and modes of vibration for the first 10
modes. Complete tables of modal participating mass ratios and modal periods and
frequencies can be found in Annexe 3.
a) b)
Figure 3- 13 (a) First mode of vibration (T=0.28sec). (b) Second mode of vibration (T=0.23sec).
In the first two modes of vibration (Figures 3-13) both towers are excited in X-
direction. This mode has an effective mass of 2%, confirming that is a particular vibration
mode of the towers.
The third mode (figure 3-14) acts along X-direction exciting one side of the
structure, affecting both tower and the module C. This mode has an effective participating
mass of 12%, becoming an important mode for the structure.
61
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 2
The fourth mode (Figure 3-15) is related to the excitation of the structure in Y-
direction having a torsional effect on the structure. In addition a sagging effect can be
localized in the walls of the main vault. The constant lateral movement of the structure origin
the development of compressive and tensile stresses in the adjacent buildings, which act as
buttresses of the main body. The differences in height between the two buildings cause a
negative interaction in the walls and cupola. This mode is the most important for the
response of the structure against seismic actions because it affects 44% of the total mass.
The following modes are less important, and basically excite the structure in a
torsional mode.
62
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 3
The seismic area under which the Church is stand was found B, corresponding to a
C=0.3g. The predominant type of soils corresponds to a very high load capacity soil formed
mostly by rocks.
The response spectrum was defined according to the Mexican code (Mexican-
Norms, 2007), which is the one used in the local area (Figure 3-16). The response spectrum
analyses were carried out in both longitudinal and transversal directions. The internal forces
were computed according to the SRSS rule and superposed with 30% in each direction in
accordance to the NTC (Mexican-Norms, 2007).
ERS type 1
0.3276
0.3024
0.2772
0.252
0.2268
0.2016
Sd(g)
0.1764
0.1512
0.126
0.1008
0.0756
0.0504
0.0252
0
0 0.5 1 1.5 2 2.5 3 3.5 4
T(s)
The code requires that the effective mass of the considered modes to be greater than
90% of the overall mass of the structure and that every mode with an effective mass greater
than 5% must be taken into account. These requirements were satisfied by using the first 50
modes.
For the study of seismic actions, the displacement of the structure was restrained in
its perpendicular direction.
63
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 4
a) b)
Figure 3- 17 (a)Shear stressa)diagram and deformed shape. (b)Stress diagramb)under earthquake loading in
principal direction.
It was observed tensional stresses lower than 1kg/cm2 in most of the structure, but
surpassed in the bottom of supports and arch connections (Figure 3-17).
The finite element model previously discussed and shown in (Figure 3-6) and
(Figure 3-7) was presented for the study of the seismic structural behaviour of the Santo
Domingo church. All the dimensions where obtained from drawings of previous repair
works.
Seismic actions are frequently variable. For earthquakes of low intensity it can be
expected that all sections present a minimum displacement, maintaining the section under
compression or perhaps with tolerable tensions. Nevertheless, in strong events, the structure
can develop high tensile stresses. In the actual condition, these tensions could origin the
collapse of the towers.
The linear response spectrum analysis of the global structure shows that the design
spectrum would produce maximum tensile stresses of around 3 kg/cm2 at the towers and
walls of module C (Figure 3-18). This clarify that with the addition of an extra section in the
Module C, the structure gained more arguments to withstand lateral forces and avoid the
collapse of one of the cupolas.
64
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 5
a) b)
Figure 3- 18 (a)Principal stresses in Y-direction due to earthquake. (b)Principal stresses in X-direction due
to earthquake.
The Figure 3-19 shows in a magnified scale, the maximum displacements originated
by the design earthquake acting in X and Y directions. The maximum displacement is
presented in the top part of the towers, and it’s about 2cms in X-direction. It can be notice
that in the main structure the maximum horizontal displacement is of 0.5cms. For the
displacements along Y-direction it shows that the vaulted body displaces around 2cms, this
behaviour can lead to the detachment between the vault and its supports.
a) b)
Figure 3- 19(a)Displacement contours originated by the design spectra in Y-direction. (b)Displacement contours
originated by the design spectra in X-direction(Contours in cms).
Estimating that all the components will remain intact, the structural system shows a
good behaviour against its own weight and low intensity earthquakes. The forces are
transmitted axially with low stresses due to the big size of the elements. This allows masonry
65
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 6
to resist moderate compressive and shear stresses. Otherwise, a large earthquake will cause
local damage causing arches and vaults to separate from its connection with the walls.
