Content uploaded by Hugón Juárez García
Author content
All content in this area was uploaded by Hugón Juárez García on Oct 08, 2014
Content may be subject to copyright.
Evaluation of Confined Masonry Guidelines for
Earthquake-Resistant Housing
Prepared by
UBC EERI Student Chapter Committee on Confined Masonry Construction:
Tim Matthews (Chair)
Zahra Riahi
Jose Centeno
Arnaud Charlet
Hugon Juarez Garcia
Clint Hoffman
Sahar Safaie
Kenneth Elwood (Advisor)
ii
EXECUTIVE SUMMARY
The prevalent form of residential construction in many countries is unreinforced masonry.
Unfortunately, unreinforced masonry housing has been observed to perform poorly in earthquake
events. To improve the seismic performance of unreinforced masonry buildings, a structural
system called confined masonry has been developed. In confined masonry housing, unreinforced
masonry wall panels are surrounded by horizontal bond beams and vertical tie columns. The
bond beams and tie columns, which are typically made of reinforced concrete, work to confine
the masonry walls and keep them intact during earthquake shaking.
Confined masonry (CM) housing is already found in many developing countries and its
popularity is growing. In particular, CM has become a popular choice for reconstruction efforts
following major disasters such as the 2004 Great Sumatra Earthquake and Tsunami and the 2005
South Asian Earthquake. While some building codes do address CM construction (e.g. Mexico
City Building Code), codes are not always effective at controlling construction methods for
housing in developing countries. Simplified guidelines that can be implemented by homebuilders
or reconstruction agencies provide a more effective means of ensuring quality, seismic-resistant
housing in such regions. However, guidelines are frequently region-specific and comparisons
between guidelines often show discrepancies in the requirements stated.
The aim of this report is to identify a CM guideline or guidelines that, modified by certain
improvements if necessary, can be used to construct safe and affordable confined masonry
housing in seismically-active regions around the world. The report defines what confined
masonry is, analyzes how confined masonry structures behave during seismic shaking, and gives
a summary of current confined masonry construction practice, including its strengths and
weaknesses. Five confined masonry guidelines are described and compared, noting how the
guidelines address some of the weaknesses found in current CM practice. Confined masonry
research is consulted to help clear up some of the discrepancies between the guidelines, address
certain shortcomings in the guidelines, and identify areas requiring further research.
The five CM guidelines examined in the report are evaluated to determine which have the
potential to be applied generally in seismic regions around the world. Only one of the
guidelines, a document developed in Peru, is determined to have this potential.
Recommendations are proposed on how the seismic performance and applicability of the Peru
Guideline can be improved. Future research programs that could help optimize the Peru
Guideline are described.
iii
ACKNOWLEDGEMENTS
This report would not have been possible without the valuable input of Svetlana Brzev and the
EERI Student Chapter at UNAM in Mexico City. Their comments and suggestions at various
stages in the development of this report immeasurably improved its quality and perspective.
Thanks very much.
Thanks also to the external reviewers of the report: Qaisar Ali, Elizabeth Hausler and Durgesh
Rai. Their comments provided valuable insight into how confined masonry is actually practiced
in different parts of the world and helped bring to light passages in the report requiring further
clarity. The time taken by the reviewers to evaluate the report is greatly appreciated.
Finally, the encouragement and financial support of the EERI were instrumental in this report
reaching its current form. The report began as a small-scale, volunteer effort; without the
EERI’s generosity and vision, the document would never have obtained its current depth and
scope.
iv
TABLE OF CONTENTS
EXECUTIVE SUMMARY ...............................................................................ii
ACKNOWLEDGEMENTS ............................................................................ iii
TABLE OF CONTENTS .................................................................................iv
LIST OF FIGURES ..........................................................................................vi
LIST OF TABLES .......................................................................................... vii
1 INTRODUCTION...................................................................................1
2 CONFINED MASONRY FAILURE MODES.....................................6
2.1 Confined Masonry Buildings – How Do They Work? .............................6
2.2 Structural Failure Modes...........................................................................9
2.2.1 In-Plane Failure Modes................................................................................. 10
2.2.2 Out-of-Plane Failure Modes ......................................................................... 17
2.2.3 Diaphragm Failures....................................................................................... 23
2.2.4 Connection Failures ...................................................................................... 24
2.2.5 Closure .......................................................................................................... 26
2.3 Non-Structural Failure Modes.................................................................26
2.3.1 Masonry Partition Walls ............................................................................... 26
2.3.2 Gable End Walls ........................................................................................... 27
2.3.3 Parapets......................................................................................................... 28
2.3.4 Chimneys ...................................................................................................... 29
2.3.5 Roof Tiles...................................................................................................... 29
2.4 Summary Tables......................................................................................29
3 CURRENT PRACTICE .......................................................................32
3.1 History.....................................................................................................32
3.2 Summary of Current Practice..................................................................33
3.3 Shortcomings of Current CM Construction............................................34
3.3.1 Seismic Design Flaws................................................................................... 34
3.3.2 Construction and Post-Construction Concerns ............................................. 37
3.3.3 Lack of Regulation........................................................................................ 38
3.4 Positive Aspects of Current Practice.......................................................39
3.5 Closure.....................................................................................................39
v
4 GUIDELINE COMPARISON .............................................................41
4.1 Introduction to CM Guidelines ...............................................................41
4.1.1 City University Guideline............................................................................. 41
4.1.2 IAEE Guideline............................................................................................. 42
4.1.3 Peru Guideline .............................................................................................. 43
4.1.4 Schacher Guideline ....................................................................................... 43
4.1.5 UNESCO Guideline...................................................................................... 44
4.1.6 Guideline Summary Table ............................................................................ 45
4.2 Discussion of CM Guidelines .................................................................46
4.2.1 General Comparison ..................................................................................... 46
4.2.2 Comparison of Individual Items ................................................................... 49
4.3 Closure.....................................................................................................55
5 RESEARCH...........................................................................................56
5.1 Wall Openings.........................................................................................56
5.2 Tie Columns and Bond Beams................................................................58
5.3 Masonry Wall-Tie Column Connections................................................60
5.4 Minimum Wall Density...........................................................................62
5.5 Closure.....................................................................................................63
6 EVALUATION OF GUIDELINES & RECOMMENDATIONS ....65
6.1 Potential of CM Guidelines for General Applicability...........................65
6.2 Peru Guideline - Suggested Improvements.............................................69
6.2.1 Suggested Changes for Improved Performance............................................ 70
6.2.2 Suggested Changes for Wider Applicability................................................. 75
6.3 Optimizing the Peru Guideline - Future Research..................................77
6.4 Closure.....................................................................................................81
7 CONCLUSION......................................................................................82
8 REFERENCES......................................................................................84
8.1 References Cited .....................................................................................84
8.2 References Consulted..............................................................................87
vi
LIST OF FIGURES
Figure 1.1 – Components of a typical confined masonry wall ........................................... 1
Figure 1.2 – Comparison of masonry infill and confined masonry .................................... 2
Figure 2.1 – Idealization of earthquake demands on a structure ........................................ 7
Figure 2.2 – Load transfer path for inertial loads on a CM structure ................................. 8
Figure 2.3 – Division of earthquake excitation into x-axis and y-axis components........... 9
Figure 2.4 – Force transfer mechanism from which diagonal shear failure occurs .......... 10
Figure 2.5 – Tensile stresses and resulting cracking in the compression strut ................. 11
Figure 2.6 – Sliding shear failure in a CM wall................................................................ 14
Figure 2.7 – In-plane bending failure in a CM wall ......................................................... 15
Figure 2.8 – Vertical out-of-plane behaviour of a CM wall, horizontal cracking ............ 18
Figure 2.9 – Vertical out-of-plane behaviour of a CM wall, rocking ............................... 19
Figure 2.10 – Effect of vertical force during wall rocking ............................................... 19
Figure 2.11 – Horizontal out-of-plane behaviour of a CM wall: elevation view ............. 21
Figure 2.12 – Horizontal out-of-plane behaviour of a CM wall: plan section view......... 22
Figure 2.13 – Wood plank floor without and with a horizontal truss ............................... 23
Figure 2.14 – Gable end wall in a CM building ............................................................... 28
Figure 3.1 – Confining elements added by local construction workers............................ 32
Figure 3.2 – Failure at an opening due to no tie column .................................................. 35
Figure 3.3 – Failure due to tie column missing in corner ................................................. 35
Figure 3.4 – Shear failure of a tie column ........................................................................ 35
Figure 3.5 – Inadequate rebar detailing: stirrups not hooked properly............................. 35
Figure 3.6 – Inadequate roof-to-wall connection.............................................................. 36
Figure 3.7 – Gable end wall failure due to lack of support and confinement................... 36
Figure 3.8 – Poor mortar joint workmanship.................................................................... 37
Figure 3.9 – Unsoaked masonry units absorbing water from mortar ............................... 37
Figure 4.1 – Spectrum of CM guidelines.......................................................................... 47
Figure 6.1 – Possible top of partition wall lateral support connection ............................. 70
vii
LIST OF TABLES
Table 2.1 – Failure mode summary table, part 1 .............................................................. 30
Table 2.2 – Failure mode summary table, part 2 .............................................................. 31
Table 4.1 – Guideline summary table, part 1.................................................................... 45
Table 4.2 – Guideline summary table, part 2.................................................................... 46
Table 6.1 – Inspection checklist table, part 1.................................................................... 73
Table 6.2 – Inspection checklist table, part 2.................................................................... 74
Table 6.3 – Inspection checklist table, part 3.................................................................... 75
1
1 INTRODUCTION
Since the dawn of human civilization, masonry has been used to construct all types of buildings,
bridges, roadways and other civil engineering works. Even today, more masonry is used in the
construction of buildings than any other building material. There are many reasons for the
continued popularity of masonry. It is durable, easy to construct, fire-resistant, aesthetically
pleasing, and above all, economical. Thus, it is not surprising that masonry is the material of
choice for residential construction in many parts of the world.
Like all building materials, though, masonry has its weaknesses. In particular, unreinforced
masonry (URM) buildings have proven to be quite vulnerable in seismic events, with significant
building damage and numbers of fatalities occurring in URM buildings during past earthquakes.
To enhance the seismic performance of URM systems, different methods for reinforcing
masonry walls have been attempted over the years. Of these methods, two have proven to be
particularly effective: reinforced masonry (RM), which involves the placement of reinforcing
steel and concrete in the hollow cells of special masonry blocks, and confined masonry (CM).
Confined masonry is a structural system consisting of unreinforced masonry wall panels
surrounded by horizontal and vertical “confining” members called bond beams and tie columns.
Figure 1.1 illustrates a typical confined masonry wall. The masonry wall panel consists of
masonry units bonded with mortar; the masonry units themselves are usually concrete blocks or
clay bricks. The elements confining the masonry panel are nearly always constructed of
reinforced concrete (RC), although timber and other materials have been used occasionally. In
some cases, the units in a masonry wall are staggered or “toothed” at tie column locations (as
shown in Figure 1.1) to create better interlock between the wall and tie column.
Toothing of
Wall and Tie
Column
Unreinforced
Masonry Wall Tie
Column
Bond Beam
Figure 1.1 – Components of a typical confined masonry wall
In a confined masonry system, the masonry wall panels are relied upon to transmit all lateral
loads, including earthquake loads, and gravity loads to the building foundation. The bond beams
2
and tie columns work to hold the wall together under earthquake shaking and distribute out
lateral load within and between wall panels. The RC confining elements also help improve wall-
to-wall, floor-to-wall and roof-to-wall connections, enabling the structure to better act together as
a unit during a seismic event. The net effect of the presence of confining members is that
confined masonry wall systems have somewhat higher strength and considerably higher
deformation capacity than unreinforced masonry wall systems. This translates into better
seismic performance.
Confined masonry is sometimes confused with another type of building system: reinforced
concrete frames with masonry infill walls. These two systems often look the same (concrete
beams and columns enclosing masonry walls) but behave very differently. In confined masonry
systems, the masonry wall carries the structural forces while the RC members just confine the
masonry wall. As well, the RC members are usually the same thickness as the masonry wall in
CM construction, and the sequence of construction is such that the masonry walls are built first
and the tie columns and bond beams cast around the masonry wall.
With reinforced concrete frames and masonry infill walls, the RC members carry all the
structural loads and the masonry walls are non-structural elements. The columns and beams in a
RC frame are normally thicker than the masonry infill wall, and the RC members are constructed
first and the masonry wall is then built inside the finished concrete frame. Figure 1.2 below
illustrates the difference in construction sequence between confined masonry and RC frames
with masonry infill walls.
Figure 1.2 – Comparison of masonry infill and confined masonry
Source: AIS, 2001
Confined masonry housing is quite popular in many countries with significant seismic hazard.
CM construction can be found in Central and South America, Central and South Asia, Eastern
Europe and Indonesia. In many of these regions, CM housing is rapidly increasing in popularity
for a number of reasons. First, the methods and techniques used in CM construction are very
similar to those used in building URM structures, making it easier for builders to switch from
conventional masonry to CM. Second, it is only marginally more expensive to construct a CM
house than a conventional URM house. Lastly and most importantly, CM buildings exhibit
Masonry
I
nfill
Confined
M
asonry
3
better seismic performance than conventional masonry buildings. In fact, well-constructed CM
homes have been observed to incur little to no damage in moderate and even severe earthquakes.
The confined masonry system has generally evolved empirically. Over the years, builders have
tried to improve the seismic performance of first URM buildings and then CM buildings through
trial and error. In some countries, such as Mexico, engineers have become involved and
developed codes for CM construction. Despite the development of codes, the vast majority of
CM construction is built in a non-engineered fashion according to historical local construction
practice.
There are a number of reasons why much of the CM housing around the world is not built
according to the available codes. Many of these reasons stem from the lack of resources in many
regions where CM is popular. In many cases, CM codes are not enforced by an organized
governmental or professional body because of insufficient resources and political will. On many
housing projects, there is also not enough money available to pay an engineer to develop a code-
compliant design. Lastly, the provisions in many CM codes tend to be quite conservative; to
adhere to these provisions would in some cases make construction of a CM house prohibitively
expensive.
To encourage proper detailing and proportioning of members in non-engineered confined
masonry houses, guidelines have been developed. Confined masonry guidelines are usually
based on codes, but simplify the design of simple CM buildings by reducing complex code
design equations to simple rules and recommendations. Since technical calculations are not
required, CM guidelines can be implemented with less effort and by individuals without
extensive engineering backgrounds, for instance, homebuilders. These two factors make CM
guidelines more appealing to use than codes in many instances.
On the downside, CM guidelines are usually intended for a specific region and it is sometimes
difficult to discern what level of seismic resistance a guideline provides. This can make it
difficult to apply guidelines outside the specific region for which they were developed. This lack
of applicability is a limitation of some guidelines, especially in comparison to codes, which are
usually setup to incorporate varying levels of seismic hazard and can be applied to a wide variety
of regions. Thus, a drawback of some guidelines is that the simplifications they make to the
codes they are based on reduce their general applicability.
Confined masonry housing has demonstrated good resistance to earthquakes when it’s
constructed properly. Unfortunately, much CM construction is built according to historical
practice and may have significant seismic resistance deficiencies. Codes are one answer to the
problem of insufficient seismic resistance in CM structures, but a lack of resources and
engineering expertise make codes unviable in certain situations. Guidelines provide a more
promising alternative to improving the seismic resistance of a wide range of CM housing, in
particular, if difficulties with applicability discussed above can be overcome.
The goals of this report are to conduct a thorough investigation of the confined masonry housing
system (how it works, how it is currently constructed), evaluate the CM guidelines currently
available, and suggest how the CM guidelines can be improved and then used to construct
4
seismic resistant housing throughout the world. These goals can be organized into the following
five objectives:
1. Explain how confined masonry structures behave when subjected to seismic loading and
identify potential failure modes which must be addressed in the design of CM buildings.
2. Summarize current confined masonry practice, identifying shortcomings and positive aspects
of current CM construction.
3. Compare existing confined masonry guidelines, noting the strengths and shortcomings of
each; discuss how the guidelines address shortcomings in current CM practice.
4. Examine confined masonry research to resolve discrepancies between CM guidelines and
address shortcomings in the guidelines; identify future research areas focused on issues not
dealt with by current research.
5. Evaluate CM guidelines based on their potential for general applicability, suggest specific
improvements to generally applicable guidelines, and propose future research topics aimed at
optimizing such guidelines.
The scope of the report will be limited to simple confined masonry buildings, in other words,
buildings that are one to two stories high, rectangular in plan and symmetric. There are two
reasons why the report scope was limited in this way. The first is that such simple buildings
have uncomplicated, predictable behaviour and as a result, complex technical design calculations
can be reduced to simple rules. The second reason is that a large portion of the confined
masonry building stock fits into the category chosen; limiting the scope in this way does not
exclude a large percentage of CM buildings.
Earlier in this section, reinforced masonry was introduced as another masonry system with good
seismic resistance. RM requires special hollow masonry units that are sometimes larger than
common solid clay bricks and are not available in all regions. As well, there have been problems
with quality control in small, residential RM buildings because the reinforcing steel and concrete
are hidden from view. For these two reasons, RM is generally used for larger commercial,
industrial and institutional buildings that are outside the scope of this report. Thus, the
reinforced masonry system will not be discussed further.
This report has been written with a wide audience in mind: non-governmental organizations
(NGOs) and other agencies looking to implement confined masonry in reconstruction efforts;
local and global building authorities; engineers; construction workers; researchers; and other
individuals interested in confined masonry structures. People with a limited engineering
background, may find parts of this report a bit technical and difficult to interpret. To alleviate
this problem, tables, pictures and figures have been included wherever possible to help clarify
technical concepts.
It is hoped that this report will help inform individuals and organizations currently constructing
unreinforced masonry buildings in seismic zones of the many merits of confined masonry
housing. The report also intends to provide such individuals and organizations with assistance
on how confined masonry guidelines can be applied in their particular situation to construct
houses with good earthquake resistance.
5
Finally, a cautionary note. The instructions and recommendations in this report are not meant to
be implemented blindly. Rather, this document is intended to be an aid in helping qualified
individuals apply confined masonry guidelines to local materials, conditions and regulations.
Further, this report is not in itself a guideline, but is instead a review and critique of the various
guidelines available. In itself, the report does not contain all the information required to
construct a safe confined masonry home. For complete set of instructions on the steps involved
in building a CM house, the guidelines themselves must also be consulted.
6
2 CONFINED MASONRY FAILURE MODES
Failure modes are at the very core of the structural design of buildings. A failure mode is exactly
as its name suggests: a mechanism (or mode) by which a component in a building ceases to be
able to resist the loads applied on it (i.e. it fails in its ability to act structurally). The essence of
structural design is identifying the possible mechanisms that could bring about failure, predicting
the force or deformation that failure by this mechanism will occur, and determining if this force
or deformation capacity is larger than the expected demands on the structure.
For confined masonry buildings, and in fact all buildings, there are two general types of failure
modes: structural failure modes and non-structural failure modes. Structural failure modes
involve the failure of structural members, or members that are relied upon to support other
components of a building. If a structural failure mode occurs, the result is collapse of part or all
of the building. For instance, if a structural wall collapses, in most cases the floors and roof
supported by the wall will also collapse.
In contrast, non-structural failure modes result only in collapse of the failed component itself.
This is because non-structural members do not support other members. From a structural
perspective, such members are only required to resist the forces applied to them directly, such as
self-weight or inertial forces (inertial forces will be explained shortly). An example of a non-
structural failure is the collapse of an interior partition wall under earthquake loads; such walls
do not support the roof or floor above them (they are just dividers) and therefore if they collapse,
the structural load-carrying capacity of the building is not affected.
The fact that non-structural failure modes do not lead to complete collapse of a building should
not imply that they are not of concern. The falling bricks resulting from collapse of a partition
wall or parapet are a significant life safety hazard. Thus, to reduce the overall seismic risk of
confined masonry buildings, non-structural failure modes must be considered.
As may be evident from the above introduction to CM failure modes, this section is quite
technical. Readers without engineering backgrounds may find this section challenging to fully
understand, since some of the concepts covered draw on relatively advanced topics in structural
and earthquake engineering. Thorough explanations and figures have been provided wherever
possible to help effectively describe the different aspects of CM failure modes. Nonetheless, the
section remains rather technical in places and therefore may not be useful to all readers.
2.1 Confined Masonry Buildings – How Do They Work?
Before getting into the different failure modes that are of concern for confined masonry
buildings, it would probably be helpful to explore briefly how these buildings work to resist
earthquake forces. What follows is a simplified explanation of what earthquake forces are and
how a well-constructed confined masonry home behaves in a seismic event.
Earthquakes introduce stress into a building by accelerating and displacing the foundation of the
building. Since no lateral forces are applied to the building above the ground, the walls, floors
7
and roof want to stay in their original positions. The foundation, however, wants to drag the rest
of the building along with it since everything is connected together. To do so, the structural
elements (the components connecting the building together) must apply forces on the
superstructure (all of the building above ground) to get it to move with the foundation.
The forces applied by the structural elements work to overcome the inertia, or resistance to
change in velocity (the initial velocity in this case being zero), of the superstructure, and hence
are referred to as “inertial forces”. The inertia of a component of a building is directly
proportional to its mass. Thus, the heavier a wall, floor, etc. is, the larger the force that must be
developed in the structural elements to make it move with the foundation.
Trying to deal with earthquake loads as ground displacements and accelerations can sometimes
be difficult to visualize and analyze. What is often times done instead is to consider the base of
the structure to remain still, and apply equivalent lateral loads on the building equal to the
inertial forces caused by the ground motion. The two systems are structurally equivalent, but the
latter is usually easier to understand and interpret. Figure 2.1 illustrates the two ways of
conceptualizing earthquake loads.
Inertial Loads
due to Mass of
Roof + Floor
Inertial Loads
due to Mass
of Walls
Deflected Building Shape due to Ground Motion = Deflected Building Shape due to Inertial Loads
Ground Acceleration
Figure 2.1 – Idealization of earthquake demands on a structure
It should be noted that the magnitude and direction of seismic loads continually change during
the course of an earthquake. This is often accounted for in seismic design by determining the
maximum demands in each direction and designing the structure to resist these demands.
The equivalent lateral force visualization of earthquake loads will now be used to explain how a
confined masonry structure transmits earthquake forces to the foundation. Referring to Figure
2.2 below, the inertial forces on a confined masonry building are concentrated where most of the
mass is: at the floor and roof levels, and in the masonry walls themselves. The red arrows show
the inertial forces on each component at the component’s center of mass, the blue arrows
illustrate how the inertial forces transfer to supporting elements, and the green arrows represent
forces on components transferred from other components. How a confined masonry structure
works to transfer inertial forces is described beneath Figure 2.2.
8
Direction
of Ground
Motion
Figure 2.2 – Load transfer path for inertial loads on a CM structure
The masonry walls perpendicular to the direction of shaking are much weaker and more flexible
than the walls parallel to the direction of shaking, roof and floors. As a result, these walls
transfer their inertial forces to the much stiffer elements bordering the walls. This force transfer
is accomplished by out-of-plane bending, which is bending that results in the wall deforming
perpendicular to the plane of the wall.
The roof and floor of confined masonry buildings transmit the inertial forces from their own
mass, along with the forces transferred to them from the walls perpendicular to the direction of
shaking, to the walls parallel to the direction of shaking. To do so, roofs and floors deform in-
plane (since all deformations are parallel to the plane of the roof or floor) as a diaphragm, and as
a result, are referred to as diaphragm elements.
The walls parallel to the direction of shaking have the responsibility of transferring inertial forces
from their own mass, the diaphragms and the walls perpendicular to shaking down to the
foundation, where it is transmitted to the surrounding soil. These walls perform this function by
deforming in-plane in shear and bending.
In explaining how confined masonry buildings work, an earthquake acting parallel to one of the
principal axes of the structure was assumed for simplicity. In reality, the direction of ground
shaking is often at an angle to the walls in a structure. For this case, the earthquake excitation
can be divided into components in line with the principal building axes, as shown in Figure 2.3.
The structure must be able to resist the inertial forces induced by both excitation components
simultaneously. In other words, walls have to be able to act in-plane and out-of-plane at the
same time.
9
Direction of
Earthquake
Excitation
Y-Axis
X-Axis X-Axis
Component of
Excitation Y-Axis
Component of
Excitation
Figure 2.3 – Division of earthquake excitation into x-axis and y-axis components
It should be noted that the force transfer paths discussed above can not occur if the elements or
connections between elements do not have sufficient capacity to resist the inertial forces on the
structure. In poorly constructed or designed confined masonry structures, certain connections or
members fail prematurely, hampering the performance of the building during earthquake events.
Next, we will look at the different ways that parts of confined masonry structures can fail. In
other words, we’ll examine the different confined masonry failure modes.
2.2 Structural Failure Modes
The structural failure modes observed in confined masonry structures can be divided into five
broad categories: in-plane failures, out-of-plane failures, diaphragm failures, connection failures
and foundation failures. Although definitely a concern, the foundation failures observed in
confined masonry buildings are not unique to this structural system and are in fact the same as
those seen in other types of buildings. Consequently, there is much information on foundation
failures elsewhere and they will not be discussed further in this report.
The other four types of confined masonry structural failure modes will be discussed as if they
occur independently. The goal of this approach is to try to explain these somewhat complex
concepts clearly. In reality, damage resulting from one failure mode can reduce the capacity of
the structure in another failure mode, and two failure modes can act simultaneously. The
interaction between different failure modes is quite complicated, however, and not well
understood; it will therefore not be explored at length.
Earlier in this section it was mentioned that the direction of the inertial forces acting on a
building will change during the course of an earthquake. For simplicity, however, possible CM
failure modes will be explained in this report considering loads in only one direction. When
reading the discussion and figures that follow, one should keep in mind that mirror images of the
10
force transfer and damage described will result when inertial forces reverse and start acting in the
direction opposite to that shown.
2.2.1 In-Plane Failure Modes
In-plane failures occur in walls parallel to the direction of earthquake shaking. Confined
masonry walls are quite strong and stiff in the in-plane direction; therefore, resisting forces in the
in-plane direction is an efficient way for a structure to transfer lateral earthquake loads to the
foundation. As a result, not only inertial forces from the mass of the in-plane walls themselves,
but also inertial forces from the floors, roof and walls perpendicular to the earthquake direction
are transferred through in-plane behaviour.
In-plane failure modes tend to dominate in structures with short, stocky walls, where the ratio of
the distance between lateral supports (roofs and floors in the vertical direction and intersecting
walls in the horizontal direction) to the wall thickness is relatively small. For more slender
walls, out-of-plane failure modes are more of a concern.
Three failure modes can result from confined masonry walls carrying lateral loads in-plane:
diagonal shear failure, sliding shear failure and in-plane bending failure.
2.2.1.a Diagonal Shear Failure
Diagonal shear failure is the type of in-plane mechanism observed in well-proportioned confined
masonry structures. In this type of failure, cracks form along the main diagonal of the wall panel
and eventually the wall moves along this joint. Several approaches can be used to idealize the
force transfer that brings about diagonal shear failure. Figure 2.4 illustrates one approach in
which the force transfer is achieved through a compression strut and a tension tie. The vertical
forces acting on the wall from gravity loads it supports are omitted from Figure 2.4 for clarity.
Concentrated
Intertial
Force
Tension
Tie Force
Idealized
Compression
Strut
Compression
Strut Force
Figure 2.4 – Force transfer mechanism from which diagonal shear failure occurs
11
Referring to Figure 2.4, the concentration of horizontal inertial forces from upper floors and the
diaphragm at that level is transferred from the top of the wall to its base by dividing the
horizontal force up into a diagonal compression strut in the masonry and a vertical tension tie in
the tie column. This arrangement works well because masonry carries compression well while
the reinforcement in the tie column is good at resisting tension.
Strictly speaking, most confined masonry design guidelines stipulate that tie columns are not to
be relied upon to resist any loads. To comply with this stipulation, the tension tie can be
visualized as being in the masonry next to the tie column. Although, masonry is weak in tension,
the vertical load on the wall normally more than compensates for the tensile stresses in the
tension tie, enabling the masonry to be visualized as a tension tie.
A property of materials under stress is that perpendicular to the direction of compressive force,
tensile stresses form. Figure 2.5 shows this concept. The tensile stresses that form at right
angles to the compression force cause, with increasing stress, zig-zag cracks to form in mortar of
the masonry wall. These cracks form in the mortar because this material is very weak in tension.
When masonry units are very weak, as is sometimes the case for hand-made units, it is also
possible to get diagonal cracks in the units themselves. For good quality materials, though, the
mortar is normally the weak link and is the first place where damage forms.
Tension Stresses
Resulting from
Compression Force
Compressive Force
in Strut
Cracks in Mortar
from Tension Stress
Section of Compression Strut
Figure 2.5 – Tensile stresses and resulting cracking in the compression strut
As diagonal cracking increases, the compression strut becomes less effective in transferring
compressive force. To compensate, the tie-columns work to distribute the horizontal inertial
force off the main diagonal, in effect widening the compression strut. As inertial forces continue
to deform the wall further, cracking will increase, forcing the tie-column to distribute more and
more force.
If the horizontal force at the floor level continues to increase, eventually the top of the tie column
will crack or masonry units in the compression strut will start to crush. When either condition
occurs, the other closely follows. After this point, the wall typically begins to lose strength, but
can still deform and transmit lower magnitudes of force. This capacity to deform beyond the
peak strength of the wall, also known as ductility, is useful for seismic loading, since maximum
inertial forces only occur for a short time before diminishing and often reversing direction.
12
Thus, even if the inertial loads on a wall exceed its strength capacity, if the wall is ductile, it will
stay together and “stretch” until the inertial loads lessen. In this way, the wall is prevented from
collapse.
During the “stretching” phase beyond maximum strength resistance, the wall continues to
deform. These additional deformations cause further damage in the masonry and tie columns. In
many cases, ultimate failure occurs when the tie columns completely fail in shear (by developing
wide diagonal cracks) and the wall splits apart along the main diagonal.
Contributing Factors
There are many factors that contribute to a confined masonry wall’s ability to resist diagonal
shear failure. The shear and tension strength of mortar, along with brick-mortar bond strength,
are generally the weak links in masonry assemblies. For diagonal shear failure, they determine
at what point the tensile stresses perpendicular to the compression strut start to cause cracking.
Other factors that have an effect on wall strength and deformation capacity include masonry unit
compression strength, vertical load on the wall (more vertical load is beneficial if less than the
compressive strength of the units and mortar because it helps to increase mortar shear
resistance), concrete-masonry interface shear strength (has a similar effect to vertical load) and
tie column shear strength. This last parameter is a function of tie column cross-sectional area,
concrete strength, rebar strength, rebar anchorage and amount of longitudinal and transverse
reinforcing. It is the resultant of all these tie column characteristics that is relevant for the
behaviour of CM walls, so for brevity, just tie column shear strength will be referred to hereafter.
Two other factors that affect wall capacity are the number of tie columns and the structural wall
density in the direction of shaking. If there are too few tie columns, the system may behave
more as an unreinforced masonry wall than a confined masonry wall. Unreinforced masonry
walls typically have lower lateral strength capacities, much lower deformation capacities and
considerably lower ductility. Structural wall density is defined as the plan area of the structural
walls stretching in a particular direction. The wall density, which is the total length of structural
wall in one direction, influences in-plane wall behaviour simply in that the more wall there is in
the direction of shaking, the less inertial force each wall has to transmit to the foundation.
Typical Damage
As described slightly earlier, the initial damage characteristic of diagonal shear behaviour is
diagonal cracking in the mortar or masonry units along the compression strut. With increased
loading, this diagonal cracking widens and spreads. At larger deformations, one sees crushing of
masonry bricks, cracking in the tie columns and eventually tie column shear failure. The last
damage item is followed by wall collapse, and should be avoided.
Situations Where Failure Mode Dominant
Confined masonry buildings with height-to-length wall panel ratios of between 1:1 and 1:2 are
quite common. Panel height is defined as the dimension between bond beams and panel length
13
as the dimension between tie columns. This height-to-length ratio is also known as the panel
aspect ratio. In addition to being popular, confined masonry buildings with panel aspect ratios
between 1:1 and 1:2 are also considered to be “well proportioned” [Riahi, 2007]. For these “well
proportioned” structures, diagonal shear failure is the dominant in-plane failure mode.
Normally, diagonal shear failures occur in the first story where lateral loads are highest.
Ways to Improve Resistance to Failure Mode
Since diagonal shear failure is the predominant failure mode in properly constructed CM houses,
the goal when designing a confined masonry structure is often to maximize the capacity of a
confined masonry building against diagonal shear failure and then check to see if any of the other
failure modes have become dominant. This being the case, there are a number of ways in which
a confined masonry structure can be improved to increase resistance to diagonal shear failure.
The most obvious way to increase resistance to diagonal shear failure is to improve upon the
factors that contribute to this failure mode: mortar strength, masonry-mortar bond strength,
masonry unit compressive strength, tie column shear strength, etc. as listed previously.
Some research and field observations suggest that placing wire reinforcing in the horizontal
mortar joints at certain ratios can also be beneficial. Wire reinforcing has been noted to have an
effect similar to higher mortar strength, and also appears to help with widening the compression
strut, thereby spreading out diagonal cracking. Despite these benefits, the extra cost associated
with placing wire reinforcing in the horizontal mortar joints may not justify the benefits;
improving other factors may prove more cost-effective.
Finally, as is the case for all failure modes, good workmanship, in particular uniform and full-
thickness mortar joints with good bond, has a large impact on wall capacity. Constructing
confined masonry walls properly will help ensure adequate resistance to diagonal shear failure.
Effects of Openings
One other item that must be addressed when discussing diagonal shear failure is the effect of
wall openings. Wall openings weaken the in-plane resistance of walls by decreasing the size of
the wall cross-section at opening locations. Wall openings also cause stress concentrations at
their corners, which promote the formation of diagonal cracks. These cracks start at the opening
corners and progress in a zig-zag fashion through the mortar joints or diagonally through the
masonry units in the wall. This cracking can further reduce wall capacity.
Large, unconfined openings can cause confined masonry walls to behave similar to unconfined
masonry. Wall regions with large unconfined openings may not have the increased strength and
ductility characteristic of confined masonry, thereby making them weak points in the structure.
Conversely, if wall openings are confined by tie-columns on both sides, diagonal cracks from
stress concentrations are resisted by the stronger reinforced concrete, confined masonry
behaviour is preserved, and the only major effect on the load-carrying capacity of the building is
that the length of structural wall is reduced.
14
2.2.1.b Sliding Shear Failure
The second type of in-plane failure mode observed in confined masonry structures is sliding
shear failure. In this type of failure, a horizontal crack forms in the mortar joint across a portion
or the entire length of the panel and then extends into the tie columns. Once the tie columns
have failed in shear, failure occurs by the wall sliding along the horizontal joint. Figure 2.6
illustrates the sliding shear failure mode.
Concentrated
Intertial
Force
Shear Failure in
Mortar Joint
Shear Failure in
Tie Column
Figure 2.6 – Sliding shear failure in a CM wall
Sliding shear failure only occurs if the resistance of the horizontal mortar joint is very weak.
Even when the mortar fails, the shear resistance of the tie-columns must be overcome for overall
wall failure to occur. Thus, sliding shear failure is much less likely in confined masonry than in
unreinforced masonry.
Contributing Factors
The factors that contribute to sliding shear failure are essentially the same as for diagonal shear
failure: mortar shear strength, brick-mortar bond strength, tie column shear strength, number of
tie columns, and structural wall density in the direction of shaking.
Particularly important for sliding shear is the vertical load on the wall. Vertical load increases
the frictional resistance of the horizontal mortar joints. Unless the mortar shear strength is
exceptionally low and the vertical load on the wall is very small, a horizontal crack will not
spread across the entire wall panel. Instead, cracks will move down the wall in a zig-zag, as per
diagonal shear failure.
Typical Damage
As mentioned previously, the damage typifying sliding shear failure is horizontal cracking
through the mortar and tie-columns.
15
Situations Where Failure Mode Dominant
In unreinforced masonry systems, sliding shear has been observed in the upper floors of
buildings where the vertical load is smallest. In these instances, the mortar joints have had poor
shear resistance, either because of poor materials, poor workmanship or both.
For confined masonry buildings, though, even structures with light roofs have not been observed
to fail as a result of sliding shear. This is particularly the case for one and two story confined
masonry buildings. Thus, as long as good quality materials and workmanship are present and an
adequate number of tie-columns are included, sliding shear should not be a concern for this class
of building.
Ways to Improve Resistance to Failure Mode
As just noted, sliding shear is generally not a concern in one and two story confined masonry
buildings. Strong mortar shear strength, good tie column shear resistance and reasonably spaced
tie columns should ensure this failure mechanism is not a problem.
2.2.1.c In-Plane Bending Failure
The third type of in-plane failure mode is in-plane bending failure. For this type of failure, the
wall bends about an axis perpendicular to the wall, causing compression stresses at one end of
the wall and tensile stresses at the other. Assuming the tie column does not participate, the wall
fails when the tension strength of the mortar at the tension end is exceeded, causing the wall to
tip over. Figure 2.7 shows a schematic of what an in-plane bending failure looks like.
Cracking from mortar
tension stength being
exceeded
Concentrated
Intertial
Force
Vertical Loads on wall from floor/roof
slab + walls from higher levels
Tension sideCompression side
Figure 2.7 – In-plane bending failure in a CM wall
16
Depending on the strength of the bond between the masonry and concrete tie-columns, some
bending forces may be transferred from the wall through to the reinforced concrete tie columns.
Masonry has reasonably good compressive strength, so force transfer on the compression side is
usually not a problem. On the tension side, however, the tie-column is significantly better at
resisting tension than the masonry since the former is reinforced with steel. Because of this,
confined masonry walls can be significantly better at resisting in-plane bending failure than
unreinforced masonry walls.
Contributing Factors
Likely the most important factor determining if a wall is susceptible to in-plane bending is the
panel aspect ratio. Only if the aspect ratio is quite large (ie. the panel height is much larger than
the panel length), will the wall behave as a bending element rather than a shear element prone to
sliding or diagonal shear failure. The greater the aspect ratio, the more significant the tension
and compression stresses at the ends of the wall.
There are a number of factors that will influence a wall’s resistance to in-plane bending. On the
material side, the two most important are likely mortar tension strength and concrete-masonry
interface bond strength. These two parameters determine the tensile stresses from bending that
the wall can resist.
Another crucial factor on in-plane bending resistance is the vertical load on the wall. Vertical
load works to counteract the tensile stresses from bending, but compounds with the compression
stresses from bending. Since the compressive strength of masonry is generally much greater
than the tensile strength, the net effect of vertical load is generally beneficial. In extreme cases
the sum of the compression stresses from bending and vertical load will approach the
compressive strength of the masonry unit, making masonry crushing a possible failure mode.
Typical Damage
Two types of damage can result from in-plane bending behaviour. The most likely damage is at
the tension end of wall, where horizontal tension cracks in mortar, separation between the
masonry wall and tie-columns, and even horizontal cracking in the tie column can all occur. The
second type of damage from in-plane bending can only occur if the vertical force on the wall is
high. In such cases, crushing of the masonry bricks at the compression end of the wall is
possible.
Situations Where Failure Mode Dominant
In-plane bending failure of confined masonry walls only occurs when wall panels are very tall,
not very long, and loaded very lightly with vertical loads. Poorly constructed mortar joints can
contribute to these types of failures by reducing tension resistance.
In unreinforced masonry systems, in-plane bending failure sometimes occurs in the upper floors
of buildings where vertical load is smallest. However, in-plane bending failure is not very likely
in confined masonry buildings, especially those of one and two stories. For confined masonry
17
structures where the masonry wall is interlocked with the tie columns, in-plane bending failure is
even less likely; this setup facilitates significant force transfer across the wall-tie column
interface, allowing the tie column to carry much of the tensile stress from bending.
Ways to Improve Resistance to Failure Mode
As indicated above, in-plane bending failure is usually not an issue in confined masonry
buildings of two levels or less. This being said, confined masonry structures that interlock the
masonry wall and tie columns or contain rebar across the masonry-tie column interface are even
less likely to suffer from this failure mode. Wall-tie column interlock and interface rebar
facilitate significant force transfer across the wall-tie column interface, allowing the tie column
to carry much of the tensile stress from bending rather than the masonry wall with its
comparatively low tension strength.
2.2.2 Out-of-Plane Failure Modes
Out-of-plane failures occur in walls perpendicular to the direction of earthquake shaking.
Confined masonry walls are quite weak and flexible in the out-of-plane direction. As a result,
only inertial forces from the mass of the walls themselves are transferred by this behaviour.
Confined masonry walls span vertically and horizontally out-of-plane to floors, roofs and
intersecting walls, transferring their inertial loads to these stiffer elements. In some CM designs,
tie columns not laterally supported by intersecting walls and bond beams not laterally supported
by floors or roofs are relied upon to act as supports for masonry walls bending out of plane. In
such situations, the tie columns and bond beams must act as bending members, making their
bending strength and stiffness critical parameters. The fact that many confining elements have a
small cross-section and must span relatively long distances means that these members are quite
flexible, in some cases too flexible to act as lateral supports for the masonry wall. For this
reason, only bond beams at floor or roof levels and tie columns at wall intersections will be
considered lateral supports for masonry walls.
The ratio of how much of a wall’s inertial loads will be transferred out-of-plane vertically as
compared to horizontally is influenced by the relative length of the span in each direction. If the
vertical dimension between lateral supports (floors and roofs) is much less than the horizontal
dimension between lateral supports (intersecting walls), most of the wall’s out-of-plane inertial
loads will be transferred in the vertical direction.
Out-of-plane failure modes tend to dominate in structures with tall, thin walls, where the ratio of
the distance between lateral supports (roofs and floors in the vertical direction and intersecting
walls in the horizontal direction) to the wall thickness is quite large. As discussed above, in-
plane failure modes are the primary concern for more stout walls.
Vertical and horizontal out-of-plane failure modes for confined masonry walls are somewhat
different. For simplicity, each will now be discussed separately, as if the contribution from out-
of-plane behaviour in the other direction was nonexistent. The real out-of-plane behaviour of
18
confined masonry walls, however, is a combination of bending in both the vertical and horizontal
directions.
2.2.2.a Vertical Out-of-Plane Failure
Vertical out-of-plane behaviour will be the dominant out-of-plane failure mode when the
distance between horizontal lateral supports (floor and roof diaphragms) is less than the distance
between vertical lateral supports (intersecting walls). When a wall fails out-of-plane vertically,
there are 2 distinct stages.
In the first stage, the wall bends until the tensile strength of the mortar is exceeded, resulting in a
horizontal crack in the mortar bed joint. This is a consequence of mortar being much less strong
in tension than either the masonry units or mortar in compression. As mentioned previously, in
some instances the masonry unit is weaker than the mortar, in which case a horizontal crack
would occur in the unit rather than the mortar joint; this situation, however, occurs much less
frequently than mortar cracking. For a wall supported at both top and bottom, horizontal
cracking usually occurs at the supports and near mid-height, whereas for walls supported at the
bottom only, cracking occurs just at the base. Figure 2.8 illustrates this damage pattern.
Tension
Capacity of
Mortar
Exceeded
Concrete Bond
Beam at
Floor/Roof
Level
Concrete Bond
Beam at Floor
Level
Inertial
Forces
from mass
of masonry
Tension
Capacity of
Mortar
Exceeded
=> Horizontal
Crack Forms
Inertial Forces
from mass of
masonry
Figure 2.8 – Vertical out-of-plane behaviour of a CM wall, first stage: horizontal cracking
In the second stage of failure, the wall continues to stay intact after formation of a horizontal
crack by rocking. During rocking, the wall rotates as a rigid body, with the weight of the wall
acting as a restoring force (refer to Figure 2.9). If, however, the inertial forces acting on the wall
cause it to displace too far, the weight of the wall changes from a restoring force to a
destabilizing force, resulting in collapse.
19
Inertial
Forces from
Mass of
Masonry
Gravity Forces
work to tip wall
over
Imminent Collapse
Inertial
Forces from
Mass of
Masonry Wall Gravity
forces work to
restore wall to
vertical
Stable Rocking
Figure 2.9 – Vertical out-of-plane behaviour of a CM wall, second stage: rocking
It should be noted that vertical forces transferred to the top of confined masonry walls by floors
and roofs help increase vertical out-of-plane capacity. Intuitively, adding more compressive
force will help to counteract the tensile forces resulting from bending prior to horizontal
cracking. Referring to Figure 2.10, vertical load is also beneficial after cracking, acting as an
additional restoring force during rocking.
How wall
wants to rotate
All vertical
forces work
to restore
wall to
vertical
Wall Gravity
Forces
Additional vertical
force from floor/roof +
walls from higher levels
Vertical force from
stiff bond beam
pushing down on
wall as it rotates
How wall actually
rotates because
of force applied
by bond beam
Inertial
Forces from
Mass of
Masonry
Figure 2.10 – Effect of vertical force during wall rocking
Also outlined in Figure 2.10 is the beneficial effect that concrete bond beams at the roof level
have on vertical out-of-plane behaviour. Stiff roof bond beams push back on the top of the wall
as it attempts to rotate and lengthen while bending. This effect is equivalent to increasing the
vertical load on the wall.
20
Contributing Factors
There are a number of factors that contribute to the resistance of a wall to vertical out-of-plane
bending. One of the most significant is the ratio of the wall height between horizontal lateral
supports to the wall thickness, or h/t ratio. For high h/t ratios, large mortar tensile stresses from
bending can result from even small lateral inertial forces. After horizontal cracking, the amount
of rotation a wall can undergo during rocking before becoming unstable decreases with
increasing h/t ratio.
Related to h/t ratio is the number of horizontal lateral supports. If a masonry wall is only
laterally supported at its base, the h/t ratio is effectively doubled. This underlies the importance
of proper connections of the masonry wall to bond beams, and bond beams to horizontal
diaphragms. Failure of the above connections eliminates the lateral support at afflicted locations,
possibly inducing out-of-plane failure in walls that would otherwise have adequate resistance to
this failure mode.
Other parameters that improve vertical out-of-plane resistance are increased vertical load on the
wall, stiffer roof bond beams and greater mortar tension strength. The benefits of vertical load
and bond beam stiffness are discussed above. Higher mortar tension strength will enable the
wall to carry larger inertial forces before the onset of horizontal cracking.
Typical Damage
When the mortar tension strength is exceeded, horizontal cracks form in bed joints at the top,
bottom and mid-height of walls supported top and bottom. For walls supported at the bottom
only, cracking results only at the base.
After a wall has cracked and begun rocking, when the wall rotates past the point of stability, the
wall collapses along with the floors and roof the wall supports vertically. This is a very
catastrophic type of failure.
Situations Where Failure Mode Dominant
Vertical out-of-plane failure is dominant in walls with high h/t ratios, even higher l/t ratios (l/t
ratio will be defined shortly), low vertical loads and flexible roof bond beams. Since vertical
load is a factor, this failure mode is more problematic in upper stories. Walls in upper stories are
also more susceptible because accelerations (and therefore inertial forces) are greater near the top
of CM buildings. Sometimes vertical out-of-plane failure follows failure of a horizontal lateral
support connection; the connection failure increases the h/t ratio, bringing about vertical out-of-
plane collapse.
Ways to Improve Resistance to Failure Mode
Resistance to vertical out-of-plane bending can be increased by improving upon the factors that
contribute to this failure mode. Structures with smaller distances between horizontal lateral
supports, thicker walls, greater vertical loads, and stiffer roof bond beams will be more resistant
21
to vertical out-of-plane failure. Buildings with adequate wall-to-bond-beam-to-diaphragm
connections will also have better resistance to out-of-plane failure. If these contributing factors
can not be improved upon, reducing the distance between vertical lateral supports can help by
increasing the proportion of inertial forces transferred by horizontal out-of-plane behaviour.
2.2.2.b Horizontal Out-of-Plane Failure
Horizontal out-of-plane behaviour will be the dominant out-of-plane failure mode when the
distance between vertical lateral supports (intersecting walls) is considerably less than the
distance between horizontal lateral supports (floor and roof diaphragms). This is often the case
when walls only have a horizontal lateral support at their base.
Horizontal out-of-plane failure is a simpler mechanism than its vertical counterpart because
rocking is not possible in the horizontal direction (gravity can only act as a restoring force
vertically). When the wall bends horizontally to the point where the tensile strength of the
mortar is exceeded, vertical or zig-zag cracks will form in the mortar head joints, as shown in
Figure 2.11. Again, for very weak masonry units, cracking may form in the units themselves.
Zig-zag crack at midspan
Vertical crack at support.
More likely if masonry
wall is not interlocked
with tie column
follows mortar joint
Masonry wall interlocked
with column
Figure 2.11 – Horizontal out-of-plane behaviour of a CM wall: elevation view
Figure 2.12 shows that for a wall supported at both ends, vertical cracking usually occurs at the
ends and halfway between intersecting walls. For walls supported at only one end, cracking
occurs just at the support. Similar to roof bond beams, stiff concrete tie columns at corner
locations push back on the wall edge as it attempts to rotate and lengthen while bending. This
effect is equivalent to adding a horizontal compressive force on the wall, and helps increase the
bending strength.
22
Tension Capacity of mortar exceeded
-> Vertical or Zig-Zag cracks form
Inertia forces from mass of masonry
Concrete Tie Column
at Wall intersection
No lateral support on this side,
possibly due to a doorway
Figure 2.12 – Horizontal out-of-plane behaviour of a CM wall: plan section view
Contributing Factors
The factors contributing to horizontal out-of-plane bending resistance are similar to those for
vertical out-of-plane behaviour. Rather than the h/t ratio, it is the ratio of the wall length
between vertical lateral supports to the wall thickness, or l/t ratio, that is important for horizontal
out-of-plane bending. For high l/t ratios, large mortar tensile stresses from bending can result
from even small lateral inertial forces, making walls with such ratios susceptible to horizontal
out-of-plane collapse.
The number of effective vertical wall supports also influences horizontal out-of-plane stability in
that the l/t ratio for masonry walls laterally supported at only one end is twice that for walls
supported at both ends. This characteristic emphasizes the importance of proper masonry wall to
tie column and tie column to intersecting wall connections.
Stiffer corner tie columns and higher mortar tension strengths also improve horizontal out-of-
plane bending resistance.
Typical Damage
When the mortar tension strength is exceeded, vertical or zig-zag cracks form in the mortar joints
or masonry units at the midspan and ends of walls supported at two ends. Cracking occurs only
at the support of walls supported at just one end.
Situations Where Failure Mode Dominant
Horizontal out-of-plane failure is a concern for walls with high l/t ratios, even higher h/t ratios,
and flexible corner tie columns. Horizontal out-of-plane failures are sometimes due to the
increase in l/t ratio resulting from failure of a vertical lateral support connection.
Ways to Improve Resistance to Failure Mode
As common sense would suggest, increasing resistance to horizontal out-of-plane bending can be
achieved by improving the factors that influence this failure mode: distance between vertical
lateral supports, wall thickness and corner tie column stiffness. Some research and field
23
observations indicate that installing wire reinforcement in the bed joints of the masonry wall can
also improve horizontal out-of-plane behaviour.
2.2.3 Diaphragm Failures
Diaphragm failures occur in the horizontal structural elements of a confined masonry building:
the roof and floors. The horizontal force in a diaphragm comes from the inertial forces
transferred to the diaphragm element from walls perpendicular to the ground shaking that span
vertically out of plane, plus the inertial forces originating in the diaphragm element itself.
Diaphragms must span between walls that are parallel to the ground shaking and act in-plane;
these-in plane walls serve as lateral supports for diaphragms.
A diaphragm fails when the inertial forces on a roof or floor exceed the capacity of the
diaphragm to transfer horizontal forces to its lateral supports. The specific mechanism by which
a diaphragm fails depends on the flooring/roofing material and system. Diaphragm failures are
normally not an issue for concrete floors and roofs in regularly-shaped buildings. Concrete slabs
have large capacities for transferring forces through diaphragm action.
Diaphragm failures are much more of a concern for wood floors. For plywood roofs and floors
to act as diaphragms, the plywood panels and connections between panels must be strong enough
to transfer the applied diaphragm forces. For wood plank roofs and floors, a horizontal truss in
the plane of the roof or floor is usually needed for the element to work as a diaphragm; it is very
difficult for the connections between planks to transfer diaphragm forces, resulting in the need
for a horizontal truss. Figure 2.13 illustrates this situation. Similarly for sloped wooden roof
trusses, a horizontal truss between the top chords of the roof truss is needed for this roof system
to act as a diaphragm.
Tension tie to
prevent middle joist
from moving up
Diaphragm
supports top
of wall
Wood Plank Floor with horizontal truss
Inertial Forces
at floor level
Floor Joists
Nailed
connection can
rotate easily
Wood Planks
Wood Plank Floor without horizontal truss
Diaphragm
does not
support top
of wall
Compression strut to
prevent middle joist
from moving up
Nailed connection
does not have to
resist rotation
Figure 2.13 – Wood plank floor without and with a horizontal truss
24
Contributing Factors
There are several parameters that influence diaphragm resistance. The distance between the
lateral supports of a diaphragm is one factor. The larger the dimension between in-plane support
walls, the greater the diaphragm forces and the lower the inertial loads the diaphragm can resist.