I would like to mention that the real structure shows a cracking in the middle span of
the vault. According to these findings it should be considered that the geometry of the vault
is suitable enough to avoid a separation between the vault and walls, but instead of this, a
plastic hinge is developed at the middle span.
As result, the computations with the response spectrum method show a good
structural behaviour of the church against low seismic events, but for the case of high
intensity earthquakes, local failures leading to local collapse in the vault are substantially to
take place and must be studied by refined methods.
Time History analyses were carried out to verify the results from the response
spectrum analysis and to compute the internal forces and displacements as a function of time.
For this purpose, recordings of the 21 of October of 1995 earthquake on hard ground at a
distance of about 100 km from the epicentre and with a magnitude of Ms=6.2 were used.
The record samples the near field strong motions that triggered damage to some buildings in
the city. The duration of the recording of the motion is about 165 sec. and the maximum
accelerations are 348.61cm/s2 in the N-S direction and 441.95cm/s2 in the E-W direction
(Figure 3-20). For the computation the accelerograms are applied in two horizontal
directions as in the response spectrum analysis.
a) b)
Figure 3- 20 (a) 1995 earthquake accelerograms in N-S direction (b) 1995 Earthquake accelerograms in E-W direction
Tensile stresses of 2kg/cm2 above the limit of the proposed material strength are
observed in the two towers (Figure 3-21). Tensile stresses are observed in the connection
between the main body and the adjacent module that can lead in the disconnection in some
areas.
66
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 7
a) b)
Figure 3- 21 (a)Stresses diagram in Y-direction [95 Earthquake]. (b)Stresses diagram in X-direction [95
Earthquake].
The next figure shows the development of relative low tensional stresses in the
intrados of the main arches. To reduce the detachment of units at the extrados face, the
addition of a GFRP plate is proposed to increase the tensile capacity of the structure and to
add an extra bonding between the constitutive elements.
The Figure 3-22(b) shows the stresses contours of the main facade, having
significant tensional stresses in the towers, from about 1kg/cm2.
a) b)
Figure 3- 22 (a)Stresses diagram in X-direction [95 Earthquake]. (b)Maximum envelop of stresses in X-
direction [95 Earthquake].
67
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 8
a) b)
Figure 3- 23 (a)Stresses diagram in Y-direction [95 Earthquake]. (b)Stresses diagram in Y-direction [95
Earthquake].
The Figure 3-24 shows the safety of the arches in the mezzanine and vault section.
No high tensile stresses were found to occur during the applied accelerations, it means that
the landfill above the mezzanine and the big size of the sections establish a good system
against this type of excitations. Again, this analysis was carried out without considering any
sliding in the joints.
a) b)
Figure 3- 24 Maximum envelop of principal stresses in X-direction [95 Earthquake].
The time history analysis shows relative displacements in the central arches of
building-A (vaulted body); the stresses are not significant, around 0.4 kg/cm2. The local
displacement is about 0.2cms and can not lead to any sliding in the supports and activate a
typical mode of failure (Figure 3-25(a)). At difference with the response spectra, this
earthquake actives a typical torsional effect in the structure and towers, where the central
arches suffer from lateral displacements (Figure 3-25(b)). This zigzag movement can cause
an out of plane mechanisms in the top part of the longitudinal walls affecting the connection
with the vaulted roof.
68
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 6 9
Joint 1339
Joint 1344
a) b)
Figure 3- 25 (a) Maximum envelop of principal stresses in Y-direction. (b) Deformed shape under recorded
accelerations [95 Earthquake].
The maximum displacement found occurs at the arches of the vaulted body
(0.22cms) (Figure 3-26). The difference in displacements between the top and bottom part
indicates why this section of the structure presents higher concentration of tensional stresses.
a) b)
Figure 3- 26 (a)Displacement contours originated by the design spectra in X-direction. (b)Displacement
contours originated by the design spectra in Y-direction(Contours in cms).
The complete results of the displacement distribution in the 4 selected joints over the
time can be found in Appendix D.
Therefore, the results obtained by the time history analysis prove to develop similar
tensional stresses to those found with the response spectrum case, but with the only
difference that the torsional vibration mode is more obvious in the time history case.