Building shape regularity also affects diaphragm resistance. If a building is non-rectangular or
has large recesses and projections, the diaphragm shape will be irregular, potentially leading to
massive increases in the diaphragm forces.
The strengths of diaphragm materials and connections will of course influence the ability of
floors and roofs to transmit inertial forces. Thick floor and roof structures generally have greater
capacities than thin ones; concrete is generally stronger than wood. If a horizontal truss is used
for wood systems, the strength, size and configuration of the truss elements will factor into the
diaphragm capacity.
Typical Damage
Damage indicative of diaphragm failure includes buckling or cracking of diaphragm panels,
connection failure and rotation of panels, and compression or tension failure of horizontal truss
members. If diaphragms fail, the out-of-plane masonry walls that rely upon the diaphragms for
lateral support may in turn fail from the increase in wall h/t ratio caused by diaphragm failure.
Situations Where Failure Mode Dominant
Buildings with irregular building shapes and long distances between lateral diaphragms supports
(in-plane walls) may have diaphragm problems. Buildings with wood plank diaphragms not
reinforced with horizontal trusses or plywood diaphragms with poor connections between
plywood panels are also prone to diaphragm failures.
Ways to Improve Resistance to Failure Mode
To ensure diaphragm failure does not govern, building shapes should be kept regular and lateral
diaphragm supports should be spaced at reasonable distances. Good diaphragm detailing,
connections and workmanship are also essential.
2.2.4 Connection Failures
Failures of connections in CM buildings are chiefly of concern in that they alter the force transfer
path in a structure and result in a loss in lateral support for some elements. This latter
consequence can cause failure by one of the modes discussed previously in this section.
There are many different types of connections in a confined masonry building, each of which can
have detrimental effects. A few of the more critical connection failures, and the damage
characteristic of each, will be discussed below.
25
Tie column to bond beam connection failures generally come about as a result of in-plane
loading. Premature cracking and failure at the ends of tie-columns and bond beams typify this
failure mode. This type of connection failure can be caused by improperly anchored rebar,
which results in reduced tie column tension and shear strength. Pull-out of inadequately
anchored longitudinal rebar is particularly problematic for CM structures without continuous
connections between floors/roofs and walls. For such buildings, rebar pull-out can induce not
only in-plane failure but also out-of-plane collapse due to the fact that the affected bond beam
loses its ability to laterally support the top of the masonry wall.
Masonry wall to tie column connection failure can occur under in-plane or out-of-plane loading.
Indicators of this connection failure type are cracking and separation along the masonry wall-tie
column interface. Masonry wall to tie column connection failures affect in-plane behaviour by
preventing transfer of tension forces from the wall to the column, possibly leading to tension
cracks in the masonry mortar. This connection failure type affects out-of-plane behaviour in that
it removes a lateral support for horizontal out-of-plane bending, thereby increasing the l/t ratio
for the wall and greatly diminishing the wall’s capacity to transfer inertial forces by this
mechanism. The end result of masonry wall to tie column connection failure may be increased
vertical or zig-zag wall cracking or possibly horizontal out-of-plane wall collapse.
Diaphragm to masonry wall connection failure can also occur as a result of either load transfer to
in-plane walls or load transfer from out-of-plane walls. Cracks along the diaphragm-bond beam
interface are typical of this type of connection failure. If failure is along a bond beam connected
to an in-plane wall, the change in behaviour is similar to a diaphragm failure: force transfer to in-
plane walls is lost. Without this force transfer, the diaphragm can no longer act as a horizontal
lateral support, possibly resulting in damage in or collapse of out-of-plane masonry walls. In
extreme cases, the diaphragm may slide right off the building.
If instead a diaphragm to masonry wall connection failure occurs along a bond beam connected
to an out-of-plane wall, the effects are potentially more localized. The connection failure
removes a lateral support for vertical out-of-plane bending, possibly leading to horizontal
cracking in the wall and in some cases collapse.
Contributing Factors
The first factor that influences the likelihood of a connection failure is the magnitude of force
that must be transferred. Having greater structural wall density, smaller floor heights, lesser
distances between intersecting walls with tie-columns, and lower building mass will reduce the
forces that must be transferred through connections.
Connection detailing and workmanship are also important. If connections aren’t designed and
detailed properly, connection strength may be inadequate. Improperly placed concrete and rebar
can also lead to connection failures.
Situations Where Failure Mode Dominant
It is difficult to pin down the types of buildings where connection failures will be of concern.
26
Structures with poor connection detailing and workmanship tend to be susceptible to this failure.
Ways to Improve Resistance to Failure Mode
One of the biggest advantages of confined masonry buildings over unreinforced masonry
buildings is the quality of the diaphragm-to-wall intersection connections characteristic of the
former. These connections are typically a weak spot in unreinforced masonry structures. The
concrete bond beams in confined masonry buildings enable strong wall-to-diaphragm
connections for an assortment of diaphragm systems. For the case of concrete diaphragms, the
bond beams and floor/roof slabs are cast at the same time, promoting excellent force transfer.
Confined masonry buildings also tend to have better wall intersection connections than
unreinforced masonry structures. The presence of tie columns at wall intersections in confined
masonry buildings help improve connection performance. In particular, when the masonry wall
is interlocked with the tie column and horizontal rebar is installed across the wall-column
interface, wall intersection connections behave quite favourably.
2.2.5 Closure
For the failure modes discussed above, there are common factors that help improve behaviour for
pretty much all cases. Higher material strengths and quality, greater structural wall density, and
shorter spans between supports will generally improve the capacity of all failure modes. Good
workmanship is also essential for confined masonry elements to reach their maximum potential
capacity. Without good workmanship, even a well-designed confined masonry building will not
perform satisfactorily in an earthquake. Similar to the effects of poor workmanship are the
detrimental influence of damage from past earthquakes and deterioration due to aging. The
current condition of a building must be considered when evaluating its seismic resistance.
2.3 Non-Structural Failure Modes
Having discussed the different mechanisms by which the structural elements in a confined
masonry building can fail, it’s now time to examine the failure modes of non-structural
components. Since there are a wide variety of non-structural elements found in confined
masonry housing, it would be rather arduous to discuss the failures observed for all these
components. Instead, this report will look at the failure mechanisms of several heavy non-
structural components, which because of their weight, represent significant life safety hazards.
2.3.1 Masonry Partition Walls
Partition walls are divider walls in the interior of a house that do not support vertically floors or
roofs. As such, if partition walls are constructed of masonry, they are not confined. Interior
walls that support roofs and floors vertically are structural walls and are therefore confined by tie
columns and bond beams, or at least they should be.
27
Masonry partition walls fail by out-of-plane mechanisms and are thereby subject to the
contributing factors and damage types discussed in the out-of-plane failure mode section.
Because partition walls do not support vertically floors or roofs, partition walls do not receive
beneficial vertical load from these elements, decreasing the vertical out-of-plane resistance of
partition walls.
The tops of masonry partition walls are often poorly connected or not connected laterally at all to
diaphragm supports. This forces the partition walls to transmit their out-of-plane inertial forces
by horizontal bending. One situation where partition walls are not usually supported at their tops
is at the underside of sloped roof trusses. The bottom chords of sloped roof trusses are often not
tied together with a horizontal truss and can not therefore work as a lateral support diaphragm.
Thus, it makes no sense to laterally connect the wall to the roof trusses because the roof trusses
can’t act as lateral supports anyway.
Because of the inability for many partition walls to span vertically, these elements rely heavily
on horizontal out-of-plane bending to transmit inertial loads. Unfortunately, in many cases
partition walls are not intersected at both ends by other partition walls or structural walls. Also,
there are no tie columns at partition wall intersections, so intersection connections are sometimes
suspect. Because of the problems partition walls have in spanning both vertically and
horizontally, these elements are quite prone to out-of-plane failure.
Despite the lateral support problems discussed above, there are ways to ensure that partition
walls behave adequately under earthquake loading. Vertical out-of-plane bending can be relied
upon to resist inertial forces, as long as partition walls are laterally connected at their tops and all
lateral supports are able to act as horizontal diaphragms (critical for the case of sloped roof
trusses).
Horizontal out-of-plane bending can be used to transmit inertial forces if the distance between
intersecting walls is reasonable, the span between intersecting walls is not interrupted by large
openings, and the connections at intersecting walls are properly detailed and constructed.
Partition walls can be converted to structural walls, which will generally help improve both
vertical and horizontal out-of-plane behaviour. Alternately, it may be possible to construct
partition walls from materials that are lighter than masonry, which reduces the inertial forces on
the wall and makes it less likely that people will be injured if the wall fails.
2.3.2 Gable End Walls
In buildings with sloped roofs, the triangle-shaped piece of wall at the ends of the building is
called a gable end wall. In many confined masonry buildings, gable end walls are constructed of
masonry units that sit on the roof bond beam and extend to the underside of the sloping roof line.
Figure 2.14 illustrates a typical gable end wall.
It is often quite difficult to connect the tops of masonry gable end walls to the sloping roof
systems. As a result, gable end walls are often only supported at their base, making them
28
susceptible to out-of-plane failure. Many failures of this type have been observed in past
earthquakes.
To prevent failure of gable ends walls, two strategies have been observed. The first strategy is to
construct sloping bond beams on top of gable end walls and extend tie columns from below up to
the sloping bond beams, effectively making the gables a confined element. The connections of
the sloping bond beams must be detailed correctly, which is sometimes an issue. As well, the
sloping bond beams must be laterally connected to the roof to be effective; fortunately, effective
bond beam-roof connections are often much easier to achieve than effective masonry-roof
connections.
Roof
bond
beam
Roof
Masonry Gable End Wall Difficult to connect
gable end wall to roof
Figure 2.14 – Gable end wall in a CM building
The second strategy to prevent masonry gable end wall failures is to replace the masonry with a
lighter material. A gable end wall made from a material lighter than masonry will have lower
inertial forces, is less likely to injure people if the gable fails, and may be easier to connect to the
roof at the top of the gable. Another strategy is to eliminate gable end walls completely by
switching to a hipped roof style.
2.3.3 Parapets
Parapets are defined as the continuation of an exterior wall above the roof bond beam. They are
quite common for buildings with non-sloped roofs. Unreinforced masonry parapets only have
lateral supports at their base, and are therefore susceptible to out-of-plane failure. This helps
explain why masonry parapet collapse is one of the most common types of non-structural failure
modes observed in the aftermath of seismic events.
29
A number of approaches have been developed to mitigate the problem of masonry parapet
failure. In some cases, the height of a masonry parapet is limited to keep the stresses at the
parapet base low. Sometimes, the parapet is eliminated altogether or constructed from lighter
materials. If higher parapets are necessary, the tie columns from the wall below can be extended
into the parapet, and the parapet capped with a concrete bond beam; in this system, the bond
beam and tie columns act as bending members (not simply as confining members as is the case
below the roof bond beam) and must be designed accordingly. If the tie columns are not strong
enough in bending, wing walls perpendicular to the parapet wall can be added to laterally support
the parapet at tie column locations.
2.3.4 Chimneys
On many confined masonry homes, rectangular masonry chimneys are constructed to exhaust
oven and furnace fumes. Chimneys are only laterally supported at their base, and their
slenderness leaves them prone to bending failure and collapse.
It is preferable in seismic zones to replace masonry chimneys with either sheet steel chimneys or
chimneys made of other light materials. If chimneys must be built in masonry, it is beneficial to
make them as short as possible and make the cross-sectional dimensions as large as possible.
2.3.5 Roof Tiles
In many areas where confined masonry construction is popular, heavy terra cotta shingles are the
roofing material of choice. In earthquakes, these shingles sometimes become detached and fall
down off the roof, potentially causing injury.
From a seismic design perspective, using lighter roofing materials would lessen the load on the
shingle to roof connections and lower the overall inertial forces acting on the roof. If using
lighter shingles is not possible, special care should be taken to ensure the connections holding
down the terra cotta shingles are adequate for the expected earthquake loads.
2.4 Summary Tables
Having looked at the individual confined masonry failure modes in detail, the failure modes will
now be summarized in tables outlining typical damage, contributing factors, situations where
dominant, and ways to improve resistance to the failure mode.
30
Contributing Factors Situations wh ere Domina nt Ways to Improve Resistance
Shear and tensile strength of mortar
and masonry units
Improve the shear and tensile
strength of individual panel materials
Brick-mortar bond strength Enhance the masonry-mortar bond
Masonry unit c ompression strength Utilize masonry units with s uperior
compressive strength
Tie column shear strength Improve the tie c olumn shear
resistance
Tie column rebar anchorage
Applied vertical loads
Tie column-masonry wall interface
shear capacity
Enhance masonry wall-tie column
interface shear transfer
Number of tie columns Provide s ufficient wall density in both
principal directions
Tie column spacing Openings: minimize s ize/number,
carefully locate, c onfine
Wall density (length and thickness
of the panels)
Improve workmanship and
construction quality
Size and location of wall openings
Workmanship
Construct panels with aspect ratios
near 1
Masonry compression strength
Mortar tension strength Interlock the tie c olumn and adjacent
masonry panels
Tie column-masonry wall interface
shear capacity
Provide t ie columns with sufficient
longitudinal reinforcement
Vertical load (counteracts for
tension stresses from bending)
Provide c onnection rebar along the tie
column-masonry wall interface
Mortar shear and tensile strength Use mortar with adequate shear
resistance
Brick-mortar bond strength
Workmanship and construction
quality
Vertical load (contributes to
frictional resistance of mortar joints)
Provide panels with a sufficient
number of tie columns of adequate
shear strength
Structural wall density Do not interupt walls with large
unconfined openings
Tie column shear strength
Number of tie columns
Mortar tensile st rength Use high tensile strength mortar
Wall s lenderness ratio (H/t) Decrease wall slenderness ratio
Tall, thin walls
Stiffness of the bond beams
Connections (wall to bond beam
and bond beam to diaphragm)
Walls with flexible roof bond
beams
Mortar tensile st rength
Walls that are perpendicular to
the direction of motion
Utilize high quality mortar and
improve masonry-concrete bond
Tie column-masonry wall bond
strength Long, thin walls
Reduce the panel length-to-thickness
ratio
Wall length-to-thickness ratio
Walls with poor masonry-to-tie
column connections
Connections (wall to tie c olumn &
tie column t o intersecting wall)
Walls with flexible corner tie
columns
Type Typical Dama ge
1- In -plan e S tru c tura l failu re m od es
Di a g o n al S h ea r Fai lur e
Initial diagonal cracks in the middle of the walls
and/or at opening corners .
Masonry crushing in t he middle of the panels or
at corners; eventually splitting apart of wall along
main diagonal
Horizontal bending cracks in tie columns
S lid ing S h e ar Fai l u re
Horizontal cracks in horizontal mortar joints
across entire panel length and into tie c olumns
Walls that are parallel to the
direction of the ground motion
Extension of initial diagonal cracks along the
principal diagonals through mortar joints and/or
masonry units, depending on the relative
strength/stiffnes s of the materials
Panels that are under the
influence of significant lateral
and vertic al forces (first story
walls)
Provide t he panels with adequate and
properly distributed confinement
Cracking at tie c olumn ends, eventually leading
to tie c olumn shear failure
Well-proportioned panels
Separation of masonry panels and tie columns
along their vertic al interface
Provide wire reinforcement in the
horizontal mortar joints
In -pl an e B end ing F ai l u re
Tension side: Horizontal cracking at the base of
the panel from mortar tension strength being
exceeded
Wall aspect ratio (walls with large
panel heights and small panel
lengths are susceptible)
Walls with high aspect ratios
(bending behaviour dominates
shear behaviour) Use high tensile capacity mortar &
high compressive strength units
Compression side: Masonry crushing This type of failure is not
common for well-confined
walls that are properly
interconnected with tie
columns
Panels with low quality mortar
and insufficient confinement
Walls that are under the effect
of low vertical loads but
relatively high shear forces
(lower parts of upper story
panels) Improve workmanship and
construction quality
Enhance unit-mortar bond strength
Reduce the magnitude of force
transferred vertically out-of-plane by
reducing distance between
intersecting walls
Vertical loads Upper story panels with lower
vertical st resses
Provide s tiff tie columns that are
properly connected to both masonry
panels and intersecting walls
Tie column stiffness at corner wall
intersections
Walls with much les s distance
between intersecting walls
Walls that are perpendicular to
the direction of ground motion
Provide s tiff bond beams that are
properly connected to both panels
and diaphragms
Walls with closer spacing
between floors/roof than
Distance between lateral supports
2-o u t-of-p lan e s tru c tu ra l failu re m o de s
V erti cal o ut-o f-pl a n e fail u re
Horizontal cracks in mortar joints near the wall
supports and mid-height; for panels only
constrained at their base, cracks concentrate
near the support
Collapse of wall and the subsequent collapse of
roofs/floors it s upports
Hor i z o n ta l ou t-of -p l an e
fai l u re
Vertical or zigzag cracks in mortar joints at wall
ends and halfway between intersecting panels;
for walls supported only at one end, cracks
concentrate at the support
Collapse of wall and the subsequent collapse of
roofs/floors it s upports
Table 2.1 – Failure mode summary table, part 1
31
Contributing Factors Situations whe re Dominant Ways to Improve Resistance
Type of diaphragm s ystem Irregular buildings with many
large projections and recesses Keep the building shape regular
Material strength and workmanship Buildings with heavy roof/floors Construct the diaphragm with proper
materials
Building configuration and regularity Buildings with flexible
diaphragms
Buildings with poor diaphragm
connections
Place lateral diaphragm supports at
reasonable distances
The distance between diaphragm
lateral supports
Rigidity of the flooring/roofing
system
Shear resistance of tie columns Anchor column longitudinal rebars
properly
Reinforcement detailing Design and detail connections
properly
Workmanship Increase wall density or reduce the
distance between tie columns
Tie column-masonry wall interface
force transfer capacity
Long panels with t oo few tie
columns
Interlock tie columns and masonry
panels by toothing end of panel
The distance between and number
of tie columns Low quality mortar Provide c onnection rebars across the
wall-tie column interface
Amount of force transferred by
connection
Inadequate bond between tie
columns and walls
Increase wall density and reduce the
distance between tie columns
Detail the connections between
panels and diaphragms properly
Slender partition walls Connect partition walls adequately to
their lateral supports
Reduce the distance between lateral
supports/slenderness ratio
Partition walls with one free
end
Construct partition walls from lighter
materials
Gable end walls not properly
confined along borders
Laterally connect t op of gable end
wall back to roof structure
Extend tie columns into parapet and
cap with a bond beam
Limit the height of the parapets
Replace masonry parapet with lighter
materials
Cross sect ional area Reduce the height of chimneys and
make their cross -section large
Height
Construct chimneys from light
materials such as sheet steel
Weight of cover materials Make roof coverings as light as
possible
Connection detailing Connect them properly to their
supporting roofing system
Buildings with wood plank
diaphragms not reinforced with
horizontal trusses
Select a diaphragm system effective
at transfering inertial loads
Properly detail connections between
diaphragm components
The quality of connections between
different diaphragm parts
Buildings with long distances
between lateral diaphragm
supports
Provide wood plank floors/roofs with
horizontal trussesStrength, size and configuration of
horizontal trusses for wood
floors/roofs
Type Typical Dama ge
3- D iap hr a g m Fa ilu re M o d es
Buckling or cracking of diaphragm panels
Rotation of diaphragm panels
Connection failure between different diaphragm
components
Tension or compression failure of horizontal truss
members (wooden floors/roofs)
4-C on ne c tion F a ilure M od e s
Tie co lum n-
B on d b ea m
Premature cracking and shear failure at the end
of tie columns
Tie column-bond beam joints
with improper detailing
(Longitudinal rebars do not
have adequate anchorage)
Wa ll-Tie co lum n
Vertical or zigzag cracking along the masonry
wall-tie column interface
Horizontal out-of-plane failure of the panel
W a ll-D iap hr agm
Cracking along the diaphragm-bond beam
interface
The magnitude of loads being
transferred
Large connection loads must
be transferred Keep story height and the distance
between diaphragm lateral supports
reasonable
In extreme cases, diaphragm can slide off the
building
Workmanship and connection
detailing Improper c onnection detailing
Vertical out-of-plan failure of the masonry wall
panel
The distance between lateral
diaphragm supports
Excessive distance between
lateral diaphragm supports
Provide t he building with sufficient
wall density
5- N on -S tr u c tu ral F ailu re M o de s
P ar ti ti o n W all s
From vertical out-of-plane behaviour, horizontal
cracking at the wall base, mid-height and top; if
wall not laterally connected at top, horizontal
cracking concentrated at wall base
Factors that contribute to t he out-of-
plane wall behaviour except tie
columns and bond beams (partition
walls not confined) and vertical load
(partition walls not load bearing)
Gab le E n d
W al l s
Horizontal bending cracks from vertical out-of
plane behaviour Same as for partition walls
P ar ap et s
Horizontal bending cracks from vertical out-of-
plane behaviour and out-of-plane collapse Same as for gable end walls
Partition walls with improper
connections t o their lateral
supports
From horizontal out-of-plane behaviour, vertical or
zig-zag cracking at wall ends and mid-length
Enhance bond with intersecting walls
with connection rebar or toothing
Collapse of partition wall
Long and tall gable end walls Provide a sloping bond beam on top
of gables and extend tie columns up
Collapse of gable end wall Confinement from bond beams and
tie columns are also factors
Gable end walls from heavy
materials
Replace gable end walls with li ghter
materials
Tall, unreinforced parapets
constructed from heavy
material
Ch im n ey s
Horizontal cracks at base of chimney; collapse
Tall, slender, unreinforced
chimneys made of heavy
masonry units that are not
properly anchored to the main
structure
Provide longitudinal reinforcement
along the chimney perimeter
Reinforcement (Confinement)
Ro of
co v er ing s
Detachment from the roof
Heavy coverings with improper
connections t o their
supporting roof
Table 2.2 – Failure mode summary table, part 2
32
3 CURRENT PRACTICE
Before looking at the various resources that indicate how confined masonry construction should
be practiced, it is necessary to first investigate how this building system is currently practiced.
The latter is the objective of this section. First, a brief history of confined masonry will be
presented, showing how confined masonry has evolved into its current form. Next, the different
types of confined masonry houses currently being built in different parts of the world will be
described and compared. Finally, the shortcomings and strengths of current confined masonry
housing will be discussed. This last item will form the bulk of the section.
3.1 History
Confined masonry housing is generally seen as an extension and improvement upon more
conventional unreinforced masonry construction. Unreinforced masonry structures have been
around for thousands of years, with the monumental architecture of Ancient Egypt, Cambodia
and Mesoamerica being constructed of unreinforced masonry. Even today, a huge number of
buildings, in particular residential buildings, are made of unreinforced masonry.
Unfortunately, unreinforced masonry has been observed to behave poorly during earthquake
events. To mitigate this poor seismic behaviour, around a century ago, local constructors started
installing bond beam and tie column members where they observed cracking and damage in
unreinforced masonry buildings during past earthquakes. Adding confining elements helped to
tie the building together and prevent out-of-plane wall failures. Confining members were also
installed at connection failure locations to enable more stable connections between walls and
diaphragm elements.
In Iran and India, wood confining elements were used to strengthen unreinforced masonry
housing. In several Latin American countries, concrete bond beams and tie columns were
implemented. Figure 3.1 shows a non-engineered confined masonry wall in Mexico. Further
refinement of the concrete confining element system by constructors, engineers and researchers
has culminated in the confined masonry structures being built today.
Figure 3.1 – Confining elements added by local construction workers to improve seismic performance
33
The idea that inspired local constructors to install confining elements in unreinforced masonry
buildings is older than a mere 100 years, however. Much earlier than 1900, Mesoamericans used
the concept of confining elements in the bajareque houses they constructed. The walls of
bajareque houses consisted of wooden or bamboo tie columns entwined with leather straps to
form two meshes. Between these meshes were installed mud, rocks and stones. The end result
was a rudimentary confined masonry system that performed quite well in earthquakes.
3.2 Summary of Current Practice
To get an idea of what confined masonry construction practices are currently in use, the World
Housing Encyclopedia (WHE) was consulted. The WHE is a web-based database of the housing
types found in seismically-active regions around the world. The WHE is operated and
maintained by the Earthquake Engineering Research Institute (EERI) and the International
Association for Earthquake Engineering (IAEE). The individual reports that make up the
encyclopedia are compiled by volunteer engineering professionals. Contributors visit housing
sites during construction and after seismic events to record the characteristics of each housing
type and its performance during earthquakes. In many cases, the engineers submitting reports
are local to the area being reported upon.
The value of the WHE lies in the fact that it provides a broad listing of housing types in a
common format. This enables housing types to be easily compared and contrasted, often with
respect to earthquake resistance. The WHE contains both engineered and non-engineered types
of construction. There are also reports for a wide variety of building materials, including
reinforced concrete, steel, adobe masonry, stone masonry, unreinforced masonry, reinforced
masonry and confined masonry. Each report details a style of house in a particular area and
contains information on construction type and materials, structural system, building layout,
seismic design flaws, construction quality, performance in past earthquakes, regulating bodies
and professional involvement, and socio-economic characteristics.
At the time this report was written, there were ten reports in the WHE on confined masonry
housing in seven different countries: Argentina, Chile, Iran, Kyrgyzstan, Peru, Serbia &
Montenegro and Slovenia. Table A.1 in Appendix A contains a detailed description of these
reports and provides a brief comparison. Confined masonry is also practiced in a number of
countries other than those listed above, including El Salvador, India, Indonesia and Mexico.
That being said, the reports in the WHE represent a decent cross-section of the confined masonry
housing in existence and provide good insight into the current practice of CM construction.
Looking at the WHE reports, the current practice of confined masonry varies considerably from
country to country with respect to materials (masonry unit composition and size; floor and roof
type), layout (building plan and number of floors), confining elements (size, location and
reinforcement), construction practices, regulatory influence and cost. Despite all these
differences, confined masonry buildings in all these places are still at their core relying upon the
same structural system to resist earthquake forces: unreinforced masonry walls with reinforced
concrete elements that work to confine the masonry walls and help them transmit in-plane and
out-of-plane lateral forces better than unreinforced masonry on its own.
34
3.3 Shortcomings of Current CM Construction
Although the practice of confined masonry varies considerably from country to country, analysis
of the WHE reports reveals some recurring problems. Where these problems occur and what
failure modes they contribute to, will now be presented for each type of deficiency. The
shortcomings in current confined masonry housing can be roughly organized into 3 categories:
seismic design flaws, construction and post-construction concerns, and lack of regulation.
3.3.1 Seismic Design Flaws
Seismic design flaws refer to components in the design of a building, that even if the building
were constructed with good workmanship and proper materials, would cause the building to
perform poorly in an earthquake. From the WHE reports, five recurring design flaws are evident
in current confined masonry housing construction practice.
The first seismic design flaw is irregular building configuration. With respect to plan layout,
irregular building configurations consist of projections from the rectangular core of the building,
and unequal lengths or unsymmetrical distributions of structural walls in the two principal
building directions. These irregularities cause the building to go into torsion under earthquake
loading, increasing the complexity of force transfer in the structure, forcing the horizontal
diaphragms and their connections to work harder, and potentially increasing the lateral forces
that must be resisted by some in-plane walls. These torsional effects can contribute to diaphragm
failures, diaphragm-to-wall connection failures and in-plane wall failures. Plan irregularities in
building layout have been noted in confined masonry housing in Iran, Kyrgyzstan and Peru.
What could be considered a vertical building irregularity for non-engineered confined masonry
structures is a building greater than two stories in height. Such buildings have more complex
load transfer paths and higher lateral forces than shorter structures. Thus, they are more prone to
in-plane shear failure. Confined masonry houses three to six stories tall were identified in Chile,
Iran, Peru and Serbia & Montenegro.
It should be noted that irregular building configurations can be mitigated when custom
engineering design of a confined masonry house is performed. The problems arise when
principles and guidelines for regular, symmetric, one or two story buildings are applied to
irregular buildings. In Chile, Peru and Serbia & Montenegro, engineer involvement in design
and construction is noted in the WHE reports. Presumably then, the irregular and multi-story
nature of these buildings has been explicitly designed for in these cases. At this point, however,
the buildings cease to be non-engineered and thus move beyond the main focus of this report.
For non-engineered confined masonry structures, building irregularities remain a design flaw.
The second recurring design flaw noted in confined masonry buildings by the WHE reports is
inadequate confining elements. Inadequate confining elements can mean tie columns missing at
large wall openings, excessive spacing between tie columns and tie columns missing at wall ends
and intersections, inadequate tie column shear strength, and poor confining element rebar
detailing. Buildings with inadequate confining elements are more prone to in-plane and
35
connection failure modes. Below are photos of the damage resulting from each of the types of
inadequate confining element, along with the countries these design flaws have been observed in.
Note that some of these photos are from Mexico, a country where WHE confined masonry
reports could not be located.
Figure 3.2 – Failure at an opening due to no tie
column; design flaw noted in Chile and Mexico
Source: Zabala et al., 2004
Figure 3.3 – Failure due to tie column missing in
corner; design flaw noted in Chile, Mexico, Peru
and Slovenia
Source: EERI, 2003
Figure 3.4 – Shear failure of a tie column;
design flaw noted in Chile and Peru
Source: EERI, 2003
Figure 3.5 – Inadequate rebar detailing: stirrups not
hooked properly; design flaw noted in Chile, Peru and
Serbia & Montenegro
Source: Blondet et al., 2004
Failures at roof-to-wall connections are the third recurring design flaw noted in the WHE reports.
In some confined masonry houses in Chile and Peru, the roof is not connected to the masonry
wall in a way that provides lateral support to the top of the wall. The lack of lateral support at
the roof level contributes to vertical out-of-plane damage and failure. Figure 3.6 shows a photo
of a roof-to-wall connection that does not provide lateral support.
36
In general, the confined masonry houses in the WHE reports are relatively “stocky” (have walls
with low h/t and l/t ratios) and are therefore not that prone to out-of-plane failure modes unless
connection problems are an issue. In Indonesia, however, confined masonry walls are more
slender and may be subject to out-of-plane problems. Unfortunately, information on how
confined masonry structures perform in Indonesia could not be found in the WHE.
The fourth design flaw observed in the WHE confined masonry building reports is floor and roof
systems that do not work as horizontal diaphragms. The confined masonry system relies on the
roof and floors to transfer inertial loads through diaphragm action to the in-plane walls. In most
confined masonry structures, concrete or concrete-masonry composite roofs and floors fulfill the
role of horizontal diaphragm quite effectively, although there is some concern with the rigidity of
these elements in Serbia & Montenegro.
Figure 3.6 – Inadequate roof-to-wall
connection
Figure 3.7 – Gable end wall failure due to lack of support and
confinement along roof line
Source: French, 2004 Source: Lutman and Tomazevic, 2002
In Chile and Iran, however, wooden members are sometimes used to support the roofs and floors,
and the resulting floor/roof system is not able to transfer inertial loads as a diaphragm. As
detailed in the failure modes section of this report, ineffective horizontal diaphragms can lead to
out-of-plane wall failures and walls or floors sliding off the building. In some cases, wooden
roofs in these countries support heavy clay tile shingles, which impose large inertial forces on the
diaphragm during an earthquake, increasing the probability of damage or failure.
The fifth design flaw noted in the WHE reports is inadequately supported masonry gable walls, a
non-structural failure concern. This problem appears to be particularly troublesome in Slovenia,
where masonry gables are not confined and supported along their top edge, facilitating an out-of-
plane failure mode. Figure 3.7 illustrates this design flaw.
37
3.3.2 Construction and Post-Construction Concerns
Construction and post-construction concerns are items or activities that prevent well designed
confined masonry buildings from behaving adequately in seismic events. Construction concerns
include poor material quality and bad workmanship; housing owners or tenants modifying
structural components without professional guidance is a post-construction concern.
Material quality is a major concern noted in the WHE reports from Chile, Iran, Kyrgyzstan and
Peru. The poor quality of mortar was emphasized in the confined masonry buildings inspected.
Poor mortar quality, and in effect poor material quality overall, affects every kind of failure
mode. Thus, improving material quality is one of the most effective ways to enhance seismic
performance.
Bad workmanship was another chief criticism of the confined masonry houses reported on in
Chile, Iran and Kyrgyzstan. Workmanship was also a slightly less urgent concern in the Peru
and Slovenia WHE reports. Poor mortar joint detailing and neglecting to soak masonry units
prior to installation (so they don’t seep water out of the mortar) were two workmanship
deficiencies that showed up in a number of reports. Figure 3.8 and Figure 3.9 illustrate these
problems. Similar to material quality, quality of workmanship affects every type of failure mode
and is therefore a concern on every building.
Figure 3.8 – Poor mortar joint workmanship
Source: Blondet et al., 2004
Figure 3.9 – Unsoaked masonry units absorbing water
from mortar, weakening mortar and mortar-unit bond
Source: Lutman and Tomazevic, 2002
In many of the WHE reports, modifications to the original building were noted. These
modifications included adding new floors, extending the building in plan (often into an L-shape
or other irregular configuration), adding interior partition walls, removing interior structural
walls, and adding openings for doors and windows. The modifications were generally
undertaken without professional advice and all have the potential to decrease the seismic
performance of the structure. In essence, modifications performed without professional
consultation could transform a building that originally had good seismic behaviour into a
structure with inadequate resistance to earthquakes.
38
3.3.3 Lack of Regulation
The presence of design flaws, material problems and poor workmanship in some of the confined
masonry houses in the WHE reports can be partially attributed to the lack of regulatory and
inspection systems for confined masonry in those regions. There is no mechanism in place to
catch bad designs, materials and workmanship, which often means that these problems only
come to the surface after seismic events occur, when it is too late.
The first way in which regulation for confined masonry housing is lacking is the absence of
applicable codes and guidelines. In the WHE reports, the buildings reported on were subject to
certain building and construction codes in all cases. In many instances, though, the governing
standards are seismic or masonry codes that provide little guidance on how confined masonry
buildings should be designed. As a result, there is often little guidance available to help in the
design of confined masonry buildings.
To fill this gap, generic guidelines addressing how to design and construct confined masonry
buildings have been developed in several countries; some of these guidelines will be examined
later in this report. Guidelines, however, do not exist in all the areas using confined masonry and
even if they do exist, they are often not utilized or adhered to. The same could be said for
confined masonry building codes. Thus, even in regions that have applicable codes or
guidelines, the absence of mechanisms to enforce governing regulations results in many confined
masonry houses being built without any consideration to regulations.
For many types of construction other than confined masonry, a three-pronged regulatory
enforcement infrastructure is in place: building design review, material quality control and
construction inspection. This infrastructure ensures that building design, material quality and
workmanship are all at an acceptable level. In the WHE reports, the lack of design and
construction control and inspection for confined masonry housing is explicitly stated in the
reports from Kyrgyzstan, Serbia & Montenegro and Slovenia. It is likely that inadequate quality
control and inspection is an issue in many other regions as well. Functioning design review,
material quality control and construction inspection mechanisms would go a long way in
improving the seismic performance of confined masonry buildings.
It should be noted that the need for applicable codes or guidelines and regulation enforcement
infrastructure can be alleviated if there is intimate professional involvement in the design and
construction of a confined masonry building. If an engineer designs and takes responsibility for
the layout of a confined masonry house, this negates the need for a governing code or guideline.
Similarly, if an engineer inspects the material quality and workmanship of a confined masonry
structure during construction and certifies that these items are adequate, it would be no longer
necessary for a regulatory body to demand material test reports and conduct building inspections.
To require engineers to perform the above functions on every confined masonry house would far
exceed the current level of involvement and be economically unfeasible in many cases. In some
instances, such as irregular building shapes and three or more story buildings, engineering
participation is essential, especially in the design phase. For simpler structures, however,
adequate seismic performance can often be more economically ensured through codes or
39
guidelines enabling non-engineered design, design review by a regulatory body, and
requirements for material quality control and construction inspection.
3.4 Positive Aspects of Current Practice
Although there are some aspects of current confined masonry practice that need to be improved
upon, there are also some areas where the current practice does quite well. For instance, all the
WHE reports reviewed indicate that the foundations for confined masonry houses appear to be
performing adequately. Also, the WHE reports suggest that there are generally enough structural
walls in current confined masonry buildings to resist the earthquake loads on the structures. If
these walls are properly confined, symmetrically distributed and constructed with good materials
and workmanship, the buildings should exhibit favourable behaviour during seismic events.
In some regions, confined masonry buildings are already behaving excellently under seismic
loading; Argentina comes to mind in this respect. Even in locations where deficiencies have
been noted, confined masonry buildings are still vastly out-performing their unreinforced
masonry counterparts. Table A.2 in Appendix A lists observations of building damage from
earthquake events in regions utilizing confined masonry construction and other types of housing
systems. As an extreme example, in one Chilean earthquake, 16% of confined masonry houses
in the quake zone partially or completely collapsed, as compared to collapse percentages of 57%
for unreinforced masonry buildings and 65% for adobe masonry structures. Thus, even
imperfectly constructed confined masonry buildings are reducing seismic risk for their
inhabitants. With improvements in confined masonry practice, this risk can be further reduced.
Given that confined masonry generally provides better seismic performance than unreinforced
masonry, the fact that confined masonry housing has several characteristics fostering growth of
the building system is also quite promising. Since confined masonry buildings require many of
the same materials and techniques as unreinforced masonry construction, the shift for home
builders from the latter system to the former is not a considerable one. As well, the cost of
confined masonry is only marginally greater than unreinforced masonry, the premium for
confined masonry essentially being the cost for its reinforced concrete confining elements.
These two characteristics, along with superior earthquake resistance, are making it easier to
convince constructors, and in turn owners and tenants, that confined masonry structures are a
feasible housing alternative.
3.5 Closure
Confined masonry housing has come a long way since the first attempts were made at improving
unreinforced masonry structures with wood and concrete confining members. Even in its current
form, confined masonry has a considerable advantage over unreinforced masonry with respect to
seismic resistance.
To this point, however, much of the evolution of confined masonry has come about empirically,
with local builders and inhabitants adding and changing the location of confining elements to get
40
better seismic behaviour. The practice of confined masonry could benefit significantly from
greater professional involvement and the application of engineering principles.
As mentioned several times previously in this report, confined masonry design guidelines have
been developed to help improve the seismic behaviour of this building system. The guidelines
were generated by taking current confined masonry practice and applying engineering
knowledge to see how the building system could be altered to improved earthquake resistance.
The guideline approach has been adopted in many cases rather than a code approach, because the
former can be quicker to implement and provides recommendations directly to builders. In
comparison, codes can take a considerable amount of time to develop and formalize, and often
require interpretation by a professional in the design process. In the next section, several
confined masonry guidelines will be examined and compared.
41
4 GUIDELINE COMPARISON
As mentioned previously, only confined masonry houses that are non-engineered and two stories
or less will be considered in this report. There are two reasons why the report scope was
narrowed in this way. First, a large portion of the global confined masonry building stock fits in
this category. Second, these types of buildings are usually simple; i.e. geometrically regular,
rectangular and symmetric along both principal directions. It is much easier to predict the
behaviour of simple buildings, and as a result, such buildings can be designed effectively by
prescriptive methods, rather than rigorous engineering analysis. This prescriptive approach is the
methodology utilized by confined masonry guidelines.
4.1 Introduction to CM Guidelines
In this report, guidelines developed by City University [
Virdi and Raskkoff, 2005
], Catholic
University of Peru [PUCP and SENCICO, 2005], the International Association of Earthquake
Engineers (IAEE) [IAEE and NICEE, 2004], UNESCO [Ghaidan, 2002], and Tom Schacher
[Schacher, 2006], a Swiss architect, will be discussed. A comprehensive comparison of the
guidelines is presented in Table A.3 of Appendix A. A brief description of each of the
guidelines is included here, along with a table summarizing the recommendations of the different
guidelines for some of the more critical CM housing parameters.
4.1.1 City University Guideline
The City University guideline is primarily based on Eurocodes 6 and 8 [European Committee of
Standardization, 2003 and 2002] and often refers to these standards. The guideline addresses
unreinforced masonry, reinforced masonry and confined masonry buildings. Both engineered and
non-engineered structures are taken into account, and a prescriptive design procedure is
presented for buildings called “simple buildings”. The guideline assumes that confining elements
(tie columns and bond beams) carry no lateral or gravity loads. Rather, the masonry walls are
relied up to resist all of the gravity and lateral loads applied to the system.
This guideline targets primarily engineers; the information presented is quite technical and may
be difficult to interpret for those without substantial engineering education. For example, in
addition to prescriptive construction rules for “simple buildings,” design information is also
provided for calculating the resistance of confined masonry walls in more complex structures.
In comparison to some of the other guidelines, the City University Guideline is not fully
comprehensive, in that it does not provide all the information required to design a confined
masonry house. Items such as foundations, site selection, and concrete element detailing are not
dealt with in the guideline, although references addressing these topics are provided.
The guideline allows for a reasonable amount of flexibility in building design and gives detailed
recommendations for simple cases. The basic principles employed are clearly outlined, giving
42
engineers the building blocks to design larger and more complex structures. Design information
is also provided on non-standard layouts, and configurations such as non-structural elements,
balconies, and gable end walls.
Level of seismicity is explicitly taken into account in the City University Guideline, with sites
being classified according to peak ground acceleration (PGA). Three different ranges are
considered (PGA < 0.2g, 0.2g < PGA < 0.3g and PGA > 0.3g), with different prescriptive
parameters provided for each case. No consideration is given to other seismic parameters such
as soil conditions.
4.1.2 IAEE Guideline
The IAEE guideline offers basic concepts and construction techniques to improve the earthquake
resistance of commonly built non-engineered houses. General concepts of earthquake resistant
design are addressed in Chapter 3, while masonry buildings are addressed in Chapter 4 of this
guideline. Construction rules and structural details mainly focus on solid fired brick masonry
units and the confined masonry system as a whole; hollow block masonry and reinforced
masonry walls are also discussed, however.
Similar to the City University guideline, the language of the IAEE guideline is quite technical
and may not be easily understood by builders or developers without engineering backgrounds.
Throughout the guideline, description of the dominant failure modes both in terms of the non-
structural elements, whose failure could be quite hazardous, and structural components provide
guidance on how to reduce overall risk.
To provide better insight, comprehensive discussion of building location, wall properties,
confining elements and non-structural components are supported in the guideline by figures and
tables. Although the basic design principles and assumptions required to analyze less typical
cases are not explicitly stated in the IAEE guideline, some flexibility is provided. Building
layouts of varying size and height (up to 4 stories) are presented, as are design rules for simple
buildings up to two stories.
The IAEE guideline differs from the other guidelines in that the tie columns suggested have
small cross-sections (as small as 75mm x 75mm), only one longitudinal rebar and no stirrups.
Larger cross-section tie columns (the smallest specified being 150mm x 150mm) with 4
longitudinal rebars minimum and stirrups are advocated by the other guidelines.
Seismicity is taken into account in the IAEE guideline by considering Modified Mercalli
intensity rather than peak ground acceleration. Intensity is broken down into four seismic zones;
the design procedure then takes into account the applicable seismic zone, soil type and
importance to classify buildings into 4 categories (I to IV, refer to Appendix A). The building
category is a factor in the required longitudinal rebar in tie columns and bond beams.
The IAEE Guideline was developed by the IAEE itself and is not based on any one code. It is
thus code-independent, which makes it easier to apply the guideline in different countries.
43
4.1.3 Peru Guideline
The Peru Guideline for the Construction and Maintenance of Masonry Houses outlines the
different earthquake hazards that exist in Peru. The guideline identifies the important properties
that an earthquake resistant house should possess and describes the ideal locations where the
house should be situated. Only one and two story non-engineered houses utilizing the confined
masonry system are considered.
The aim of the Peru guideline is to help builders construct earthquake resistant dwellings. The
sequence and details of construction are clearly presented and advice on maintenance after
construction is given. Throughout the guideline, simple illustrations are provided to help the
reader visualize what is being explained. For instance, there are figures showing good examples
of the proper and unacceptable techniques associated with the construction of masonry houses.
There is little flexibility in the guideline for altering the building layout or detailing of structural
elements; the guideline utilizes a prescriptive approach to the construction sequence and the
structural detailing of the confined masonry elements. On the other hand, the Peru guideline is
completely comprehensive: it provides all the information necessary to construct a confined
masonry house. That being said, if one strays from the sample building plans, it is not exactly
clear where tie columns should be located, and the potentially hazardous effects of non-structural
masonry elements are not explicitly addressed.
The guideline identifies three different seismic zones in Peru: Z1-low seismicity, Z2- medium
seismicity, and Z3 high seismicity. It appears, however, that varying seismicity is not considered
and one design is presented for the all three zones. It is assumed that the guideline uses a
conservative approach when considering seismic hazard. The lack of definition to the seismic
zones makes it difficult to apply this guideline in another region with different seismicity.
The Peru Guideline was developed using a number of sources. The guideline is based on good
CM building practice, the knowledge of the individuals involved in its development, research
projects conducted at PUCP, and the Peruvian National Building Code.
4.1.4 Schacher Guideline
This guideline discusses simple confined masonry buildings that are up to two stories tall and
embody the characteristics (regular in plan and elevation, symmetric, rectangular) that make
them exempt from seismic verification according to Eurocode 8, one of the resources the
guideline is based on.
Attempts have been made in the Schacher guideline to make the language as accessible as
possible. As well, the guideline contains comprehensive figures to illustrate construction rules,
helping the user visualize what is being explained. The presence of some technical concepts and
advanced construction procedures, however, necessitate that the reader have some engineering
background and familiarity with construction works. Consequently, the average homeowner
may have difficulty implementing the guideline on their own.
44
Although the guideline is quite rigid with respect to building layout, it allows for horizontal and
vertical extensions of the core house, for the vertical case, to a maximum of two stories. These
extensions enable the prescribed confined masonry system to be used not only for dwellings, but
also for public buildings such as health centres and offices. In addition, even more so than the
Peru guideline, the Schacher guideline is completely comprehensive, containing all the
information one requires to build a confined masonry house.
Different levels of seismicity are not considered by the Schacher guideline. Rather, a single
design capable of withstanding a peak ground acceleration of 0.35g is presented. This limitation
may make it difficult to apply the guideline in regions were higher accelerations are expected.
The guideline is primarily based on Eurocode 8 and Swiss Norms 261, 262 and 266; it also
satisfies the Iranian Seismic Code of Practice, No. 2800.
4.1.5 UNESCO Guideline
Although the UNESCO Guideline is based extensively on Eurocode 8, it does not contain all the
information required to be considered a complete guideline. It is, however, a good resource on
the topic of CM housing and has therefore been included in this report for completeness.
In the UNESCO Guideline, dominant modes of failure for structural and non-structural
components are explained, as is ground failure. A number of measures are also presented for
enhancing the seismic performance of non-engineered masonry buildings. The UNESCO
Guideline provides general recommendations relevant to all categories of masonry, which makes
it flexible with respect to building configuration. For the purpose of this report, only the parts of
the UNESCO guideline that pertain to confined masonry buildings have been extracted.
The UNESCO Guideline considers confined masonry buildings up to three stories high; this
differs from most of the other guidelines, which limit the number of stories to two. Although not
clearly stated, it appears that the recommendations in the UNESCO guideline are intended for
“simple” structures, i.e. regular and symmetric buildings.
The language in the UNESCO guideline is not very technical, however it refers to basic technical
concepts (e.g. shear, stresses) which necessitate the reader to have some minimum engineering
background. Illustrations are provided to help describe aspects such as building location, wall
construction, openings and non-structural elements.
The UNESCO guideline lacks information on foundations and tie column reinforcement and
size, making it less comprehensive than other guidelines. The lack of detailing information
means that a complete and consistent seismic design can not be achieved solely from this
guideline. Furthermore, it is not clear how seismicity is accounted for in the guideline; thus, it is
unknown for what level of seismicity the UNESCO guideline is applicable
45
4.1.6 Guideline Summary Table
As mentioned previously, Table A.3 in Appendix A provides an extensive list of the information
given in each of the guidelines and compares how this information agrees and differs between
guidelines. Table 4.1 and Table 4.2, below, summarize Table A.3 by including only the most
critical items of information in each of the guidelines. These tables were developed to provide
some additional insight into the nature of the guidelines within the body of the report itself.
City University IAEE Peru Schacher UNESCO
Guideli ne
Basis Eurocode 6 and 8 Developed by IAEE; code
independent
Local practice, P eru Code,
research
Eurocode 8, Swiss Norms
260, 261, 262, 266 Eurocode 8
Seismsicity
Design values differ depending
on Peak Ground Acceleration
Design val ues differ depending
on Modified Mercalli Intensity
Same design for all 3 seismic
zones in Peru
Proposed building designed
for a PGA of 0.35g Design seismi city not stated
Site
Sele ction
Pick sites free of risk from
ground rupture, slope
instability or liquefaction
Away from unstable slopes,
sensitive clays, liquefiable
sands
Away from landslide areas,
landfills, uncompacted fill,
flood plains, river beds,
irrigation ditches
Away from steep slopes,
retaining walls, river beds
Away from expansive clays ,
loose sands, unst able hills
Building
Configuration
Regular in plan & elevat ion;
aspect ratio < 4; walls &
openings symmetric in both
directions; 4 floors max Low
PGA, 2 floors max Hi gh PGA
Rectangular; walls & openings
symmetrical in plan and
elevati on; aspect ratio < 3;
2 floors max
Rectangular; walls & openings
symmetrical; aspect ratio < 3;
2 floors max
Simple, rectangular,
symmetric al ; aspect ratio < 3;
2 floors max
Walls & openings symmetric
in plan & elevation; aspect
ratio < 4; 3 floors max
Wal l
Openings
Align openings vert . & tops of
openings horiz.; openings
symmetric in plan and
elevation; min adjacent wall
length > 1/3 height of opening;
install ty pical tie c olumns if
opening area > 2.5m2; RC
lintels with 250mm bearing
length recomm.; no info on
lintel reinf.
Openings symmetric &
preferably small; c omplicated
criteria for size, min s pacing
and min end distance for
unconfined openings; if criteria
not met, provide vert. & horiz.
one-bar concrete confining
members; bond beam at top
of door elev.; l ocate all
openings at bond beam
Align openings vert . & tops of
openings horiz.; openings
symmetric & away from wall
ends; minimize openings;
opening width < 1/2 wall
length; typical tie columns
recomm. at openings; locat e
openings at/near bond beams;
extra reinf. at lintels s pec ifed,
200mm lintel bearing
Specific openings detailed;
top of openings aligned;
openings in middle of walls;
vert . confinement is rebar in
masonry; RC sills and lintels
with 200mm bearing;
guidance on other openings
not given
Align openings vert. & t ops of
openings horiz.; openings
symmetric in plan; min end
distance > 600mm; install
typical t ie columns if opening
area > 2.5m2; 250mm lintel
bearing
Projections,
Recesses,
Cantile vers
Max projection or recess is
15% of parallel building
length; max c antilever is
1.25m if monolithic with s lab
or 0.5m if only anchored to
bond beam
Max projection or recess i s
1/3 of parallel building length
for I-s haped buildings;
cantilevers to be reinforced
and properly tied to floor
No projections, recesses or
cantilevers
No projections, recesses or
cantilevers shown or
suggested
Max projection or recess i s
25% of parallel building
length; max cantil ever is 1.2m
if monolithic with slab or 0.5m
if only anchored to bond beam
Foundations
No info provided; presumably
design foundation using
Eurocode
No info on foundation design;
continuous foundations
beneath walls; foundation
depth greater than freezing or
shrinkage crack depth
Footing height: 800mm
embedded 500mm in native
soil; footing width: 400mm on
rock, 500mm wide on sandy
clay, 700mm wide on sand;
Plinth: wall width by 300mm
high with 4-3/8" long. rebar
and 1/4" stirrups at 200mm
Footing height: 800mm
embedded 500mm in native
soil; footing width: 500mm
wide on clay and 800mm wide
on sand; Plinth: wall width by
250mm high with 4-10M long.
rebar and 6M stirrups at
200mm
No info provided
Max h/t Ratio 15 20 2400 high/140 wide = 17.1 3000 high/220 wide = 13.6 4000 high/240 wide = 16.7
Masonry Wal l
- Tie Column
Connections
Toothing in figures and 6M
connection rebar at 600mm
embedded 250mm in wall
One-bar tie columns enclosed
in wall: no toothing or
connection bars
Toothing or 2-4.1M connect.
wires every 2nd course
embedded 500mm in wall
Toothing and 2-8M hooked
connection bars at 450mm
embedded 500mm in wall
No toothing in figures and 6M
connection rebar at 600mm
anchored properly
Mech./Ele c.