69
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 0
The principal stresses where slightly lower and it was found that the shear stresses
do not have a significant effect in the structure
According to the findings, the actual condition of the church is safe against gravity
loads, settlements and low intensity earthquakes. But the seismic analyses demonstrate that
one common failure mechanism in this type of structure is the tendency of the tower to
separate from its base triggering in failure at the connections between the main facade and
the vault; causing towers to continue working as isolated structures and as consequence a
collapse.
For the arches a strengthening based on layers of GFRP fibres in both extrados and
extrados faces, is proposed, to give the sections more capacity to resist moments, ductility
and lateral movements.
The remedial measures were considered taking into account the modern principles of
architectural heritage protection, specifically, minimal repair, unobtrusiveness, removability,
stability, durability and compatibility.
70
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 1
4 CONCLUSIONS
Historical buildings and monuments are an important link between our time and
history. Unfortunately, most historical buildings have already disappeared; damaged or
collapsed because of several external factors. For this reason, it is very important to protect
them immediately, taking into account the importance of choosing the best rehabilitation
technique based on analytical and numerical assessments, so action can be taken ascertaining
that any work will lead to a better behaviour of the structure.
The thesis try to demonstrate by analytical tools and failure comparisons, how to
predict and prevent an eventual failure or collapse of the structure and corroborate that the
basic principles and criteria of structural engineering are valid for any type of construction
and that the methods applied on modern buildings, with clever modifications, can be used in
historic masonry buildings.
For the investigation of the seismic performance of the Santo Domingo church, a
three dimensional numerical model was created and analysed using an advanced earthquake
simulation software (SAP2000®). The model then was corroborated by the measurement of
its natural frequencies. The first mode period is considered a local mode of the towers
(T=0.28sec) and the third (T=0.23sec) and fourth (T=0.22 sec) modes were considered as the
fundamental modes of the structure shaking it in X and Y directions respectively. Such
modes were not expected for a heavy and medium high structure; it means that the structure
oscillate around 4.5 times per seconds. But these findings were correlated with other works
in the field (Meli, 1998) and reviewed by its deformed shapes and then by correct
interpretation of the stress concentrations taken from the numerical results.
From the static analysis was found that the structure of the church is conceived to
support gravitational loads. The geometric form of arches, cupolas and vault is enough to
distribute the forces in a good manner, allowing it to resist actions with the employed
materials. But, in contrast, individual elements as the towers in the front of the building are
flexible against lateral loads. In general is a slender structure which its relationship between
base and height induces a heavy change in the stiffness.
Two main types of analyses were carried out. A linear static response spectrum
analysis (based on (Mexican-Norms, 2007)) and a time history linear analysis (based on
accelerograms from the 21 of October of 1995 seismic event). The results of the static and
dynamic analysis cases demonstrate differences in the deformed shapes. For the response
71
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 2
spectrum case the modules A & B basically moves in the Y-direction and the towers and
module C in X-direction. While for the time history analysis, the module A becomes in
torsion, a zigzag effect is developed in the middle section of the vaulted body due to the high
stiffness in Y-direction of the mezzanine and facade in relationship with this part.
These analyses provided us with very useful information related to the vibration
modes and stress diagrams. The response spectrum analysis shows stresses of around
3kg/cm2 in the towers and in one wall of the adjacent section, stresses that will lead to
cracking. The maximum displacements were found to take place in X-direction (2cms) while
for the whole structure does not surpass the 5 mm. These are considered conservative results.
For the Time history case tensional stresses are developed in the intrados of the arches of
less than 1kg/cm2. Again high tensile stresses around the walls of the adjacent section are
taking place.
Additionally it was found that general elastic behaviour of the structure is relatively
safe against gravity and minor lateral loads. This is attributed to precedent repair works, in
which the addition of buttresses and resizing of walls was correctly decided.
It is agreed the use of FRP for the strengthening works in the towers as additional
reinforcement bars and for the arches based on layers of GFRP fibres in both extrados and
extrados faces.
As commented due to the lack of time and information, the numerical model and
parameters were not fully studied.
The previous work and strengthening recommendations only must be realized under
an extended investigation, in which is recommended:
72
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 3
73
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 4
References
Basilio, I., Oliveira, D. V., & Lourenco, P. B. (2004). Optimal FRP Strengthening of
Masonry Arches. 13th International brick and block masonry conference. Eindhoven:
Eindhoven University of Technology.
BBC-NEWS. (2007, August 18). BBC. Retrieved August 23, 2007, from A. BBC Web site:
http://news.bbc.co.uk/1/hi/world/americas/6952985.stm
Bennett, B. (2002). Lime Plaster and Render Reinforcement. The building conservation
directory .