Slee ves No info No info Details: embedded conduit &
pipes in walls, holes in plinth
Don't embed servic es in walls,
no sleeves in shear walls No info
Max Di stance
Betwe en
Structural
Wal ls
Dependent on PGA: 8m for
High Seismic Zones
7m for 2-story buildings less
than 40* wall thickness (t )
4.5m implied by max slab
span 3.3m implied by drawings 8m
Min W all
Density
Dependent on PGA: 2% Low
PGA, 4% Mid PGA, 5% High
PGA
No info
Dependent on soil: 1% Rock,
1.2% Hard Sandy Clay, 1.4%
Loose Sand or Soft Clay
Only one design provided: wall
density not c ons idered 3%
Tie Col umns
Locate at wall corners, ends,
intersections, openings >
2.5m2 and 4m max s pacing;
150mm x 150mm min s ize;
Long. rebar: 4-8M Low PGA,
4-10M Mid PGA, 4-12M High
PGA; Trans. rebar: 6M at
200mm; anchor rebar
Locate at wall corners,
intersections, openings; 1/2
brick t x 1/2 brick t; Long.
rebar: 1-16M Low Cat egory, 1-
20M High Category (2
stories); No Trans. Rebar
Sample plans suggest at wall
corners, ends, intersect ions,
openings and 4.5m max
spacing; 250mm x wall t min
size; Long. rebar: 4-3/8";
Trans. rebar: 1/4" at 100mm
at ends, 250mm in m iddle;
anchor rebar
Sample plans suggest at wall
corners, ends, intersect ions &
3.3m max s pac ing; wall t x
wall t min s ize; Long. rebar: 4-
10M small c olumns, 8-10M
large columns; Trans. rebar:
6M at 100mm at ends,
250mm middle; anchor rebar
Locate at wall corners,
intersections, openings >
2.5m2; no other info
Confined Ma sonry Guide lines
Item
Table 4.1 – Guideline summary table, part 1
46
City University IAEE Peru Schacher UNESCO
Bond Beam s
Locate at floor and roof levels;
4m max s pacing; 250mm
wide x 150mm high min siz e;
Long. rebar: 4-8M Low PGA,
4-10M Mid PGA, 4-12M High
PGA; Trans. rebar: 6M at
200mm; cast wi th slabs;
anchor rebar
Locate at floor, roof and top of
door elevations; wall t wide x
75mm high min size; Long.
Rebar varies from 2-10M to 4-
16M depending on category
and span between inter. walls;
Trans . rebar: 6M at 150mm
Locate at floor and roof levels;
2.4m max s pacing; 250mm
wide x 200mm high min siz e;
Long. rebar: 4-3/8"; Trans .
rebar: 1/4" at 100mm at ends,
250mm in middle; anchor
rebar; cast with s labs
Locate at floor and roof levels;
3.0m max s pacing; wall t wide
x 220mm high min s ize; Long.
rebar: 4-10M small beams, 6-
10M large beams; Trans.
rebar: 6M at 100mm at ends,
250mm in middle; anchor
rebar; cast with s labs
Locate at floor and roof levels;
4m max spaci ng; 150mm
wide x 150mm high min siz e;
Long. rebar: 4-12M ; Trans.
rebar: 6M at 200mm; cast
with slabs; anchor rebar
Floor an d
Roof Systems
Floors/roofs must work as
horizontal diaphragms;
connections must transfer
lateral forces; RC slabs
recomm to ensure adequate
diaphragms & connections;
light roofs preferred to lessen
seismic loads
Floors/roofs must work as
horizontal diaphragms and be
well connected to wall
Concrete floor/roof slab:
100mm wide x 200mm deep
concrete joists between
lightweight ceiling bricks with
50mm concrete cover slab;
reinf. depends on joist span
and continuity; 4. 5m max
span
Concrete floor/roof s lab: pre-
fab. 220mm deep 8M rebar
trusses with 100mm wide
precast concrete bott om
chord between hollow bri cks
with 60mm RC cover slab;
extra reinf. at bond beams;
rain roofs, shade structures
and roof terraces detailed
Floors/roofs must work as
horizontal diaphragms;
connections must transfer
lateral forces; Timber roofs not
recommended unless highly
skilled carpenters and good
diaphragm and connection
details
Non-
Structural
Eleme nts
Gable end walls: confine with
tie columns and diagonal
bond beams, no mention of
tying top of gable back t o roof;
partition walls: vert. & horiz.
restraints suggested;
chimneys: don't use masonry;
veneers and architectural
details: anchor to s tructure
Gable end walls: confine with
diagonal bond beams or
construct from light m aterials,
no mention of tying top of
gable back to roof; parapets:
anchor to roof structure
Guideline does not encourage
most non-structural elements
so does not detail t hem;
partition walls: hollow clay tile
suggested but no lateral
connection details
Guideline does not encourage
most non-structural elements
so does not detail t hem;
partition walls: s ame as
structural walls but no
confining elements; parapets:
confine with tie columns and
bond beams
Gable end walls: confine with
tie columns and diagonal
bond beams, no mention of
tying top of gable back t o roof;
partition walls: vert. restraints
suggested; parapets: confine
with tie columns and bond
beams
Materia l
Quality
Concrete: Eurocode Grade
C15; mortar: 1 cement, 0.25-
1.25 lime, 2.25-3*(cement +
lime) sand, water to make
workable; masonry units:
solid & non-solid as per
European Standard
Concrete: 1 cement, 2 s and,
4 gravel, 1 wat er; mortar:
various proportions listed;
masonry units: solid with
various min comp. s trength
Concrete: 1 cement, 2 s and,
4 gravel, 1 wat er (different for
footings); mortar: 1 cement, 5
sand, water to make
workable; masonry units:
solid, no other info
Concrete: 1 cement, 2 s and,
4 gravel, 1 water (different for
footings); mortar: 1 cement, 5
sand, water to make
workable; masonry units:
"strong", s olid bricks, no other
info
Concrete: 1 cement, 2 s and,
4 gravel, 1 wat er; mortar: 1
cement, 4 sand, water t o
make workable; masonry
units: s olid bricks t o national
standards
Workma nship
No information on
construction procedures or
inspection
No information on
construction procedures or
inspection
Good workmanship practices
shown step-by-step for all
construction act ivities; no
inspection tips
Good workmanship practices
shown step-by-step for all
construction act ivities; no
inspection tips
No information on
construction procedures or
inspection
Confined Ma sonry Guide lines
Item
Table 4.2 – Guideline summary table, part 2
4.2 Discussion of CM Guidelines
In this section, the guidelines discussed above will be compared to see how they agree, how they
differ, and in what areas individual guidelines lack information altogether. Some items will be
strongly supported by all of the guidelines, whereas for other items, recommendations will differ
from one guideline to the next. For items where the guidelines are at odds, the research
conducted on confined masonry to date will be examined to help determine which guideline’s
suggestions are most appropriate. This examination of available research will be performed in
Section 5.
4.2.1 General Comparison
Referring to Table 4.1 and Table 4.2, the guidelines agree on some items, while on others they
differ. In places where the guidelines concur, the common recommendations made by the
guidelines often address the shortcomings in current confined masonry construction discussed in
Section 3. It should be reiterated that the shortcomings in confined masonry practice identified
in Section 3 do not apply to all CM housing currently being constructed. Rather, these
47
shortcomings represent recurring deficiencies in some lower quality CM construction. When the
individual items in the guidelines are discussed in Section 4.2.2, areas where the suggestions of
the guidelines focus on alleviating observed shortcomings in CM housing will be explicitly
noted.
In places where the guidelines differ, one might attribute the discrepancies to the fact that the
guidelines are based on different codes and references (Table 4.1 lists the bases of the each).
However, building codes are generally similar, making it unlikely that all the differences
between the guidelines (and there are quite a few) are due to each having a different starting
point. Some of these discrepancies are more likely a result of the different approaches the
guidelines take to achieve seismic resistant CM designs. A good way to convey these
differences in approach is to put the guidelines on a spectrum, such as the one in Figure 4.1.
Figure 4.1 – Spectrum of CM guidelines
48
As shown above, the guidelines have either tried to be applicable to many situations, with the
result that they are less comprehensive and more technical, or applicable to a specific situation,
making them more of an all-in-one solution and easier to follow for non-engineers. The different
approaches adopted by the guidelines will often be referred to when the individual items in the
guidelines are analyzed in Section 4.2.2. In fact, many of the discrepancies between the
guidelines will be attributed to the difference in the approaches taken by the guidelines. For
cases where the discrepancies can not be explained by differences in approach, research on
confined masonry will be examined in Section 5 to help resolve the discrepancies.
Before proceeding with comparison of the guidelines on an item by item basis, two more general
topics must be discussed. The first topic relates to the fundamentally different structural
behaviour assumed by the IAEE Guideline as compared to that assumed by the other guidelines.
The second topic pertains to the great deal of similarity between the City University and
UNESCO Guidelines in many respects, and the possible reason behind this similarity.
The City, UNESCO, Peru and Schacher guidelines aim to achieve the force transfer mechanisms
detailed in the Failure Modes Section. In particular, these mechanisms rely on tie columns to
redistribute in-plane shear forces and help strengthen wall intersection connections; as a result,
tie columns must have sufficiently large cross-sections, multiple longitudinal bars, and ample
transverse reinforcing. Bond beams also play a crucial role in the assumed mechanisms,
distributing diaphragm forces along in-plane walls, facilitating good diaphragm-to-wall
connections and providing vertical stiffness to increase the out-of-plane bending resistance of
walls.
In contrast, the IAEE Guideline is more of a partially confined masonry system. Similar to the
other guidelines, bond beams are specified at the roof level to distribute diaphragm forces and
help with diaphragm-to-wall connections. However, these roof bond beams are wide and
shallow, not narrow and deep like the other guidelines, and are therefore not well set up for
restraining masonry walls rotating out-of-plane.
Another unique feature of the IAEE guideline is the continuous bond beam it specifies at top of
opening elevations. This bond beam works in horizontal bending to carry out-of-plane forces
from the wall to intersecting wall locations. In the other guidelines, the bond beams do not work
in horizontal bending because they are only located at diaphragm levels and are in turn supported
in the horizontal direction by the diaphragms. Similarly, the other guidelines do not rely on tie
columns to act in horizontal bending since they assume that only tie columns braced in the
horizontal direction by intersecting walls act as lateral supports for out-of-plane wall bending.
The biggest difference between the IAEE Guideline and other guidelines is the way in which tie
columns are expected to work. Except for thin walls where tie columns are spaced at 1.5m to
help with vertical out-of-plane bending, the IAEE Guideline only stipulates tie columns beside
openings and at wall corners. As well, the tie columns are specified as only half a brick thick,
contain just one longitudinal bar, and have no stirrups. In this capacity, the IAEE tie columns
provide some nominal crack control and resistance to in-plane wall rocking; they do not
distribute in-plane loads or help with connections of intersecting walls like larger tie columns do.
As a result, the smaller tie columns do not “confine” the masonry wall and in-plane force transfer
49
is similar to that in unreinforced masonry structures. From this lack of confinement with respect
to in-plane behaviour stems the presumption that the structural system in the IAEE Guideline is a
“partially” confined masonry system.
Because of the considerable difference in the expected behaviour of the IAEE Guideline system
and the systems specified in the other guidelines, it will be difficult to compare the IAEE
Guideline to these systems in many cases. This point will be noted when the individual items in
the guidelines are analyzed in Section 4.2.2.
Switching now to the City University and UNESCO Guidelines, there is an uncanny similarity
between many of the recommendations made by these two guidelines. The UNESCO Guideline
references a publication dealing with Eurocode 8, which suggests that this guideline may be in
some way be based on the City University Guideline (itself, mainly an interpretation of parts of
Eurocode 8) or a precursor to this guideline. This being the case, the UNESCO Guideline may
have been an attempt at creating a more simplified version of the City University Guideline that
is focused on non-engineered buildings of three stories or less; the UNESCO Guideline will be
evaluated with this in mind.
4.2.2 Comparison of Individual Items
Having taken a broad look at some of the general similarities and differences between the
guidelines, the more critical items of information provided by the guidelines will now be
examined one at a time. As discussed previously, it will be noted if items the guidelines agree on
how to address observed shortcomings in some current CM construction. On items in which the
guidelines differ, it will be stated if the discrepancy can be explained simply by considering the
different approaches adopted by each guideline, or if existing confined masonry research must be
consulted.
Consideration of Seismicity
How seismicity is considered varies from guideline to guideline. The IAEE and City University
Guidelines consider multiple hazard ranges. As a result, the building design can be scaled up or
down to reflect the seismic hazard. The Peru and Schacher Guidelines, on the other hand,
provide only one design/set of design guidelines and therefore consider only one seismic hazard.
Thus for these guidelines, the specified design could be unconservative for seismic hazards
greater than the maximum design hazard and overbuilt for lower seismic hazards. For the
UNESCO Guideline, there is no indication as to what seismic hazard has been considered.
The different ways that seismicity is considered appears to be a function of the approaches taken
by the guidelines: prescriptive (Peru and Schacher) or flexible (City University and IAEE). The
approach selected by each guideline will influence its applicability, however, and is not arbitrary
in this case. For instance, the one-design-fits-all approach utilized by the Peru and Schacher
Guidelines could mean that these guidelines may be unsafe or uneconomical in regions with
earthquake intensities significantly different than the design intensity. Furthermore, it is difficult
to apply the UNESCO Guideline anywhere because there is no indication what it has been
50
designed for. The question of applicability will be a significant concern for any case where one
of the guidelines is used in a region that it was not originally developed for.
Site Selection
All the guidelines agree that it is imperative to locate CM houses away from hazardous sites with
loose soils, unstable slopes and other problems. Houses constructed on sites with the
aforementioned problems may be susceptible to foundation failure, which can lead to the
collapse of the entire building.
Building Configurations
With respect to building configurations, the guidelines all agree that configurations must be
regular and provide similar measures to ensure regularity is achieved. Irregular building
configurations were one of the shortcomings in current CM housing (a recurring design flaw)
noted Section 3.
The guidelines suggest simple building shapes that are primarily rectangular and have maximum
plan aspect ratios (the length of the building compared to the width) between 3 and 4. Plan
symmetry is also emphasized: walls and openings are to be located symmetrically and distributed
evenly in both horizontal directions. Vertical continuity is assured by provisions recommending
that wall locations and thicknesses, as well as opening locations, be the same from story to story.
Lastly, to ensure simple seismic building behaviour, the number of stories is limited to two
stories by the guidelines with a couple exceptions (the City University Guideline allows four
story buildings in low seismic zones and the UNESCO Guideline permits three story structures).
By including all the above requirements, the guidelines help ensure regular building
configurations, thereby facilitating the construction of structures that are less susceptible to
diaphragm failures, diaphragm-to-wall connection failures and in-plane wall failures.
Wall Openings
On the topic of wall openings, the guidelines agree in some ways and differ in others. All the
guidelines require openings to be aligned vertically, located symmetrically in plan and elevation,
and offset a specific distance from other openings and wall ends. These measures are to ensure
favourable behaviour around opening locations. As well, all the guidelines, including IAEE,
suggest or stipulate some kind of confinement at large wall openings. A lack of confinement at
wall openings was noted as a design flaw (refer to paragraph on inadequate confining elements)
and shortcoming of current CM construction in Section 3.
Where the guidelines differ on wall openings is in the confinement they specify and the
situations where confinement is required. Discrepancies exist with respect to what type of sill
and lintel is most suitable, what type of vertical confinement is most appropriate, and at what
point openings require confinement. To see how the guidelines differ in these areas, refer to
Table 4.1. The fact that different approaches were taken by the guidelines does not explain the
above discrepancies. As a result, the confined masonry research available will be examined in
51
Section 5 for guidance as to what methodology is most appropriate at wall openings. Exactly
how the guidelines differ on wall openings will be further discussed at this time.
Building Projections, Recesses and Cantilevers
The guidelines handle building projections, recesses and floor cantilevers in different ways; the
method used is related to the flexibility of the guideline. The Peru and Schacher Guidelines
don’t include these building irregularities in their prescriptive design methodologies, thereby
implying that they are not permitted. The more flexible City University, IAEE and UNESCO
Guidelines provide recommendations on how large projections and recesses can be, as well as
maximum dimensions and anchorage requirement for cantilevers. The recommendations given
by the latter three guidelines are reasonably consistent. All in all, the guidance on projections,
recesses and cantilevers provided by the guidelines looks to be a function of the different
approaches taken by each and is not likely that significant to the overall building behaviour.
Foundation Information
Information on foundation construction for confined masonry housing is explicitly included by
some of the guidelines and not extensively discussed by others. The Peru and Schacher
Guidelines provide extensive information on how foundations should be built; the information
provided by these guidelines is quite similar. The other guidelines do not cover the topic of
foundations in depth, advising the reader to consult references devoted to foundation design.
The discrepancy between the guidelines with respect to foundations is a consequence of the
degree of comprehensiveness adopted by each and will not be explored further.
Maximum Wall Slenderness (h/t ratio)
The guidelines explicitly or implicitly agree on a maximum height to wall thickness (h/t) ratio of
between 15 and 17. Maximum story heights vary considerably from guideline to guideline, but
the guidelines with large story heights also have larger minimum wall thicknesses, resulting in a
consistent slenderness ratio between guidelines. The exception is the Schacher Guideline which
specifies slightly stockier walls. The stockier Schacher walls are probably less a function of
trying to prevent out-of-plane failure (the purpose of limiting h/t ratios), and more a consequence
of selected brick size.
Masonry Wall to Tie Column Connections
There is disagreement among the guidelines on how masonry wall to tie column connections
should be detailed. All the guidelines provide some kind of connection detail: a few suggest that
the masonry wall be “toothed” and interlocked with the tie column; others specify only
horizontal rebar across the masonry wall-tie column interface. The size, spacing and embedment
of horizontal interface rebar also varies between the guidelines. Current confined masonry
research will be consulted in Section 5 to determine the most appropriate masonry wall to tie
column connection. Note that for this item, the IAEE guideline was not considered since its
small tie columns do not allow for the type of wall-tie column connections specified by the other
guidelines.
52
Mechanical and Electrical Sleeves
Inclusion of guidance on mechanical and electrical sleeves is related to the comprehensiveness
targeted by the guidelines. The Peru and Schacher Guidelines include rules on the permitted size
and location of sleeves and the others do not. Since this discrepancy stems from differences in
approach, this topic will not be discussed in any greater detail.
Maximum Distance Between Structural Walls
The guidelines are grouped into two camps with respect to the maximum distance permitted
between structural walls. The Peru and Schacher Guidelines imply quite low maximum
distances (4.5m and 3.3m) on their sample plans. This is likely due to the fact that the structural
walls also support the gravity load on the floors and roof; the maximum distance given is
probably the maximum gravity load span of the floor/roof system specified, not the maximum
wall spacing for earthquake loading.
The other guidelines specify (not imply) much larger maximum distances between structural
walls (7-8m; larger for the City University Guideline in low seismic zones). This may be a result
of specific floor/roof systems not being detailed in these guidelines; in fact, smaller structural
wall spacings may be necessary for gravity loads. Larger maximum distances between structural
walls place higher demands on diaphragms and connections, which may be a concern. For the
most part, however, the discrepancy in maximum distance between structural walls appears to be
a function of different approaches taken by the guidelines, and therefore research will not be
examined for information on this topic.
Minimum Wall Density
Wall density is the plan area of a building occupied by structural walls divided by the total floor
area. This requirement is a simple means to ensure a structure has sufficient strength to resist
expected seismic forces. The guidelines differ in both the magnitude of the wall density they
prescribe and the factors that influence the required wall density. Wall density has a fairly
significant influence on how confined masonry buildings behave; for this reason, confined
masonry research will be consulted in Section 5 for recommendations on wall density. For the
IAEE Guideline system, wall density is not a parameter.
Confining Elements
With respect to confining elements, all the guidelines (except for the IAEE Guideline) recognize
the need to adequately confine masonry walls with tie columns and bond beams (thereby
addressing one of the seismic design flaws noted in Section 3), but differ in the confining
elements they specify. As discussed above, the partially confined system in the IAEE Guideline
doesn’t utilize tie columns to confine walls and will therefore not be considered here.
In the four fully confined masonry guidelines, there is much agreement on confining members,
especially on a more general level. Tie column locations are either explicitly prescribed in
sample plans, or mandatory locations (structural wall ends, corners and intersections) and a
53
maximum spacing dimension (4m – 4.5m) are specified. The problem of poor rebar detailing
identified in Section 3 is dealt with in the guidelines by providing diagrams of proper detailing,
and stressing the need for adequate anchorage of longitudinal reinforcement at the ends of bond
beams and tie columns. Another problem from Section 3, inadequate tie column shear strength,
is implicitly addressed in the guidelines by specifying longitudinal reinforcing, stirrups, and
minimum concrete cross sections, or in the case of the UNESCO Guideline, implying these items
on diagrams.
The differences between the guidelines arise in the specifics they recommend, particularly for
rebar detailing and the parameters affecting confining element shear strength: cross-sectional
area, longitudinal reinforcing and transverse reinforcing (stirrups). The rebar details in some
guidelines call for U-bars at confining element intersections, whereas in other guidelines, hooked
bars are specified, sometimes with additional diagonal bars. With any of these details, it is
possible to achieve proper anchorage of longitudinal reinforcing provided that a sufficient length
of concrete is present between the hook and face of beam in the pullout direction. Placing hooks
and U-bars at the back face of the beam intersection should ensure a sufficient length of concrete
to resist pullout. Thus, the differences between the guidelines on this topic appear to be a result
of differences in approach and will not be investigated further.
For the case of tie column and bond beam properties, the different values suggested by the
guidelines do not appear to be merely a result of the different approaches taken by the guidelines.
The reasons for the discrepancies in the cross-sectional area, longitudinal rebar area and stirrup
spacing specified by the guidelines are not apparent. Considering the impact these parameters
have on confining element shear resistance, a critical factor in the capacity of a confined
masonry wall, the CM research available will be examined in Section 5 to determine appropriate
values for each of the above parameters.
Floor and Roof Systems
The guidelines take two different approaches with respect to floors and roofs. The Peru and
Schacher Guidelines provide fully detailed roof and floor designs, while the others just address
connection and diaphragm requirements. The systems defined in the Peru and Schacher
Guidelines are different (concrete joist-lightweight masonry composite slab vs. concrete-rebar
truss-masonry composite slab), however, and only the Schacher Guideline provides additional
guidance on shade structures, rain roofs and roof terraces. Similar to the case for maximum
distance between structural walls, the differences in the roof and floor recommendations
presented by the guidelines appear to be a function of variation in approach: comprehensiveness
vs. flexibility.
When it comes to roof-to-wall connections, the guidelines all agree on the importance of sound
roof-to-wall connections in preventing out-of-plane damage and failure. Inadequate roof-to-wall
connections were noted as one of the shortcomings in current CM housing in Section 3. The
need to laterally restrain walls at the foundation, roof and all floor levels is stressed in the
guidelines, and bond beams are specified at roof and floor locations to facilitate good
connections. Concrete floors and roofs cast monolithically with the bond beams are suggested as
a way to further ensure adequate wall connections.
54
The guidelines also agree on the need to avoid inadequate floor and roof diaphragms, another
shortcoming of CM construction discussed in Section 3. By requiring adequate horizontal
diaphragms, the guidelines hope to reduce the likelihood of out-of-plane failures and loss of
inertial force transfer. As mentioned above, the guidelines recommend concrete floors and roofs.
These structures typically perform well as horizontal diaphragms. The City University, IAEE
and UNESCO Guidelines permit non-concrete diaphragms but do not fully detail these systems.
Non-Structural Elements
With respect to non-structural elements, the discrepancies between the guidelines are again a
function of the different approaches adopted by each. The Peru and Schacher Guidelines do not
allow such items such as facades and veneers due to their prescriptive nature. The Schacher
Guideline explicitly indicates how non-structural partition walls and parapets should be
constructed, but the Peru Guideline does not detail these elements or warn against the dangers of
constructing such elements out of unreinforced masonry; this item will be touched on again later
in the report. The other guidelines provide some direction on how to support non-structural
elements, although the suggestions given are relatively vague. In most cases, the guidelines
simply say to anchor architectural elements adequately to the structure; little detail is given on
how.
Focusing on inadequately supported masonry gable walls, a shortcoming in current CM housing
identified in Section 3, the guidelines all agree that this problem must be addressed. How to
prevent inadequately supported masonry gable walls, though, varies from guideline to guideline.
The Peru and Schacher Guidelines dispose of the problem by only specifying non-pitched roofs,
thereby avoiding gable end walls. The other guidelines indicate how gables should be confined
with additional bond beams and tie columns to reduce the likelihood out-of-plane failure. These
guidelines do not, however, specifically mention that the tops of gables need to be laterally
supported at the roof level, a critical factor in preventing out-of-plane failure.
The differences in the information provided on non-structural elements does not seem that
significant, and as previously mentioned, appears to be due to differences in approach. As a
result, research on this topic will not be examined and discussed in Section 5.
Material Quality
Material quality is a construction issue that affects resistance to all failure modes, and was a
concern with current CM housing noted in Section 3. The guidelines help promote good material
quality by specifying what material characteristics are acceptable. For mortar and concrete,
either a minimum strength is specified or mix ingredient proportions are provided. For masonry
units, acceptable types of brick are given (usually solid brick). Unfortunately, the guidelines do
not address the testing of materials. Thus, the guidelines provide assistance with respect to
material “design”, but can not and do not help out with material “inspection”.
It should be noted that there is some variation in the brick sizes, material properties and material
mix proportions from guideline to guideline. These variations reflect differences in the building
materials commonly produced in different regions. As long as the materials used are good
55
quality and in line with the national standards of that particular country, the similarity in national
standards of different countries should ensure that the variation in the materials specified by the
different guidelines are not that significant.
Quality of Workmanship
Workmanship quality is another factor strongly affecting resistance to failure modes. This item
was also highlighted as a shortcoming in current CM construction in Section 3. Some of the
guidelines (Peru & Schacher, in particular) address this issue by showing good workmanship
practices and discouraging bad workmanship techniques. Similar to the case for material quality,
however, the guidelines only specify what good workmanship is; they do not encourage or
provide tips for inspections, items that would help ensure that good workmanship takes place.
Guideline as a Design Aid
All the guidelines partially tackle the problem of lack of CM housing regulation noted in Section
3 by acting as design aids for the construction of seismic resistant confined masonry buildings.
The increase in the number and popularity of confined masonry guidelines, including those
discussed here, is ensuring that more and more designers and builders have the tools they need to
make sound confined masonry housing designs.
4.3 Closure
In this section, five different confined masonry design and construction guidelines were
discussed and analyzed. First, the general character and information contained in the City
University, IAEE, Peru, Schacher and UNESCO Guidelines were presented. Next, the
guidelines were compared on a general level, noting the conceptual similarities and differences
between them. Lastly, the individual items in the guidelines were compared. Instances where
the guidelines agree were noted, and special mention was give to cases where this agreement
addresses the shortcomings in current CM housing identified in Section 3. On items where the
guidelines differ, the discrepancies were analyzed and divided up into those that are mainly a
result of the different approaches taken, and those that are more difficult to reconcile. For the
latter, the research conducted to date on confined masonry will be consulted to determine which
guidelines handle certain items best. Research will be sought on the following topics:
1.
Wall Openings
2.
Tie Column and Bond Beam Cross-Sectional Area, Longitudinal Rebar and Transverse
Reinforcing
3.
Masonry Wall-Tie Column Connections
4.
Minimum Wall Density
The application of research to the differences between confined masonry guidelines will be
conducted in the next section.
56
5 RESEARCH
In the previous section, five confined masonry guidelines were compared. Of the topics that the
guidelines disagreed on, four were identified as requiring further investigation: wall openings, tie
columns and bond beams, wall-tie column connections, and wall density. In this section, the
research published on each of these topics will be examined and compared with the
recommendations in the guidelines. This is not intended to be a comprehensive literature survey
of all research on confined masonry, rather a focused review of studies related to the
discrepancies noted above. In addition, future research topics will be proposed that would help
clear up unresolved discrepancies between the guidelines.
5.1 Wall Openings
The subject of wall openings is one area where the guidelines examined disagree. There is
variation amongst the guidelines with respect to sill and lintel details, vertical confining elements
at openings, and maximum unconfined opening size.
Only the Schacher and IAEE Guidelines suggest a confining sill under non-full height wall
openings, a category consisting primarily of windows. The Schacher Guideline calls for a
concrete beam that is one brick course high, is reinforced with a single 8mm diameter bar, and
extends 200mm past the edge of the opening. The IAEE Guideline suggests a sill with a
minimum member height and end bearing length of 75mm; the reinforcing required in the sill is
not clear, however.
All the guidelines provide information on lintels, although only the Peru and Schacher Guideline
deal with reinforcing; the others simply mention minimum widths and end bearing lengths, and
suggest lintels be cast monolithically with the next bond beam up, requiring the designer to look
to other sources for the necessary lintel reinforcing. The Peru Guideline also suggests that the
tops of openings be located at or near the underside of bond beams (i.e. bond beams act as
lintels) and stipulates different amounts of reinforcing depending on the width of the opening.
The Schacher Guideline on the other hand, calls up lintels well below the bond beam level and
specifies the reinforcing required by these lintels.
The research examined on sills and lintels for CM wall openings [Alcocer et al., 2004; Yaiiez et
al., 2004] provides little guidance on how lintels should be constructed. In most cases, the
openings considered in this research are not overly wide and are located at the underside of bond
beams, which serve as the lintel. The research thus suggests that it is probably a good idea to
push openings to the underside of bond beams; reinforcing as recommended by the Peru
Guideline is probably suitable for such situations. If openings can not be moved to the underside
of bond beam elevation, the detail suggested by the Schacher Guideline may be more
appropriate.
With respect to sills, both the sources consulted employ reinforced sills below window openings.
This would suggest that a reinforced sill may be appropriate for all CM buildings. In the Yaiiez
57
et al. [2004] research project, the sills are not reinforced concrete but consist rather of 2 – 4mm
diameter rebars embedded in the horizontal mortar joint one brick course below the bottom of
each opening. Thus, the concrete sill detail in the Schacher Guideline may be excessive, but in
lieu of conclusive evidence proving such, it is recommended to include concrete sills on confined
masonry buildings. If nothing else, surrounding the sill reinforcing with concrete should help
improve resistance to corrosion, a problem observed in some CM housing with wire reinforcing
placed in mortar beds.
The research found on the topic of sills and lintels does not specifically investigate the impact
these elements have on CM wall behaviour. What the research has determined is that sill and
lintels help alleviate cracking at the corners of wall openings. Future research investigating what
types of sill and lintel details are most appropriate (reinforced concrete or just rebar in the
masonry wall, amount of reinforcement) is required.
When it comes to vertical confining elements at CM wall openings, most of the guidelines
suggest that their typical tie columns be used, one-bar tie columns for the IAEE Guideline and
four-bar tie columns for the City University, Peru and UNESCO Guidelines. The Schacher
Guideline, however, calls for just 1 – 8mm diameter vertical rebar to be embedded in a
continuous vertical mortar joint in the masonry wall on both sides of the opening. With respect
to the length of vertical confining members besides openings, the IAEE and Schacher Guidelines
only require members to extend for the height of the opening, whereas the others call for tie
columns that are the full height of the wall. Thus, there is a great deal of inconsistency from
guideline to guideline when it comes to specifying vertical confined elements at wall openings.
The experimental testing done on CM walls with openings generally suggests standard four-bar
tie columns be used as vertical confining elements at wall openings. There is some research,
though, that has utilized reinforcing grouted in hollow cores [Yaiiez et al., 2004; Andaya et al.,
2004] or single rebar tie columns with spiral transverse reinforcing [Yoshimura et al., 2004a] as
vertical confining elements. These research projects would seem to suggest that the demands on
tie columns at openings are considerably less than elsewhere. However, other research [Alcocer
et al., 2004; Ishibashi et al., 1992] recommends that transverse reinforcing be installed at close
spacing in tie columns near the corners of openings. Thus, it would appear that tie columns at
wall openings require significant shear resistance. Research explicitly comparing the effects of
different types of vertical confining elements at openings is required. Until such research is
conducted, however, it is recommended to install four-bar tie columns as vertical confining
elements with extra ties near opening corners.
There is considerable disagreement between the guidelines on what the maximum size of
unconfined openings should be. The City University and UNESCO Guidelines suggest openings
up to 2.5m
2
in area do not require confinement. The Peru Guideline recommends tie columns at
openings (presumably all openings) but does not mandate them, whereas the Schacher Guideline
specifies vertical embedded rebar confining elements at all openings. The IAEE Guideline has
complex criteria on when confinement is needed.
The variance in the maximum unconfined opening size found in the guidelines is reflected in the
research completed on this topic. Alcocer et al. [2004] proposes that all wall openings be
58
confined. Yaiiez et al. [2004] suggests that openings smaller than 11% of the wall area don’t
require confinement, while Radovanovic [1998] indicates that confinement is only necessary for
openings larger than 1.5m
2
. The 2.5m
2
maximum in the City University and UNESCO
Guidelines comes from work done by Tomazevic and Klemence [1997a].
To clear up the discrepancies in the maximum opening size suggested by different sources,
research is required that focuses directly on the size at which unconfined openings are no longer
a concern and the types of elements that adequately confine different sizes of openings. If
elements less expensive than four-bar tie columns (e.g. one-bar or two-bar tie columns, rebar
embedded in masonry) can be found that adequately confine smaller openings, or perhaps even
larger openings, this would likely make it economically feasible to confine all openings. Some
research [Liu and Wang, 2000] has already suggested that this might be a feasible option. In the
meantime, it is recommended to confine all openings except small penetrations for mechanical
and electrical work (600mm x 600mm maximum). Confinement should be attained at the very
least with embedded rebar and preferably with typical four-bar tie columns.
5.2 Tie Columns and Bond Beams
Moving from confining elements around openings to the general case of confining members,
there are considerable discrepancies between the tie columns and bond beams specified by each
of the guidelines. Specifically, the guidelines differ on minimum cross-sectional area, required
longitudinal reinforcing, and appropriate quantities of transverse reinforcement. All three of
these factors influence the shear capacity of confining members, a value that has a substantial
influence on the strength and, more importantly, the deformation capacity of CM walls under in-
plane loading.
As discussed in the Guideline Section of this report, the IAEE Guideline assumes different
structural behaviour than the other guidelines and consequently has different types of tie columns
and bond beams. The “conventional” CM guidelines will be compared with each other first and
the IAEE Guideline will be addressed afterwards.
The minimum tie column and bond beam sizes specified by the guidelines relate to the width of
masonry units each assumes. The City University Guideline says to match the wall width but
puts a lower bound of 150mm x 150mm on tie column cross-sections and 250mm wide x 150mm
high on bond beam cross-sections. The Peru Guideline specifies minimum tie column
dimensions of 250mm x the wall width (140mm minimum on the plans at the back of the
guideline) and minimum bond beam dimensions of 250mm x the slab thickness of 200mm.
Similarly, the Schacher Guideline sets minimums of 220mm x the wall width (220mm
minimum) for tie columns and the wall width x the slab height (220mm) for bond beams. The
UNESCO Guideline does not address tie column sizes but stipulates a minimum bond beam
cross section of 150mm x 150mm.
Considering the minimums stated in the guidelines appear to be based on masonry unit sizes and
rebar placement limitations (it’s difficult to fit four longitudinal bars in a member smaller than
150mm x 150mm), research justifying the minimum cross-sectional areas necessary for adequate
59
structural performance would be very helpful. Unfortunately, only one research study was found
commenting on minimum member size [San Bartolome et al., 2004] and this study only
addresses tie columns. The San Bartolome study suggests that the area of a tie column should be
no less than 150mm x the masonry wall thickness. This study also mentions, however, that
column size may have to be increased from this minimum to provide adequate resistance to shear
and concrete crushing. The minimum from research coincides quite closely with the City
University Guideline. From this perspective, the minimum cross-sections called up in the Peru
and Schacher Guidelines are likely reasonable. Additional testing on the effect of cross-sectional
behaviour on tie column and bond beam performance would be worthwhile.
How tie column and bond beam longitudinal reinforcement is specified in the guidelines varies
in a few ways. The Peru and UNESCO Guidelines specify one reinforcement scheme regardless
of seismicity or member cross-section. The Peru Guideline stipulates 4 – 3/8” diameter bars in
all tie columns and bond beams; the UNESCO Guideline recommends 4 – 12mm diameter bars
in bond beams but does not address tie column reinforcing. The Schacher Guideline gives
different reinforcing requirements depending on member size: small (220mm wide) tie columns
and bond beams require 4 – 10mm diameter bars, while large (340mm wide) bond beams get 6 –
10mm diameter bars and large (340mm x 340mm) tie columns receive 8 – 10mm diameter bars.
In the City University Guideline, member cross-sectional area does not influence the specified
reinforcing, but as the seismic demand goes up, the required size of bar increases. For low,
medium and high levels of seismic hazard, the longitudinal reinforcing for both tie columns and
bond beams consists of 4 – 8mm diameter bars, 4 – 10mm diameter bars and 4 – 12mm diameter
bars, respectively.
There are many research programs that have considered the effects of tie column and bond beam
reinforcing on the seismic performance of CM walls. Unfortunately, the conclusions drawn from
these studies are overwhelmingly qualitative in nature. Most research says that installing more
longitudinal reinforcing increases the strength and deformation capacity of CM walls. Trying to
find a minimum or optimum amount of longitudinal rebar from the research, though, is a very
difficult task. San Bartolome et al. [2004] suggests that tie column longitudinal reinforcing
should not be less than 4 – 8mm diameter bars and San Bartolome et al. [1990] indicates that to
avoid bar rupture four longitudinal bars are required in tie columns. Finding explicit longitudinal
reinforcing recommendations beyond these research studies is very difficult indeed.
Other research has been conducted, however, that may help in selecting appropriate levels of
longitudinal reinforcing in tie columns and bond beams. Based on test specimens designed in
accordance with Eurocode, the basis for both the UNESCO and City University Guidelines,
Tomazevic and Klemence [1997a] concluded that CM provisions in Eurocode were quite
conservative. This finding would tend to support the suggestion that 4 – 8mm diameter bars
might be adequate. For the time being, the longitudinal reinforcing requirement of 4 – 10mm
diameter bars or 4 – 3/8” diameter bars called up by some of the guidelines is probably a
reasonable value. Research confirming whether or not this reinforcing level can be dropped to 4
– 8mm diameter bars would be useful.
Switching from longitudinal reinforcement to transverse reinforcement, there are two basic
strategies utilized by the guidelines. The City University and UNESCO Guidelines call for
60
uniform spacing of transverse reinforcement in both tie columns and bond beams: 6mm diameter
hoops at 200mm spacing. The Peru and Schacher Guidelines on the other hand, specifically call
for closer spacing of transverse reinforcement at the ends of tie columns and bond beams
(100mm spacing) and wider spacing in the middle of members (250mm spacing). The transverse
reinforcement size stipulated by the latter guidelines is 6mm diameter, although the Peru
Guideline does call up ½” diameter stirrups in some places for bond beams. This inconsistency
is a misprint: ¼’ diameter stirrups are intended to be used everywhere in the Peru Guideline.
The research on transverse rebar in confining members is quite conclusive with respect to tie and
stirrup spacing. Many studies [Alguilar et al., 1996; Irimies, 2002; Zabala et al., 2004;
Yoshimura et al., 2004a; Astoroza and Schmidt, 2004] have concluded that closer hoop spacing
at the ends of members is essential. Exactly how close is close enough in not generally
specified. Further research pinning down appropriate hoop spacing at the ends of confining
members is still required. In the meantime, the spacing schemes proposed by the Peru and
Schacher Guidelines are likely quite suitable.
Having examined research pertinent to the discrepancies between the City University, Peru,
Schacher and UNESCO Guidelines with respect to tie columns and bond beams, additional
information will now be sought comparing these “conventional” CM systems with the one-bar
tie column system specified in the IAEE Guideline. Several research programs have looked at
one-bar tie columns [Yoshimura et al., 1995 and 1996], although the concrete and rebar area of
the one-bar tie columns tested have been similar to four-bar tie columns rather than the smaller
elements typical of the IAEE Guideline. These studies basically concluded that four-bar tie
columns perform better than one-bar systems of the same size, implying that the performance of
the smaller IAEE tie columns would be significantly poorer than conventional four-bar tie
columns.
The conclusions of the above research do not, however, rule out the possibility that there may be
a use for one-bar tie columns in confined masonry. Several research studies [Tomazevic and
Klemence, 1997a; Tomazevic et al., 2004; Casabonne, 2000] have noted that damage in CM
structures is overwhelmingly concentrated in the first story. Others have gone further [San
Bartolome et al., 2004; Radovanovic, 1998; Meli, 1991], suggesting that reinforcing in tie
columns away from critical locations (wall intersections and corners, first story) could be
relaxed. Research into scaled down tie columns in the upper storeys, perhaps containing only
one or two bars, could lead to CM housing designs that are more economical. Such research
could be completed in tandem with work investigating vertical confining members at openings, a
possible future research topic outlined in Section 5.1.
5.3 Masonry Wall-Tie Column Connections
The seismic performance of CM walls is enhanced by good connections between masonry wall
panels and adjoining concrete tie columns. There are two types of wall-tie column connections
in use: interlocking of the masonry wall and the tie column, and connection rebar across the
concrete-masonry interface. Sometimes these techniques are used in tandem. Figure 2.11 shows
how a masonry wall and tie column can be interlocked. For walls with connection rebar, the
61
rebar is placed in the horizontal bed joint near the end of the masonry wall and extended into the
tie column.
Creating an interlock by “toothing” the masonry next to tie columns is specified in the Peru and
Schacher Guidelines. This practice is implied by some illustrations in the City University but not
explicitly called up. The Schacher Guideline also mandates connection rebar between tie
columns and masonry wall panels; the Peru Guideline says to do either one or the other. The
UNESCO Guideline does not address wall-tie column connections.
The City University Guideline specifies 6mm diameter connection rebar at the wall-column
interface, vertically spaced at 600mm and extending 250mm in masonry wall mortar bed. The
Peru Guideline calls for 2 - #8 (4.1mm diameter) wires every second brick course embedded
500mm into the masonry wall. The Schacher Guideline requires 2 – 8mm diameter hooked bars
every 450mm vertically that extend 500mm into the masonry wall.
Most of the research on wall-tie column connections concentrates on connection rebar. Research
on connection rebar [Yoshimura et al., 2004a; Irimies, 2002; Liu and Wang, 2000] indicates that
providing reinforcing across the masonry-concrete interface is beneficial. Specifics on
connection rebar details are more difficult to come by. The research generally suggests 6mm
diameter bars in a U-shape configuration, with the open end of the U being in the masonry wall.
Liu and Wang indicate that connection rebar should be extended 1000mm into the masonry wall;
this 1000mm embedment, however, is likely impractical and not in line with common practice.
There is little guidance in the research on what vertical spacing connection bars should be
installed at.
One source was located [UNIDO/UNDP, 1984] that investigated the interlocking of tie columns
and walls. Interlocking connections were found to be beneficial to wall performance. A limited
study was performed by San Bartolomé and Ordóñez (2005) that compared CM walls with
interlocking columns and with those that use connection rebar. The study suggests that the two
types of walls exhibit similar lateral stiffness and strength. Walls with wire reinforcement were
found to provide larger deformation capacity. Both connection details were observed to result in
CM walls with good lateral performance, however, and the study does not conclusively
recommend that wire reinforcement be used rather than interlocked columns.
Additional research focused specifically on different tie column-masonry wall connection details
is required to evaluate the most effective connection scheme, but it is clear that some form of
connection should be provided. Until such research is conducted, it is advisable to interlock
walls with tie columns (since the amount of extra work and cost is minimal) or provide some
form of connection rebar. The connection rebar suggested in each of the guidelines appears to be
adequate. As suggested by the above research, it would be appropriate to bend connection bars
into U-shapes at wall end locations.
62
5.4 Minimum Wall Density
The guidelines examined differ in how they specify wall density in two ways: the magnitude of
wall density stipulated and the factors influencing required wall density. The IAEE Guideline
does not address wall density, but the other guidelines all specify a minimum ratio of plan wall
area to contributing floor area. This ratio must be satisfied in both principle building directions,
and only walls providing in-plane resistance in a given building direction (walls parallel to the
chosen building axis) are considered in the calculation of plan wall area. Contributing floor area
consists of the area of floor at the level the required wall density is being computed at, plus the
floor area of all levels above. In a two story building, the contributing floor area at the second
level would be the area of the floor on the second level; the contributing floor area at the ground
level would be the floor area on both the first and second levels.
The Peru Guideline gives different minimum wall density values depending on the soil type:
1.0% for hard soils (rock and gravel), 1.2% for intermediate soils (hard clayish sand) and 1.4%
for soft soils (loose sand, soft clay); how soils are classified is covered in the guideline. The City
University Guideline does not consider soil type, but mandates different wall densities depending
on seismic hazard. The wall densities specified by the City University Guideline are
considerably higher than the Peru Guideline. For regions with an expected Peak Ground
Acceleration (PGA) less than 0.2g (g is the acceleration due to gravity), the required wall density
is 2.0%. For areas with a PGA between 0.2g and 0.3g, the minimum wall density is 4% and for
places with a PGA greater than 0.3g, the required wall density is 5%. The UNESCO Guideline
specifies a single wall density recommendation of 3%. The Schacher Guideline provides a
complete housing design rather than design values. Back-calculating wall density from this
design yields a value of 2.8%; note that this wall density is for the first floor of building designed
to be extended to two levels.
Research conducted on the topic of wall density in CM buildings suggests that required wall
density is dependent on seismicity, soil conditions and building height [San Bartolome et al.,
2004]. The guidelines already implicitly consider building height by requiring that the area of
both floors be used in calculating the plan wall area for the first floor of a two story building.
Soil conditions and seismicity are not considered together in any one guideline. This is one area
where the guidelines must be improved if they are going to be applied in locations that do not
have the specific site characteristics that each guideline was derived for.
Some of the research examined [Moroni et al., 2004; Moroni et al., 2000] prefers the use of the
parameter of wall density divided by the building weight in assessing the seismic resistance of
CM structures. This parameter is cumbersome for design purposes (since building mass must be
estimated), so sources focused on just wall density were explored more thoroughly. Meli [1991]
suggests a minimum wall density of 0.65% for one-story structures and 2.15% for two-story
structures. These values compare quite well with those from the Peru and Schacher Guidelines.
The Eurocode-based guidelines (City University and UNESCO), though, call for much higher
wall densities. Some research has been done on the wall densities specified in Eurocode 8
[Tomazevic and Klemence, 1997a] and concluded that these requirements were conservative and
could be relaxed. Thus, the wall densities specified in the Peru Guideline appear to be the most
appropriate. More research is required, however, to determine how wall density should be
63
adjusted for different floor systems (e.g. wood versus concrete), varying levels of seismicity, and
different types of soil conditions.
5.5 Closure
In this section, confined masonry research was used to evaluate the discrepancies between five
CM guidelines with respect to wall openings, tie columns and bond beams, wall-tie column
connections, and wall density. In some cases, the research examined provided some excellent
insight into which guidelines most appropriately address a particular issue. In many other
instances, though, current CM research does not conclusive resolve the conflicting
recommendations given by the different guidelines. Based on the available research, the
discrepancies between the guidelines were resolved as follows:
Wall Openings
-
Align openings with the underside of bond beams, thereby making lintels monolithic with
the bond beam; this is the recommendation in the Peru Guideline.
-
Provide reinforced concrete sills at non-full-height openings as per the Schacher
Guideline.
-
Provide tie columns at wall openings reinforced with four longitudinal bars and ties
spaced closer at opening corners, similar to the recommendations in the Peru and
Schacher Guidelines.
-
Confine all openings larger than 600mm x 600mm.
Tie Columns and Bond Beams:
-
Utilize the minimum confining element cross-sectional areas in either the Peru Guideline
or Schacher Guideline.
-
Provide 4-3/8” or 4-10mm diameter longitudinal bars in confining elements as per the
Peru and Schacher Guidelines.
-
Space ties closer at the ends of confining elements as recommended in the Peru and
Schacher Guidelines.
Masonry Wall-Tie Column Connections
-
Either interlock tie columns and masonry walls or provide connection rebar in accordance
with the recommendations of the guidelines.
-
If connection rebar is installed, ensure U-shaped reinforcing is installed at wall end
locations.
Minimum Wall Density
-
Calculate required wall density as per the Peru Guideline.
As alluded to above, in many cases, the need for future research on wall openings, tie columns
and bond beams, masonry wall-tie column connections, and wall density was identified. The
specific topics mentioned in this section requiring further investigation include:
64
Wall Openings
-
The most appropriate lintel and sill details for CM walls.
-
The most suitable longitudinal reinforcing and tie spacing for tie columns bordering wall
openings.
-
The maximum unconfined opening size for which wall behaviour is not significantly
degraded.
-
Smaller tie columns at medium and possibly large wall openings.
Tie Columns and Bond Beams:
-
The influence of confining element cross-sectional area on wall performance.
-
The adequacy of 4-8mm diameter bars as longitudinal reinforcing in confining elements.
-
Appropriate tie spacing at the ends and middle of confining elements.
-
The suitability of smaller tie columns at less critical locations such as upper storeys.
Masonry Wall-Tie Column Connections
-
The effectiveness of interlocking tie columns and masonry walls vs. the effectiveness of
providing connection rebar between these elements.
-
Appropriate connection rebar diameter, embedment and spacing.
Minimum Wall Density
-
Required wall density for different seismicities, soil conditions and floor/roof systems.
65
6 EVALUATION OF GUIDELINES & RECOMMENDATIONS
Having examined in previous parts of the report the subject of confined masonry in general, and
CM guidelines in particular, this section will focus on:
1.
Which CM guidelines have the potential to be generally applicable
2.
Improvements to promising CM guidelines that would make them more generally applicable
3.
The research necessary to optimize generally applicable CM guidelines
Each of these topics will now be discussed, drawing heavily upon material presented earlier in
the report.
6.1 Potential of CM Guidelines for General Applicability
To identify which CM guidelines have the potential to be applicable in many different regions,
the criteria of a generally applicable guideline must first be established. The criteria that will be
used to evaluate the applicability of the five CM guidelines examined are completeness and
comprehensiveness, style, consistency with research, flexibility with respect to local building
materials and housing styles, and flexibility with respect to seismic hazard.
The first two criteria listed above consider the ease with which guidelines can be implemented
by those who could have the greatest influence on the safety of CM houses: homebuilders. The
third criterion evaluates the relative accuracy of the guidelines with respect to the research
conducted on confined masonry. The last two criteria consider the transferability of the
information in the guidelines: the ability to take a guideline developed for one location and apply
it elsewhere. None of the CM guidelines examined fulfill all of the five criteria completely, but
some satisfy the criteria better than others.
For a guideline to be considered complete and comprehensive, it must contain all the information
needed to build a confined masonry house: site selection and preparation, foundations, confined
masonry walls, floor and roof systems, stairs, material mix proportions. Sample plans of houses
designed for certain common conditions are also helpful in that they provide simple examples
from which custom designs can be generated.
Guidelines written in a non-technical style are generally accessible to a larger audience and are
therefore more widely applicable. Non-technical style relates not only to the language used, but
also to the design method outlined; design from simple prescriptive calculations and tables is
preferred to engineering calculations. In addition, if step-by-step construction procedures are
contained in a guideline, this makes the guideline much more attractive to homebuilders, a major
target audience of CM guidelines.
If a guideline is consistent with research on confined masonry, this provides some confidence
that buildings designed according to the guideline will behave satisfactorily in an earthquake.
Since CM guidelines are a relatively new development, most houses built according to guidelines
66
have not experienced significant earthquakes. Thus, research is the best “measuring stick” for
CM guidelines at the moment and adherence with the conclusions of this research gives the
greatest assurance that a CM house will behave how we expect it to.
Building materials and housing styles can differ drastically from one region to another. If a
guideline is to be widely applicable, it must have the flexibility to incorporate varying wall
heights and masonry unit thicknesses (slenderness ratios), opening sizes and locations, floor and
roof systems, masonry unit properties including strength, and wall layouts. It may be difficult to
consider all possible values for these variables, but if limits are provided, this will enable
interested parties to at least determine how closely the guideline suits local conditions.
The seismic demand on a CM structure is dependent on the seismicity of the region and the soil
conditions at the building site. The level of seismic demand on a CM building dictates what the
size and composition of the structural elements must be. It is uneconomical to provide one
design for the worst possible seismic hazard and prescribe this design for all situations. Thus, it
is essential for a generally applicable CM guideline to be flexible with respect to seismicity and
soil conditions.
Having elaborated on the criteria required for a CM guideline to be generally applicable, each of
the five CM guidelines will now be evaluated for its potential for use in a wide range of regions.
IAEE Guideline
The IAEE Guideline is weak when it comes to comprehensiveness. It provides some
information on mix proportions and wall details, but does not address things like foundation,
floor and roof systems, and provides nothing in the way of sample designs or plans. The IAEE
Guideline also utilizes a rather technical style and it is somewhat difficult to apply the design
method it proposes. The guideline provides very little guidance on the actual steps in
construction, reducing its attractiveness for the homebuilder demographic.