Berto, L., Saetta, A., Scotta, R., & Vitaliani, R. (2005, March). Failure Mechanism of
Masonry Prism Loaded in Axial Compression: Computational Aspects. Materials and
Structures , 38 (2), pp. 249-256.
Binda, L., Cardani, G., & Tiraboschi, C. (2006). The Difficult Choice of Materials for the
Reconstruction of the Cathedral of Noto. In S. K. Kourkoulis (Ed.), Fracture and Failure of
Natural Building Stones (pp. 185-200). Netherlands: Springer.
Binda, L., Penazzi, D., & Saisi, A. (2003). Historic Masonry Buildings: Necessity of a
classification of structures and masonries for the adequate choice of analytical models. (T.
&. Pande, Ed.) 6th Int. Symp. Computer Methods in Structural Masonry (STRUMAS VI) ,
168-173.
Binda, L., Saisi, A., & Tedeschi, C. (2006). Compatibility of Materials Used for Repair of
Masonry Buildings: Research and Applications. In S. K. Kourkoulis (Ed.), Fracture and
Failure of Natural Building Stones (pp. 167-182). Netherlands: Springer.
Block, P., Ciblac, T., & Ochsendorf, J. (2006). Real-Time Limit Analysis of Vaulted
Masonry Buildings. Computers and Structures , pp. 1841-1852.
Briccoli, B. S., Rovero, L., & Tonietti, U. (2007, January-February). Strengthening Masonry
Arches with Composite Materials. Journal of Composites for Construction, 11 , pp. 33-41.
Cardoso, R., Lopes, M., & Bento, R. (2005, September 2). Seismic Evaluation of Old
Masonry Buildings. Part I: Method Description and Application to a Case-Study.
Engineering Structures (27), pp. 2024-2035.
Christensen, J. B., Gilstrap, J., & Dolan, C. W. (1996). Composite Material Reinforcement
of Existing Masonry Walls. J. Archit. Eng. , 2 (2), pp. 63-70.
De Stefano, A., & Ceravolo, R. (2005). Monitoring and Response of CFRP Prestressed
Concrete Bridge. In F. Ansari (Ed.), Sensing Issues in Civil Structural Health Monitoring
(pp. 85-94). Netherlands: Springer.
74
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 5
Gavrilovic, P., Ginell, W. S., & Sendova, V. (2001). Seismic Strenghtening of Byzantine
Churches in the Republic of Macedonia. International Congress Archi2000. Paris:
UNESCO.
Levy, M. (2006, October). The Arch: Born in the Sewer, Raised to the Heavens. Nexus
Network Journal , 8 (2), pp. 7-12.
Lissel, S. L., & Gayevoy, A. (2003). The use of FRP’s in Masonry: A state of the art review.
International conference on the performance of construction materials, (pp. 1243-1252).
Lourenco, P. B., Melo, A., & Carneiro, M. (2005). Remedial Measures for Cathedral of
Porto: a Post-modern Conservation Approach. 4th International seminar of structural
analysis of historical constructions.
75
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 6
Meli, R., & Sanchez-Ramirez, R. (1993, May). Studies for the Rehabilitation of the México
City Cathedral. Structural Engineering International, 7 , pp. 101-106(6).
Mistler, M., Butenweg, & Meskouris, K. (2006, October). Modelling Methods of Historic
Masonry Buildings Under Seismic Excitation. Journal of Seismology, 10 , pp. 497-510.
Modena, C., Zavarise, G., & Valluzi, M. R. (1997). Modelling of Stone Masonry Walls
Strengthened by r.c. Jackets. Proc. STRUMAS – 4th Int. Symp. On Computer Methods in,
(pp. 285-292). Florence, Italy.
Moropoulou, A., Labropoulos, K., Konstanti, A., Roumpopoulos, K., Bakolas, A., &
Michailidis, P. (2006). Susceptibility of Building Stones to Environmental Loads :
Evaluation, Performance, Repair Strategies. In S. K. Kourkoulis (Ed.), Fracture and Failure
of Natural Building Stones (pp. 291-297). Netherlands: Springer.
Oliveira, D. V., Basilio, I., & Lourenco, P. B. (2006). FRP STrengthening of Masonry
Arches Towars an Enhanced Behaviour. Proceedings of the international conference on
bridge maintenance, safety management (IABMAS'06).
Page, A. W. (1981, September). The Biaxial Compressive Strength of Brick Masonry. ICE
Proceedings , pp. 893-906.
Penazzi, D., Valluzi, M. R., Saisi, A., Binda, L., & Modena, C. (2001). Repair and
Strengthening of Historic Masonry Buildings in Seismic Areas. Archi2000 Session 5:
Architectural Heritage in Seismic Areas. Paris, France: UNESCO.
Petkovski, M. (2007, March). CIV6430: Seismic Design of Buildings. Lecture Notes. Design
of buildings in accordance with eurocode 8 (EN1998-1) . Department of Civil and Structural
Engineering: University of Sheffield.
Ramos, L. F., & Lourenco, P. B. (2005). Seismic Analysis of one Heritage Compound
Building of the Old Lisbon Town. International conference on 250th aniversary of the 1755
Lisbon earthquake (pp. 362-368). Lisboa, Portugal: LNEC.
76
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 7
SAP2000. (2007). Three dimensional static and dynamic finite element analysis and design
of structures. Version 11.0 . CSI, Computers and Structures, inc. Structural and earthquake
engineering software, Berkeley, California, USA.
Valluzzi, M. R., Binda, L., & Modena, C. (2002, June). Experimental and Analytical Studies
for the Choice of Repair Techniques Applied to Historic Buildings. Materials and
Structures, 35 , pp. 285-292.
Valluzzi, M., Valdemarca, M., & Modena, C. (2001). Experimental Analysis and Modeling
of Brick Masonry Vaults Strengthened by FRP laminates. Compos. for constr. , 5 (3), pp.
163-169.
Warke, P. A., McKinley, J., & Smith, B. J. (2006). Weathering of Building Stone:
Approaches to Assessment, Prediction and Modelling. In S. K. Kourkoulis (Ed.), Fracture
and Failure of Natural Building Stones (pp. 313-327). Netherlands: Springer.
Wenzel, F., & Maus, H. (1992, September). Repair of Masonry Structures. Meccanica , 27
(3), pp. 223-232.
Zienkiewicz, O. C., & Taylor, R. L. (1967). The Finite Element Method, Basic Formulation
and Linear Problems.
77
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 8
Appendix A
Letter of Venice
Preamble
Imbued with a message from the past, the historic monuments of generations of people
remain to the present day as living witnesses of their age-old traditions. People are becoming
more and more conscious of the unity of human values and regard ancient monuments as a
common heritage. The common responsibility to safeguard them for future generations is
recognized. It is our duty to hand them on in the full richness of their authenticity.
It is essential that the principles guiding the preservation and restoration of ancient buildings
should be agreed and be laid down on an international basis, with each country being
responsible for applying the plan within the framework of its own culture and traditions.
By defining these basic principles for the first time, the Athens Charter of 1931 contributed
towards the development of an extensive international movement which has assumed
concrete form in national documents, in the work of ICOM and UNESCO and in the
establishment by the latter of the International Centre for the Study of the Preservation and
the Restoration of Cultural Property. Increasing awareness and critical study have been
brought to bear on problems which have continually become more complex and varied; now
the time has come to examine the Charter afresh in order to make a thorough study of the
principles involved and to enlarge its scope in a new document.
Definitions
ARTICLE 1. The concept of an historic monument embraces not only the single
architectural work but also the urban or rural setting in which is found the evidence of a
particular civilization, a significant development or an historic event. This applies not only to
great works of art but also to more modest works of the past which have acquired cultural
significance with the passing of time.
ARTICLE 2. The conservation and restoration of monuments must have recourse to all
the sciences and techniques which can contribute to the study and safeguarding of the
architectural heritage.
78
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 7 9
Aim
Conservation
ARTICLE 7. A monument is inseparable from the history to which it bears witness and
from the setting in which it occurs. The moving of all or part of a monument cannot be
allowed except where the safeguarding of that monument demands it or where it is justified
by national or international interest of paramount importance.
Restoration
79
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 0
construction, the efficacy of which has been shown by scientific data and proved by
experience.
ARTICLE 11. The valid contributions of all periods to the building of a monument must
be respected, since unity of style is not the aim of a restoration. When a building includes
the superimposed work of different periods, the revealing of the underlying state can only be
justified in exceptional circumstances and when what is removed is of little interest and the
material which is brought to light is of great historical, archaeological or aesthetic value, and
its state of preservation good enough to justify the action. Evaluation of the importance of
the elements involved and the decision as to what may be destroyed cannot rest solely on the
individual in charge of the work.