On the positive side, the IAEE Guideline does allow some flexibility with respect to wall layouts
and wall height-to-thickness ratios. Flexibility with respect to seismic hazard is also
incorporated by considering different levels of seismicity. The effects of soil conditions,
however, are not taken into account.
Likely the most significant factor affecting the potential of the IAEE Guideline to be applied in
other regions, though, is the fact that the partially confined masonry system the guideline
prescribes has not been verified by extensive research. This lack of research verification makes
it difficult to assess the expected seismic behaviour of the IAEE system. This lack of
understanding of the IAEE system behaviour, combined with the non-comprehensive and
technical nature of the guideline, prevent the IAEE Guideline from being generally applicable.
UNESCO Guideline
As mentioned previously in this report, the UNESCO Guideline is severely lacking in much of
the information required to construct a confined masonry house. The guideline does not address
67
site preparation, foundations, floor/roof systems or stairs. In addition, the guideline is lacking in
terms of information on tie columns, one of the fundamental components of the confined
masonry system. The UNESCO Guideline does not give step-by-step construction instructions
and contains no information on what seismicity and soil conditions its recommendations are
good for. Thus, there is no way to determine for what level of seismic hazard the guideline can
be used.
The UNESCO Guideline does have some flexibility with respect to building materials and
housing styles in that it implies or specifies recommendations for wall slenderness, openings and
wall layout. In contrast to the IAEE Guideline, it details a typical confined masonry system.
The wall density and confining element stirrup spacing specified by the guideline, however, are
not consistent with confined masonry research.
In conclusion, the UNESCO Guideline is not appropriate for general applicability mainly
because it is incomplete, but also because it does not address varying seismic hazard and is
inconsistent with CM research.
City University Guideline
The strength of the City University Guideline is its flexibility. The guideline provides extensive
information on maximum wall height to thickness ratios, openings, wall layouts and acceptable
masonry units, making it capable of accommodating different building materials and housing
styles. The guideline also considers the effect of varying levels of seismicity by specifying
different wall density and longitudinal reinforcement values depending on the expected Peak
Ground Acceleration (PGA) at a location. On the downside with respect to flexibility, the
guideline does not take into account the effects of site soil conditions on seismic demand.
In addition to its strengths in terms of flexibility, the City University Guideline also has a
number of weaknesses. To start, the guideline is not very comprehensive. It discusses confined
masonry walls in detail but does not provide the information required to design foundations,
floors, roofs, and stairs. As well, it gives required concrete strengths rather than mix proportions
and does not contain any sample plans of CM houses designed using the guideline. For the
information it lacks, the City University Guideline does provide references to other documents
containing this information. This setup is not very convenient, though, in that an array of
different standards must be consulted rather than one source of information.
The style of the City University Guideline is very technical and does not provide any guidance
on the step-by-step tasks required for the construction of a CM house. This style works well for
the apparent target audience of the guideline, engineers, but makes it difficult for people from
non-engineering backgrounds to use the guideline. The City University Guideline reads much
like a code and as a result faces many of the same obstacles that prevent codes from attaining
wide acceptance in the CM housing industry.
In the same way as the UNESCO Guideline, the City University Guideline diverges from the
conclusions of CM research on a couple of key points. The wall densities prescribed by the
guideline are two to five times higher than the values recommended by some research studies
68
and other guidelines, suggesting that the wall densities in the City University Guideline are
excessively conservative. With respect to transverse reinforcing in confining members, the
guideline calls for uniform spacing of stirrups and does not, therefore, adhere to the
recommendation of many research projects that smaller stirrup spacing be used at the ends of
confining members.
Thus, despite the great deal of flexibility inherent in the City University Guideline, its lack of
comprehensiveness, technical style and inconsistency with CM research make the guideline a
poor candidate for general application.
Schacher Guideline
The Schacher Guideline is the polar opposite to the City University Guideline. Its strengths are
its comprehensiveness and easy-to-interpret style, and its weakness is its inflexibility.
Of all the guidelines examined, the Schacher Guideline is the most comprehensive. It provides
recommendations on site selection and preparation, foundations, CM walls, roof and floor
systems, stairs and material mix proportions. The guideline specifies a complete design of a CM
house with drawings and plans, and is detailed to the point that it contains rebar lists for
submission to a rebar supplier as an order for all the reinforcing necessary for the house. In
addition, the comprehensive information in the Schacher Guideline is presented in a non-
technical, easy-to-follow style, with illustrations and instructions for every step in the
construction process.
The Schacher Guideline prescribes a wall density in line with CM research and calls for closer
stirrup spacing at the ends of tie columns and bond beams as suggested by numerous research
studies. The guideline does, however, diverge from the research with respect to the vertical
confinement it specifies at wall openings (rebar embedded between masonry blocks at openings
rather typical tie columns), which is a cause for some concern.
In terms of flexibility with respect to seismic hazard, the Schacher Guideline states the maximum
seismicity (PGA) it is designed for, but does not provide any provisions for adapting its house
design for lower or higher seismicities. As well, except for the case of foundation width, the
guideline does not consider the effect of different soil conditions.
The Schacher Guideline is even less flexible with respect to the possibility of tailoring its design
for different building materials and housing styles. Suggestions on how the housing design
provided could be altered for different masonry unit thicknesses and properties, floor and roof
systems, openings and wall layouts are not presented.
The Schacher Guideline is a very good document and application of this guideline is highly
recommended in regions where the housing style, seismicity and available masonry unit sizes are
similar to those prescribed (PGA < 0.35g, 220mm long x 105mm wide masonry units). That
being said, the inflexibility of the Schacher Guideline makes it difficult to modify the guideline
for different situations, thereby making it difficult for this guideline to be considered generally
applicable.
69
Peru Guideline
The Peru guideline exhibits the comprehensiveness and easily interpreted style of the Schacher
Guideline but has added flexibility with respect to items such as building layout. Thus, of the
five CM guidelines examined in this report, the Peru Guideline has the greatest potential for
general applicability.
The Peru Guideline details how to select an appropriate building site, batch mortar and concrete
mixtures, and construct foundations, confined masonry walls, suspended floors, roofs and stairs.
Each of these topics is explained using simple, straightforward language and lots of diagrams
and illustrations. To provide examples of what a properly designed CM house looks like, a
number of sample plans are included at the end of the guideline.
The Peru Guideline is generally consistent with CM research, in particular with respect to wall
density and stirrup spacing in confining elements. The recommendations in the guideline at
opening locations could be improved by incorporating the findings of some research programs;
this topic will be examined in Section 6.2.
As mentioned above, there is some flexibility in the Peru Guideline, thereby enabling the
possibility of different building materials and housing styles being incorporated into the
guideline. The guideline allows the designer considerable freedom in selecting the wall layout,
opening sizes and opening locations. The guideline does not, however, give explicit guidance on
how to handle different masonry unit properties and thicknesses, wall heights, and floor/roof
systems. Improvements to the Peru Guideline that would help alleviate these concerns will be
discussed in Sections 6.2 and 6.3.
The Peru Guideline appears to consider the effects of soil conditions on seismic demand by
specifying different wall densities depending on the soil type. The guideline does not, however,
look at the issue of seismicity. Only one set of design rules is provided and the seismicity for
which these rules apply is not stated. This issue will also be discussed in Section 6.2.
In summary, the Peru Guideline is unique in its ability to incorporate both comprehensiveness
and flexibility. Because of this, the Peru Guideline is better able to satisfy the criteria established
to evaluate general CM guideline applicability than the other guidelines examined. In its current
form, the applicability of the Peru Guideline to other regions is still quite limited. The ways in
which this guideline can be made more generally applicable, is the topic of the next section.
6.2 Peru Guideline - Suggested Improvements
Overall, the Peru Guideline is an excellent resource for the design and construction of confined
masonry housing. There are some areas, however, where the guideline is currently lacking and
could easily be improved. In this section, a number of suggested improvements to the Peru
Guideline will be made. These suggestions will be divided into those that focus on improving
the seismic performance of the CM system in the Peru Guideline, and those that look to enhance
the general applicability of the guideline itself.
70
6.2.1 Suggested Changes for Improved Performance
There are several ways in which the Peru Guideline could be altered to further improve the
seismic performance of houses constructed using this guideline. These improvements have been
derived from CM research and comparison with other CM guidelines. The improvements to the
performance of Peru Guideline system that this report suggests address non-structural elements,
masonry wall-tie column connections, wall openings and construction inspection.
Non-Structural Elements
As discussed previously, the Peru Guideline does not provide specific guidance on non-structural
elements. That being the case, the lack of explicit warnings against installing some non-
structural elements may incorrectly lead some to conclude that it is safe to include them. This is
not the case: items such as URM chimneys and masonry gable end walls should not be installed
on CM buildings.
Similarly, heavy clay roof tiles and masonry parapets should be avoided if possible. If heavy
roof tiling is desired for protection against roof damage in strong winds or parapets are deemed
necessary to avoid spread of fire to neighbouring buildings, these elements must be securely
fastened to the CM building structure. For the case of parapets, it is preferable that these
elements be constructed from lightweight materials. If masonry parapets are deemed necessary,
they should be properly confined and their height minimized.
The Peru Guideline makes reference to clay tile partition walls but does not indicate the
importance of laterally supporting these elements. If not properly restrained, clay tile partition
walls represent a significant life safety hazard to the building occupants. It is important that
designers/builders of CM houses be made aware of this and be provided with suggestions on
how to mitigate this risk. Partition walls should be constructed to the underside of the ceiling
and laterally supported to the diaphragm (i.e. floor or roof system). Figure 6.1 illustrates a
possible detail for laterally supporting a masonry partition wall at the underside of a roof or floor
Wood 2"x4"
Nailed to
Concrete Slab
Wood 2"x2"
Nailed to 2"x4"
Masonry Partition Wall
Concrete Floor or Roof Slab
Figure 6.1 – Possible top of partition wall lateral support connection
71
slab. Another acceptable detail would be wire mesh secured to the slab and wall with screws and
plaster. Constructing partition walls of lightweight materials or converting masonry partition
walls to structural CM walls further reduces the seismic risk of partition walls.
Lastly, the Peru Guideline provides details for embedding mechanical and electrical piping in
masonry walls. If at all possible, it is preferable to place these services outside of the building
structure rather than embedded in the walls. This should not only improve seismic performance,
but help out in terms of constructability and project cost.
Masonry Wall-Tie Column Connections
The Peru Guideline currently calls for either wire reinforcing to be placed through the tie column
into the mortar bed joints of the adjoining masonry wall or tie columns to be interlocked with the
masonry wall. Either of these measures should ensure strong masonry wall-tie column interlock,
which helps improve resistance to in-plane, horizontal out-of-plane and connection failure
modes.
For the case of toothing, it is essential that care be taken during the casting of tie columns so that
all teeth are completely filled with concrete. Limiting tooth length to 50mm as shown in the
guideline should help ensure properly filled concrete teeth.
To optimize the wire reinforcing installed on the project, it is likely only necessary to install
connection reinforcing in the regions 500mm from the ends of tie columns, rather than along the
entire height of the tie column. The exception to this rule is for tie columns bordering window
openings, where connection reinforcing will also be required at the base of the opening to double
as the reinforcing for the concrete sill. Note that wire reinforcing will be required at this location
for toothed tie columns as well. To ensure adequate connection reinforcing when tall masonry
units are used, the vertical spacing between wire reinforcing should be limited to 200mm or two
brick courses, whichever is less.
Wall Openings
The most significant changes to the Peru Guideline recommended by this report are with regard
to wall openings. The Peru Guideline already calls for the number and size of openings to be
minimized and the tops of openings to be located at the undersides of bond beams; these items
should be emphasized for reasons of economics and structural efficiency.
The Peru Guideline recommends tie columns at openings, but does not make them mandatory.
Based on testing and post-earthquake observations, severe cracking will develop at openings
without tie columns, thereby increasing the vulnerability of the wall. It is essential that tie
columns be installed at all opening locations of appreciable size (greater than 600mm x 600mm).
As well, tie column stirrup spacing should be reduced to 100mm over a length of 500mm
centered about the bottom corners of openings. This closer stirrup spacing is required to combat
the cracking that is common at wall opening corners and the diagonal cracks that form in the
short walls created by openings.
72
In addition to tie columns, reinforced concrete sills should be installed at all windows and other
non-full height openings requiring confinement. Currently, sills are not specified by the Peru
Guideline. A possible concrete sill detail that could be used is a member cast two brick courses
high and reinforced with two #8 (4.1mm diameter) wires. The above sill reinforcing is the same
as the tie column-masonry wall connection reinforcing already specified by the Peru Guideline.
Because of this fact, reinforcing of a sill could be accomplished by simply making the wire
connection reinforcing continuous between the tie columns bordering an opening. If masonry
units taller than 125mm are used, a sill one course high is probably sufficient. For these one-
course sills, however, the wire reinforcing must be hooked inside the adjacent tie columns
because there is no bed joint which the reinforcing can be extended into beyond the tie columns.
In many cases, the behaviour of confined masonry structures will not be significantly degraded
by the presence of small, unconfined openings. This being the case, it is still advisable to try to
reduce the number of unconfined openings by enlarging the openings at windows and doorways
to enable piping, ventilation and other services to pass through the wall at these locations (for
example, making the opening for a door taller than the door frame itself and installing a vent
above the door frame). If small openings are absolutely required for mechanical and electrical
work, these openings should be limited to 600mm x 600mm and located at the underside of bond
beams. Additionally, only one unconfined opening should be located between tie columns and it
should be placed such that there is at least 1m of confined wall on each side of the opening. If
the distance between tie columns is not great enough to accommodate 1m of uninterrupted wall
on each side of an unconfined opening, the opening should not be located in that wall.
Recommending that all openings greater than 600mm x 600mm be confined with tie columns
and concrete sills does negatively impact the economy of the CM system in the Peru Guideline.
The improved behaviour of buildings with only confined openings justifies the extra expense,
however. Providing adequate confining elements (tie columns, sills and lintels) around openings
is one of the most critical parameters for favourable seismic performance in CM walls.
Furthermore, the increased expense associated with confining openings encourages fewer, larger
openings rather than many, small openings; the former situation results in better and more
predictable structural behaviour. If large openings are required for mechanical and electrical
penetrations, these openings can be incorporated in enlarged windows and door openings as
described in the previous paragraph, rather than stand-alone openings.
Construction Inspection
The issue of poor workmanship is one that has been identified in deficient CM housing
throughout the world. To help address this issue, the Peru Guideline could be supplemented by a
series of inspection checklists. These checklists could take the form of “Before you pour the
concrete…” for rebar inspection. For mortar and concrete inspection checklists, guidance on
how to remediate mishaps should be incorporated. Table 6.1, Table 6.2 and Table 6.3 represent
a starting point for the development of CM construction inspection checklists. These tables are
based primarily on information in the Peru Guideline. In some cases, the requirements in the
checklists may seem onerous, but in the interests of keeping the checklists consistent with the
Peru Guideline, such requirements have not been relaxed.
73
Longitudinal bars are correct siz e and have 250mm long hooks at the base of the footing
Splices for the longitudinal bars are in the middle 1/3 of the floor height
Splices are of adequate length for the bar diameter
Only 1/2 of the longitudinal bars are spliced at the same locat ion
Assemb ly st irrups are provided over the depth of the footing
Longitudinal bars are of correct s ize and hook ed at ends of beams; hook s should be 150mm long and
located at the back face of the b eam intersection or corner
Splices are at the locations shown in the guideline and are of adequate length
Stirrups hooks are 75mm long and bent 135
o
; corner where hook located should alternate
Only 1/2 of the longitudinal bars are spliced at the same locat ion
Extra reinforcing is installed at lintels if required
No rebar is exposed at the s urface of the concrete
If above items are not satisfied, im mediately re move poorly bonded concrete, clean re inforcing bars and patch
with small-aggregate grout/concre te (1 pa rt cement, 4 parts sand, wa ter to make w orkabl e)
Concrete Element Post-Construction Inspection Checklist
Perform after forms for plinth, tie column, bond beam or suspended slab are removed
Item
Surface of concrete is free of large voids, segregat ion or honeycombing
Beam rebar cages are clean, raised up on spacers and tied correctly. Check that:
Width and depth of formwork is adequate; form is adequately braced
Construction joint locations are marked and proper details will be followed
Concrete mix proportions are correct
Importance of vibrating concrete and forms has been discussed
Plinth and Bond Beam Inspection Checklist
Perform before pouring plinth or bond beam
Item
Inside of forms is clean and free of debris
Tie column cages are clean and tied correctly . Refer to Tie Column Rebar Checklist. In particular, check that:
Footing width is large enough for soil t ype
Total footing depth and footing depth into native soil are adequate
Bottom of trench is level and has no loose s oil; lean concrete levelling course may be needed
Construction joint locations are marked and proper details will be followed
Foundation Inspection Checklist
Perform before pouring foundation concrete
Item
Trench is clean and free of organic s oil
Importance of vibrating concrete and forms has been discussed
Tie column cages are placed plumb, in the correct locations, and on spacers at bottom s o cage not s itting on s oil
Concrete mix proportions are correct
Table 6.1 – Inspection checklist table, part 1
74
Mechanical and electrical work is not embedded in wall
If above items are not satisfied, de-construct violating part of wa ll and re-construct correctly
Masonry wall is straight and plumb
Mortar joints are fully filled and between 10mm and 15mm thick
Wire reinforcing is installed at tie columns and sills
Wall is toothed at tie columns and voids for concrete sills are provided
Two-course high concrete sills with wire reinforc ing will be constructed beneath all window openings
Mechanical and electrical work will be installed outside of the masonry walls
Perform after construction of masonry wall is complete and before tie column forms are in place
Item
A stringline will be used to keep the wall straight and a plumb-bob will be used to keep wall vertical
Mortar mix proportions are correct
Mortar joints will be fully filled to a thickness between 10mm and 15mm
The wall will be toothed at tie columns and wire reinforcing will be placed at the tie column-masonry wall interface
Maximum wall c onstruction height in any one day will be 1200mm
Masonry Wall Inspection Checklist
Perform before starting construction of masonry wall
Item
Masonry units and plinth will be wet before wall const ruction starts
Splices are in the middle 1/3 of the floor height and are of adequate length
Only 1/2 of the longitudinal bars are spliced at the same location
Stirrups hooks are 75mm long and bent 135
o
; corner where hook is located s hould alternate
Stirrup size and spacing are as shown in the guideline, and additional stirrups are placed at openings
Longitudinal bars are of correct size and clean
Tie Column Rebar Inspection Checklist
At roof level, ends of longitudinal bars are hooked 250mm at top face of bond beam
At floor levels, longitudinal bars extend up to splice in the middle 1/3 of the next level
Perform before starting construction of wall adjacent to tie column
Item
Tie Column Pre-Pour Inspection Checklist
Perform after forms for tie column are installed
Item
Rebar cage is clean, undamaged and plumb
Inside of forms is clean and free of debris
Column forms are adequately braced
Tie columns at each level will be poured in two lifts: one to mid-st ory and one to the underside of the bond beam
Concrete mix proportions are correct
Table 6.2 – Inspection checklist table, part 2
75
Size and spacing of temperature rebar is as shown in the guideline
Size and length of joist rebar is as shown in the guideline: hooked bars at exterior support walls, specified
top bars at interior support walls and extra bottom bars at midspan of slab if required
Bottom b ars are splic ed in end 1/3 of slab s pan; splic es are of adequate length
50mm high wooden screed guides have been placed on top of ceiling bricks to ensure slab is of adequate thickness
Concrete mix proportions are correct
Importance of vibrating concrete and forms has been discussed; forms have been wet
Suspended Slab Inspection Checklist
Perform before pouring suspended slab floor or roof
Item
Construction joint locations are marked; joints should be at 1/3 point of span between support walls
Formwork and rebar is set up so the slab will span between the most closely spaced support walls
Inside of forms is clean and free of debris; no pipes are embedded in the slab
Ceiling bricks are spaced at 100mm
Hollow, lightweight ceiling bricks 150mm high x 300mm wide are used in the formwork for the slab
Slab formwork is as shown in the guideline, level, and supported on even, solid ground
Slab rebar is clean, raised up on spacers and placed correctly. Check that:
Table 6.3 – Inspection checklist table, part 3
6.2.2 Suggested Changes for Wider Applicability
In addition to suggestions on how the Peru Guideline should be altered to improve seismic
performance, this report will also make a number of recommendations on how the guideline can
be made more generally applicable. Where possible, these recommendations are based on
research or existing standards. For some of the topics addressed in this section, however, there is
little to no research or guidance in existing standards available. In such cases, reasonable
engineering judgement has been relied upon to determine ways to improve the applicability of
the Peru Guideline. The recommendations presented are intended to incorporate a reasonable
level of conservatism. Additional research is very much necessary, however, to verify the
adequacy of the recommendations. The future research required to validate the presented
recommendations will be introduced in this section and then expanded upon in Section 6.3.
The Peru Guideline does an excellent job of promoting good mortar and concrete quality by
providing mix proportions for each. With respect to masonry units, though, the guideline states
only that solid units should be used; it does not specify acceptable masonry unit properties such
as strength. Correspondence with the author of the Peru Guideline has revealed that the
guideline is based on solid, clay brick units, which when made into test prisms (small specimens
of masonry units and mortar constructed in the same way as actual walls) with the mortar
specified in the guideline, result in prisms with shear strengths of at least 0.51 MPa and
compressive strengths of at least 5.4 MPa. Future research focused on the effects of different
masonry unit properties and the suitability of non-solid units is required.
76
For the time being, the Peru Guideline should be implemented using solid, clay brick units that
meet the minimum strengths stated above. Non-solid units and units made from materials other
that clay could be used if testing of prisms of these units results in shear and compressive
strengths that are greater than the strengths mentioned above for prisms made from solid, clay
brick units. For non-solid units, it is important that comparison to the minimum strength values
above be performed using force units to account for the smaller cross-section area of non-solid
units.
In some instances, determining shear and compressive strengths of masonry prisms might not be
feasible. For such cases, in lieu of not checking unit quality at all, a qualitative approach should
be used. The Indonesian non-governmental organization Build Change uses a checklist to assess
the adequacy of solid masonry units; this checklist has been included in Appendix B.
Similar to the case of masonry units is that of rebar properties. The strength and type (deformed
or plain) of rebar assumed by the Peru Guideline is not stated. Discussion with the author of the
Peru Guideline, however, has revealed that deformed bars with a yield strength of 400 MPa have
been considered. Research into the effects of different rebar types and strengths would be
helpful. In the absence of such research, if plain bars are used, splice lengths should be doubled
[ASCE, 2007]; if lower strength rebar is used, the area of rebar should be increased by the ratio
400 / rebar yield strength. For low strength bars, bar diameter should be rounded up to satisfy
the aforementioned ratio. For 400 MPa bars, if the exact rebar sizes prescribed in the guideline
are not available, the next largest diameter bar that is available should be used.
With respect to different wall thicknesses, floor heights and height-to-thickness (h/t) ratios, the
Peru Guideline does not explicitly address variance in these parameters. The most critical of the
parameters is h/t ratio. The maximum h/t ratio from the sample plans in the Peru Guideline is 17.
This h/t ratio is quite low compared to that of the more slender walls typically used in places like
Indonesia. For the Peru Guideline to be more applicable in such regions, it must be able to
incorporate larger h/t ratios. Very little research has thus far been conducted on the impact of
larger h/t ratios on CM walls; such research, including out-of-plane dynamic testing, is required.
In the meantime, guidance on appropriate h/t ratios for seismicities lower than that assumed in
the Peru Guideline (0.4g) can be obtained by applying out-of-plane research performed on URM
walls.
Analytical research by Sharif et al. [2007] suggests an approximately linear trend between
maximum h/t ratio and ground acceleration for URM walls. Results of this research show 17 as
a reasonable maximum h/t ratio for 0.4g of acceleration and 22 as reasonable ratio for an
acceleration of 0.3g. This being the case, for regions of low to moderate seismicity, it is likely
conservative to apply the Peru Guideline with h/t ratios as large as 22.
The Peru Guideline specifies a reinforced concrete joist floor/roof system that makes use of
lightweight ceiling bricks to reduce the quantity of concrete. In some regions, different roof and
floor systems are used. For instance, in Southeast Asia, wood truss roof systems that do not act
as stiff diaphragms are commonly used; refer to Build Change [2006] for details on this type of
structural system. Future research is necessary to examine how different floor and roof systems
can be incorporated into the Peru Guideline. In the meantime, the floor/roof system specified in
77
the Peru Guideline should be implemented when this guideline is used. Installing lightweight
ceiling bricks, rather than heavier conventional masonry units, is essential. In areas where
lightweight ceiling bricks are not available, ceiling bricks can be replaced by wooden block-outs.
The wooden block-out system is actually preferable to ceiling bricks because the weight of the
floor/roof structure is lower (resulting in lower seismic forces) and there is no concern with
ceiling bricks falling out during earthquake shaking.
Likely the biggest barrier to general applicability of the Peru Guideline is the issue of seismicity.
The guideline itself does not say what seismicity it is intended for; it merely implies that is good
anywhere in Peru. Consulting the Peru Construction Code, the maximum PGA in the country is
0.40g [SENCICO, 2003]. Basing seismicity purely on PGA is simplistic, but for stiff buildings
like CM structures it can be used as a reasonably accurate measuring stick of seismicity.
Research is required into whether or not seismic hazard can be addressed solely by wall density,
or whether adjustments in parameters like tie column longitudinal reinforcing are also required.
Studies looking at how minimum wall density (and possibly other parameters) should be varied
with varying seismicity must also be performed.
In lieu of the research suggested above, it seems reasonable to apply the Peru Guideline as is in
regions characterized by a PGA of 0.40g. For areas of more moderate seismicitiy (PGA of 0.30g
or less), it is recommended that the wall density values in the Peru Guideline be multiplied by a
factor of 0.75 to account for the lower seismic forces expected in these areas. As an extension of
this recommendation, the wall densities specified in the guideline should be used for regions
with PGAs greater than 0.3g. In regions of moderate seismicity, it is recommended that the other
requirements in the Peru Guideline (e.g. tie column spacing, reinforcement details, etc.) not be
changed, since these features provide the structure with additional robustness to withstand
varying levels of seismic demand.
6.3 Optimizing the Peru Guideline - Future Research
Section 6.2 lists a number of ways that the Peru Guideline could be altered to better ensure
adequate seismic performance and enable wider applicability. In this section, potential research
topics will be outlined that could help further improve the seismic safety of the CM system
specified in the Peru Guideline, and optimize this system for different site conditions.