ARTICLE 12. Replacements of missing parts must integrate harmoniously with the
whole, but at the same time must be distinguishable from the original so that restoration
does not falsify the artistic or historic evidence.
ARTICLE 13. Additions cannot be allowed except in so far as they do not detract from
the interesting parts of the building, its traditional setting, the balance of its composition
and its relation with its surroundings.
Historic Sites
ARTICLE 14. The sites of monuments must be the object of special care in order to
safeguard their integrity and ensure that they are cleared and presented in a seemly manner.
The work of conservation and restoration carried out in such places should be inspired by the
principles set forth in the foregoing articles.
Excavations
ARTICLE 15. Excavations should be carried out in accordance with scientific standards
and the recommendation defining international principles to be applied in the case of
archaeological excavation adopted by UNESCO in 1956. Ruins must be maintained and
measures necessary for the permanent conservation and protection of architectural features
and of objects discovered must be taken. Furthermore, every means must be taken to
facilitate the understanding of the monument and to reveal it without ever distorting its
meaning. All reconstruction work should however be ruled out "a priori." Only anastylosis,
that is to say, the reassembling of existing but dismembered parts can be permitted. The
material used for integration should always be recognizable and its use should be the least
that will ensure the conservation of a monument and the reinstatement of its form.
Publication
80
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 1
81
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 2
ICOMOS recommendations
Structures of architectural heritage, by their very nature and history (material and
assembly), present a number of challenges in conservation, diagnosis, analysis, monitoring
and strengthening that limit the application of modern legal codes and building standards.
Recommendations are desirable and necessary to ensure rational methods of analysis and
repair methods appropriate to the cultural context.
Principles
Therapy should address root causes rather than symptoms. Each intervention should
be in proportion to the safety objectives, keeping intervention to the minimum necessary to
guarantee safety and durability and with the least damage to heritage values. The choice
between “traditional” and “innovative” techniques should be determined on a case-by-case
basis with preference given to those that are least invasive and most compatible with heritage
values, consistent with the need for safety and durability. At times the difficulty of
evaluating both the safety levels and the possible benefits of interventions may suggest “an
observational method”, i.e., an incremental approach, beginning with a minimum level of
intervention, with the possible adoption of subsequent supplementary or corrective measures.
The characteristics of materials used in restoration work (in particular new materials)
and their compatibility with existing materials should be fully established. This must include
long-term effects, so that undesirable side effects are avoided.
Finally, a most relevant aspect is that the value and authenticity of architectural
heritage cannot be assessed by fixed criteria because of the diversity of cultural backgrounds
and acceptable practices.
82
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 3
Guidelines
The methodology stresses the importance of an “Explanatory Report”, where all the
acquired information, the diagnosis, including the safety evaluation, and any decision to
intervene should be fully detailed. This is essential for future analysis of continuous
processes (such as decay processes or slow soil settlements), phenomena of cyclical nature
(such as variation in temperature or moisture content) and even phenomena that can
suddenly occur (such as earthquakes or hurricanes), and for future evaluation and
understanding of the remedial measures adopted in the present.
83
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 4
Appendix B
a B C D E F G H
H G F E D C B a
84
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 5
85
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 6
Appendix C
86
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 7
87
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 8
88
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 8 9
Appendix D
Figure D- 1 Displacement in X-direction of joint 588 (orange) and 579 (blue) VS time.
89
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 9 0
Figure D- 2 Displacement in X-direction of joint 591 (green) and 578 (blue) VS time.
90
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 9 1
Figure D- 3 Displacement in Y-direction of joint 588 (orange) and 579 (blue) VS time.
91
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 9 2
Figure D- 4 Displacement in Y-direction of joint 591 (green) and 578 (blue) VS time.
92
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 9 3
Figure D- 5 Response spectrum curve for joint 1365 matching with the third mode of the structure at 0.23sec
Period vs. Pseudo spectral acceleration in Y-direction.
93
S e i s m i c R e s p o n s e a n d R e h a bi l i t a t i o n o f H i s t o r i c M a s o n r y B u i l d i n g s | 9 4
Figure D- 6 Response spectrum curve for joint 591 matching with the first mode of the structure at 0.3sec
Period vs. Pseudo spectral acceleration in Y-direction.
94