The research areas proposed in this section will differ from those suggested in Section 5. In
Section 5, the potential research projects identified were focused on resolving discrepancies
between the CM guidelines not dealt with by current research. In this section, future research
topics will be proposed that are concerned solely with improving the Peru Guideline.
A number of areas requiring further research were briefly identified in Section 6.2. The areas
indicated as needing more investigation included masonry unit properties, reinforcing steel
properties, masonry wall h/t ratio, different wall/floor systems and varying seismic hazard. Each
of these topics will now be described in greater detail.
78
With respect to masonry units, there is research that could be conducted on a few different topics
that would help improve the Peru Guideline. Investigating how different unit strengths, material
types (different types of clay and concrete), and partially hollow cross-sections affect the
behaviour of the Peru Guideline CM system would be quite worthwhile. The goal of research in
this area would be to refine recommendations on the compressive and shear strength required for
units constructed from different materials and/or with partially hollow cross-sections.
Complimenting the guideline with such recommendations would enable the Peru Guideline to be
used more confidently to construct CM houses from a wide variety of masonry units, thereby
increasing its applicability. Examining how the Peru Guideline system could be altered for units
with strengths greater or less than that currently assumed could also be explored; however,
implementing the results of such research could start to make the guideline prohibitively
complex. Providing a list of alternate acceptable units commonly found throughout the world
would maintain the simplicity of the guideline.
For the case of reinforcing steel, conservative methods of incorporating plain reinforcing bars
and lower strength reinforcing steel were discussed in Section 6.2.2. Research verifying and
improving upon these methods would be very helpful, particularly in light of the fact that in
many cases only one type of reinforcing bar is available in a given area. Such research would
enable the Peru Guideline to be more confidently applied in regions using plain or low strength
reinforcing.
In line with the typical construction in Peru, the Peru Guideline specifies relatively stocky walls;
i.e., walls with low height-to-thickness ratios. The prevailing failure mode in such structures,
assuming all connections and diaphragms are adequate, is in-plane diagonal shear failure. As
mentioned previously, in other parts of the world like Indonesia, more slender walls are used.
Using URM research, an increased maximum h/t ratio limit for moderate seismcities was
proposed in Section 6.2.2. What is really required, however, is research on the effect of height-
to-thickness ratio on the behaviour of the Peru Guideline system. The primary focus of such
research should be to identify the h/t ratio at which the performance of the Peru Guideline system
becomes inadequate for various ground accelerations, perhaps due to out-of-plane or combined
in-plane/out-of-plane failure. Out-of-plane dynamic testing, in particular on CM walls with high
h/t ratios, would be a necessary component of this research. Out-of-plane research on the Peru
Guideline system would help determine safe upper limits for h/t ratios for various seismic
hazards, greatly increasing the applicability of the Peru Guideline in regions that traditionally use
slender walls.
The monolithic cast-in-place concrete slab and joist system currently specified by the Peru
Guideline is a very good floor and roof structure for CM houses in seismic regions. In addition
to providing good diaphragm action and strong connections, it is structurally efficient, easy to
form and minimizes the quantity of concrete required. The only major concern with this slab
system is the potential for ceiling bricks to become dislodged during earthquake shaking and fall
on people inside the house. Research is required to ascertain if ceiling brick fallout is a problem
during shaking, and if measures can be taken to alleviate this potential problem.
It should be noted that some other concrete floor/roof systems are not suitable for CM housing in
seismically volatile areas. A crucial function of the floor/roof system is to transfer seismic forces
79
horizontally through diaphragm action. Some concrete systems, for instance floors and roofs
comprised of precast concrete slabs on precast concrete joists, are not continuously connected
together and consequently have demonstrated poor diaphragm behaviour in some cases. Non-
monolithic concrete systems are inherently less suitable for transferring horizontal forces and
should be avoided in simple, CM residential structures.
In locations where lightweight ceiling bricks are not available and workers are not experienced at
forming, pouring and finishing suspended slabs, it may be difficult to effectively construct the
Peru Guideline slab/joist system. In such locations, the applicability of the Peru Guideline would
be greatly enhanced if there were other floor/roof options in the guideline that more closely
match the prevalent construction in a region.
Research into how different floor and roof systems could be incorporated into the Peru Guideline
would improve the applicability of the guideline. Systems potentially worth investigating
include the concrete slab/joist system utilizing all wood forms mentioned in Section 6.2.2 (more
difficult to form but no lightweight ceiling bricks required), a flat concrete slab system (simple
formwork but more concrete required since less efficient), and various types of wood systems.
Research on wood floors/roofs should work towards recommendations on: plywood/plank
thickness for different types of wood and joist spacings; joist spacing for different depths and
wood types; bridging and horizontal truss layouts (to ensure floor works as a diaphragm); and
connection details that facilitate lateral and vertical force transfer to the supporting concrete bond
beams. The idea of including pitched roofs built on top of structural flat roofs could also be
included, similar to the Schacher Guideline.
In most cases, the above floor/roof systems have already been developed for other structures and
would merely have to be modified for use in CM buildings. It would still be a good idea to test
the different floor/roof systems with the CM walls from the Peru Guideline to verify their
behaviour. Such testing could identify the effects of different floor/roof systems on seismic
demand. Wood floors and concrete slabs with joists and no ceiling bricks weigh considerably
less than the floor/roof system in the Peru Guideline; consequently, use of these systems could
lower the requirements for wall density and tie column reinforcement. Anecdotal evidence from
Indonesia seems to support this possibility. One-storey CM housing with wooden roofing has
performed well during recent seismic events in Indonesia despite the fact that it has smaller
confining elements, less longitudinal rebar and larger stirrup spacing than that recommended in
the Peru Guideline.
In Section 6.2.2, it was reasoned that the Peru Guideline is based on a PGA of 0.4g, and is
therefore likely to be suitable for regions characterized by high seismicity (PGAs greater than
0.3g). For regions of moderate seismicity (PGAs less than 0.3g), a 25% reduction was proposed
in the wall density specified by the Peru Guideline to reflect the lower seismic hazard and make
CM buildings in these regions more economical. These recommendations need to be verified.
Research is necessary to confirm how differing levels of seismicity can be incorporated into the
Peru Guideline, specifically, if seismicity can be addressed in the same manner as the guideline
considers varying soil conditions: by adjusting the wall density. Research on this topic would
have to investigate if other parameters, such as tie column cross-sectional area and quantity of
80
longitudinal and transverse reinforcement, must also be adjusted at different PGA levels. The
goal of this research would be to develop specific recommendations on how the Peru Guideline
system should be altered for different seismicities. Ideally, the research would conclude that
confining element detailing can be held constant for all seismic hazards, and suggest minimum
wall densities to accommodate different PGA levels.
An important part of research on wall density would be to confirm the current relationship
between soil type and wall density in the Peru Guideline. Presently, the Peru Guideline specifies
higher wall densities for softer soils. This relationship may have as much to do with preventing
damage from differential settlement in poorer soils as ensuring adequate seismic resistance.
Consequently, research is required on how wall density must be adjusted to compensate for the
seismic demand resulting from different soil conditions. This research should stipulate if the
current wall density-soil type relationship needs to be altered to better incorporate seismic
demand, and if so, how it must be changed.
There are also a number of research topics not introduced in Section 6.2 that would help
optimize the Peru Guideline. Research looking at the confining elements and openings specified
in the Peru Guideline would be quite beneficial. In addition, research work investigating mortar
bed reinforcing and intermediate bond beams and tie columns could also prove helpful.
When research on how the Peru Guideline could be adjusted for varying seismicity was
discussed above, it was stated that the effects of confining element parameters (cross-sectional
area, longitudinal rebar, transverse rebar) must be investigated. In particular, it was indicated
that it would be ideal if the confining elements called up in the Peru Guideline could be kept
constant for all seismic hazards. A natural extension of research on varying seismicity would be
an experimental study focused on optimizing the confining elements currently called up in the
Peru Guideline. This would involve checking if cross-sectional area, longitudinal rebar area,
transverse hoop spacing and locations of closer hoop spacing (ends of members and corners of
openings) can be reduced to yield a more economical system that also provides adequate seismic
resistance. This optimization could be extended beyond just reducing the cost of typical
confining elements to investigating if smaller, one or two rebar tie columns could be used at less
critical locations such as the second level and some opening locations; this idea was discussed
more generally in Section 5. The goal of such research would be to develop specific
recommendations on how the typical confining elements in the Peru Guideline can be altered,
and if smaller tie columns can be used, what the size and reinforcing of these columns should be
and where the columns can be located. To keep the guideline simple, it would be preferable if
the revised confining elements could be used for all seismic hazards.
In Section 6.2, it was recommended that all openings larger than 600mm x 600mm be confined.
For openings slightly larger than the 600 x 600 limit, the extra cost of confining the opening may
not be justified by the improvement in structural behaviour. Research into what maximum
unconfined opening size will not adversely affect the seismic performance of the Peru Guideline
could help improve the economy of this CM system. Such research should explicitly state what
the size and shape of permissible unconfined openings are and if there are certain locations
where they can not be placed.
81
Research on larger confined openings could also help optimize the Peru Guideline. Investigating
the potential for smaller one or two rebar tie columns at some openings (as discussed previously)
could yield cost savings. It would be important for research in this area to suggest at what
openings smaller tie columns could be used at (size, shape, location) and what the properties
(cross-sectional area, rebar) of smaller tie columns should be. The effectiveness of the sill
proposed in Section 6.2 for confining openings could also be checked through research and
required revisions suggested. Research into lintels, both composite and not composite with bond
beams, would complete the investigation of openings in Peru Guideline CM walls. For lintels
composite with bond beams, research should strive to provide specific recommendations on
maximum lintel span and the extra lintel depth required at maximum spans for lintels. For lintels
not composite with bond beams, a topic beyond the scope of the current Peru Guideline,
recommendations are needed on lintel depth, reinforcing, maximum span and minimum distance
below the bond beam. Recommendations on this last topic would enable openings to be located
away from the tops of walls, thereby expanding the flexibility of the guideline.
Bed reinforcing and intermediate tie columns are two potential ways of improving CM
construction not addressed by any of the guidelines examined in this report. A number of
research projects have been conducted on these topics. Testing of CM walls with bed
reinforcement suggests some improvement in seismic performance. Whether or not bed
reinforcement would result in a great enough improvement in the Peru Guideline system to offset
concerns with availability, corrosion and greater cost would need to be investigated. Similarly,
past research on intermediate bond beams and tie columns suggests these elements improve
seismic performance, but whether their inclusion in the Peru Guideline system (possibly in the
form of smaller, one-rebar members) is advisable needs to be examined in a research program.
6.4 Closure
In this section, the five guidelines investigated in this report were evaluated based on their
potential to be generally applied in many regions around the world. This evaluation found the
Peru Guideline to be the guideline most suitable for general application. Once this conclusion
was reached, a number of measures were suggested to improve both the seismic performance and
applicability of the CM system prescribed by the Peru Guideline. Even with the improvements
suggested, however, there remains a number of ways the Peru Guideline could be enhanced and
optimized; the latter third of this section outlined the research necessary to bring about further
improvements to the Peru Guideline.
With the improvements suggested in Section 6.2, it would be reasonable to apply the Peru
Guideline generally in seismic regions using the materials and systems identified in the
guideline. Completion of the future research proposed in the section would allow the Peru
Guideline to be implemented for a variety of materials, building styles, floor/roof systems and
seismic hazards, with greater economy and confidence in the behaviour of the structure.
82
7 CONCLUSION
The overall goal of this report was to develop a framework that would provide the information
necessary to construct confined masonry housing with good seismic resistance in many regions
around the world. To achieve this goal, the report first analyzed how CM structures behave and
the different ways in which CM houses fail during seismic shaking. Next, current confined
masonry practice was investigated, noting the strengths and shortcomings of current CM
construction. After this, five CM guidelines were examined and compared, noting how the
guidelines alleviate some of the concerns with current CM practice. Comparison of the CM
guidelines yielded some irresolvable discrepancies; confined masonry research was consulted for
guidance on these discrepancies, and also to address certain weaknesses in the guidelines and
areas requiring further research.
Taking into consideration the information presented in the above tasks, the five CM guidelines
examined in the report were evaluated to determine which of these guidelines are appropriate for
application in the many different seismic regions around the world. It was concluded that only
the Peru Guideline has the potential to be applied generally. The applicability of the Peru
Guideline is somewhat limited in its current form, however. Consequently, a number of
recommendations were proposed to help improve the ability of the guideline to be applied
outside of Peru. Several changes focused on improving the seismic performance of the Peru
Guideline were also suggested. The improvements to the Peru Guideline recommended by this
report are as follows:
1.
Do not install unreinforced chimneys or masonry gable end walls.
2.
If possible, avoid heavy clay roof tiles; if tiles are installed, fasten them securely to the CM
building structure.
3.
Avoid masonry parapets; construct parapets from lighter materials and fasten them securely
to the CM building structure; if masonry parapets are installed, properly confine them and
minimize their height.
4.
Construct partition walls to the underside of ceiling and connect them to the floor/roof
system for lateral support; it is preferable to build partition walls of lightweight materials or
convert partition walls to structural masonry walls.
5.
Do not embed mechanical and electrical piping in masonry walls.
6.
If tie columns are toothed, ensure teeth are completely filled with concrete.
7.
If wire reinforcing is installed through tie columns, only place wire in the region 500mm
from the end of tie column and as necessary for opening sills; vertically space wire
reinforcing in end regions at 200mm or every second brick course, whichever is less.
8.
Install tie columns on both sides of openings greater than 600mm x 600mm; at non-full
height openings, reduce stirrup spacing to 100mm for 500mm centered about the bottom
corners of openings.
9.
At non-full height openings, install concrete sills two brick courses high to a maximum of
250mm high reinforced with two 4.1mm diameter wires.
10.
Pass piping, ventilation and other services through enlarged confined openings for
windows/doors rather than smaller unconfined openings if possible.
83
11.
If unconfined openings are necessary, limit them to 600mm x 600mm in size and one
opening per wall; locate unconfined openings at the underside of bond beam; ensure there is
1m of wall on each side of unconfined openings.
12.
Inspect CM construction using the checklists in Table 6.1, Table 6.2 and Table 6.3.
13.
Use solid, clay brick masonry units resulting in minimum prism shear strengths of 0.51 MPa
and minimum prism compression strengths of 5.4 MPa; masonry units that are non-solid
and/or made from other materials may be used if testing of such units demonstrates unit
strengths (in force units) greater than the minimums for solid, clay units.
14.
If testing of solid masonry units is not feasible, use the checklist in Appendix B to
qualitatively assess unit quality.
15.
Use 400 MPa deformed reinforcing bars; if plain bars are used, double the specified splice
lengths; if lower strength reinforcing is used, increase the area of bars by 400/bar yield
strength; select the smallest available reinforcing bar size that exceeds the required bar area.
16.
For PGA > 0.3g, maximum wall h/t ratio is 17; for PGA < 0.3g, maximum wall h/t ratio is
22.
17.
Ensure lightweight ceiling bricks are used in floors and roofs, or preferably, block out ceiling
brick locations with wood forms.
18.
For PGA > 0.3g, use minimum wall densities specified in guideline; for PGA < 0.3g,
multiply specified minimum wall densities by 0.75.
To further improve the performance, safety, economy and applicability of the Peru Guideline,
additional research is required. A number of research programs that would help optimize the
Peru Guideline were outlined. Special attention was given to what the output of these research
programs should be.
Although slightly more expensive than more conventional unreinforced masonry construction,
confined masonry housing exhibits far superior seismic performance when it is constructed
correctly. The Peru Guideline, modified by the recommendations in this report, provides a
comprehensive and simple method for constructing economical CM houses, thereby ensuring
these houses have good earthquake resistance. Further research is required to optimize the Peru
Guideline. Until such research is performed, however, the guideline, along with the suggestions
in this report, represents a means to achieve housing with good seismic safety in earthquake-
active regions around the world.
84
8 REFERENCES
8.1 References Cited
Alcocer, S.M., Arias, J.G., and Vazquez, A. (2004). "Response assessment of Mexican confined
masonry structures through shaking table tests." 13th world conference on earthquake
engineering , Vancouver , B.C. , Canada , No. 2130
Aguilar, G., Meli, R., Diaz, R., Vazquez-del-Mercado, R., (1996). "Influence of horizontal
reinforcement on the behavior of confined masonry walls, paper No 1380, 11th World
Conference on Earthquake Engineering
Andaya, Rodelio Mercado (2004). "Improvement of seismic performance of confined masonry
structures" Individual Studies by Participants at the International Institute of Seismology and
Earthquake Engineering, vol. 40, 243-255, 2004
ASCE (2007). "Seismic Rehabilitation of Existing Buildings, ASCE/SEI 41." American Society
of Civil Engineers, Reston, Virginia.
Asociación Columbiana de Ingeniería Sísmica (2001). "Manual de Construcción, Evaluación y
Rehabilitación Sismo Resistente de Viviendas de Mampostería", AIS, Bogota, Columbia,
http://www.desenredando.org/public/libros/2001/cersrvm/mamposteria_lared.pdf
Astroza, M. and Schmidt, A. (2004). "Capacidad de deformacion de muros de Albanileria
confinada para distintos viveles de desempeno (Deformation capacity of confined masonry for
different performance levels)." Revista de Ingeniería Sísmica No. 70, 59-75
Blondet, M., Duenas, M., Loaiza, C., and Flores, R. (2004). "Seismic vulnerability of informal
construction dwellings in Lima, Peru: Preliminary diagnosis" 13th world conference on
earthquake engineering, Vancouver, B.C., Canada , No. 2122
Build Change, (2006). “Earthquake-Resistant Design and Construction Guideline – For Single
Story Reinforced Concrete Confined Masonry Houses Built in the Aceh Permanent Housing
Reconstruction Program.” Build Change, Indonesia
Building Code E 030 Earthquake Resistant Design (Norma Técnica de Edificación E.030
DISEÑO SISMORRESISTENTE). Servicio Nacional de Normalización, Capacitación e
Investigación para la Industria de la Construcción (SENCICO), Lima, 2003
Casabonne, C. (2000). "Masonry in the seismic areas of the Americas, recent investigation and
developments." Progress in Structural Engineering and Materials. Vol. 2, no. 3, pp. 319-327.
July-Sept. 2000
85
Construction and maintenance of masonry dwellings for masons and builders (Construcción y
Mantenimiento de Viviendas de Albañilería para Albañiles y Maestros de Obra). EERI,
Pontificia Universidad Católica del Perú, and SENCICO, Second edition, 2005
Construction under seismic condition in the Balkan region. Vol. 3: Design and construction of
stone and brick masonry buildings (UNIDO/UNDP, Vienna, 1984)
Earthquake Engineering Research Institute (2003). "Quick report to EERI, SMIS, CENAPRED
and GIIS reading the earthquake in Colima, Mexico January 21, 2003." EERI, Oakland, USA,
http://www.eeri.org/lfe/pdf/mexico_colima_eeri_report
Eurocode 6: Design of masonry structures - Part 2: Design, Selection of materials and execution
of masonry, European committee of standardization, draft No 8, 2003
Eurocode 8: Design of structures for earthquake resistance - Part 1: General rules, seismic
actions and rules for buildings, European committee of standardization, draft No 5, 2002
French, M. (2004). "Student impressions of South American construction." Earthquake Hazard
Centre Newsletter, Vol.7 , No.4
Ghaidan, U. (2002). "Earthquake-resistant masonry buildings: basic guidelines for designing
schools in Iran." Published by UNESCO
"Guidelines for earthquake resistant non-engineered construction: Chapter 3 & 4." International
Association for Earthquake Engineering (IAEE), Tokyo, Japan and National Information Centre
of Earthquake Engineering (NICEE), Kanpur, India, 2004.
http://www.nicee.org/IAEE_English.php
Irimies, M.T., (2002). “Confined Masonry Walls: the influence of the tie-column vertical
reinforcement ratio on the seismic behaviour.” The proceedings of the twelfth European
Conference on Earthquake Engineering , No. 7
Ishibashi, K.; Meli, R.; Alcocer, S.M.; Leon, F.; Sanchez, T.A. (1992). "Experimental study on
earthquake-resistant design of confined masonry structures." Proceedings of the Tenth World
Conference on Earthquake Engineering, Madrid, Spain ,page 3469-3474
Liu, D.H. and Wang, M.Z., (2000). "Masonry structures confined with concrete beams and
columns." Proceedings of the 12th World Conference on Earthquake Engineering, Auckland,
New Zealand. No.2720
Lutman, M. and Tomazevic, M., (2002), "Housing Report – Confined Brick Masonry House.",
World Housing Encyclopedia, No. 88,
http://www.world-housing.net/whereport1view.php?id=100096
86
Meli, R., (1991). "Seismic design of masonry buildings: the Mexican practice ."Proceedings of
the 1991 International Symposium on Building Technology and Earthquake Hazard Mitigation,
City of Kunming, Yunnan Province, China, 193-210
Moroni, M., Astroza, M. and Caballero, R., (2000). "Wall density and seismic performance of
confined masonry buildings", TMS Journal, July 2000, pp 81-88
Moroni, M., Astroza, M. and Acevedo, C., (2004). "Performance and seismic vulnerability of
masonry housing types Used in chile." Journal of Performance of Constructed Facilities,Vol. 18,
No. 3
Radovanovic, Z. (1998). "Nonlinear dynamic analysis of confined masonry buildings according
to Yugoslav regulations and Eurocodes." Proceedings of the Eleventh European Conference on
Earthquake Engineering, 1998, Paris, France
Riahi, Z. (2007). "Backbone model for confined masonry walls for performance-based seismic
design." Master’s Thesis, University of British Columbia, Vancouver, Canada
San Bartolomé, A. and Ordóñez, J., (2005). "Comparación del comportamiento sísmico de un
muro de albañilería confinada tradicional y otro caravista." Libro De Ponencias Del XV
Congreso Nacional De Ingenieria Civil Ayacucho, October 2005.
San Bartolome, A. and Torrealva, D., (1990). "A new approach for seismic design of confined
masonry buildings in Peru." Proceedings of the Fifth North American Masonry Conference;
University of Illinois at Urbana-Champaign, volume 1, 37-44
San Bartolome, A., Quiun, D. and Mayorca, P., (2004). "Proposal of standard for seismic design
of confined masonry buildings." Bulletin of ERS, No. 37
Schacher, T., (2006). "Confined masonry for one and two story buildings." International
committee of the red cross (ICRC), Geneva
Sharif, I., Meisl, C. and Elwood, K.J. (2007), "Assessment of ASCE Height to Thickness Ratio
Limits for URM Walls." Earthquake Spectra, Volume 23, No. 4
Tomazevic, M., Bosiljkov, V.,and Weiss, P. (2004) "Structural behavior factor for masonry
structures." 13th world conference on earthquake engineering, Vancouver, B.C., Canada,
Accession No. 2642
Tomazevic, M., and Klemence, I. (1997a). "Seismic behavior of confined masonry walls."
Earthquake engineering and structural dynamics, vol. 26, 1059-1071
Virdi, S.K. and Raskkoff, D.R., (2005). "Low rise residential construction detailing to resist
earthquakes: Confined brick masonry"
http://www.staff.city.ac.uk/earthquakes/MasonryBrick/References.htm
87
Yaiiez, F., Astroza, M., Holmberg, A. and Ogaz, O., (2004). “Behavior of confined masonry
shear walls with large openings” 13th world conference on earthquake engineering, Vancouver,
B.C., Canada, No. 3438
Yoshimura, K. and Kikuchi, K., (1995). "Experimental study on seismic behavior of masonry
walls confined by RC frames." Proceedings of the Pacific Conference on Earthquake
Engineering; Melbourne, Australia, 20-22 November 1995. Australian Earthquake Engineering
Society, Parkville, Victoria , volume 3, 97-106
Yoshimura, K., Kikuchi, K., Okamoto, Z. and Sanchez, T., (1996). "Effect of vertical and
horizontal wall reinforcement on seismic behavior of confined masonry walls." Proc. 11th world
Conf. Earthquake Engng., Acapulco, No. 191
Yoshimura, K., Kikuchi, K., Kuroki, M., Nonaka, H., Tae Kim, K., Wangdi, R., and Oshikata,
A. (2004a). "Experimental study on effects of height of lateral forces, column reinforcement and
wall reinforcement on seismic behavior of confined masonry walls." 13th world conference on
earthquake engineering, Vancouver, B.C., Canada, No. 1870
Zabala, F., Bustos, L.J., Masanet, A. and Santalucia, J., (2004). "Experimental behavior of
masonry structural walls used in Argentina." 13th world conference on earthquake engineering,
Vancouver, B.C., Canada, No. 1093
8.2 References Consulted
Alcocer, S.M., "Implications Derived from Recent Research in Mexico on Confined Masonry
Structures.”, Structural Engineering and Geotechnical Area, National Center for Disaster
Prevention, Av. Delfín Madrigal 665, Pedregal de Sto. Domingo, Mexico, DF 04360
Alcocer, S.M., Arias, J.G. Vazquez, A. (2003). "The new Mexico City building code
requirements for design and construction of masonry structures" Proceedings, the ninth North
American Masonry Conference, Clemson, South Carolina, 656-667
Alcocer, S.M., Arias J.G., and E Flores, L. (2006). "Some developments on performance-based
seismic design of masonry structures." Institute of engineering, UNAM, Mexico
Alcocer S.M., and Klingner, R.E. (2006). "The Tecomán, México earthquake January 21, 2003."
An EERI and SMIS Learning from Earthquakes Reconnaissance Report.
Aschheim, M., Flanagan, S., Harlander, J., Pitt, C., Alfaro C., Rivas, C.,and Rodriguez, M.E.
(2006). "Improving the earthquake resistance and sustainability of confined masonry (Mixto)
dwellings in El Salvador." Proceedings of the 8th U.S. National Conference on Earthquake
Engineering, San Francisco, California, USA, No.1462
Brzev, S., Green, M., Arnold, C., Blondet, M., Cherry, S., Comartin, C., D'Ayala, D., Farsi, M.,
Jain, S., Naeim, F., Pantelic, J., Samant, D., and Sassu, M. (2004). "The web-based world
88
housing encyclopedia: housing construction in high seismic risk areas of the world." 13th world
conference on earthquake engineering , Vancouver , B.C. , Canada , No. 1677
Brzev, S., Greene, M., Bo¸stenaru, M., D’Ayala, D., Comartin, C., Jain, S., Blondet, M., Sassu.
M., Farsi, M., Naeim, F. (2005). "A web-information system for housing types in urban areas
exposed to strong earthquakes around the world:The World Housing Encyclopedia."
Geophysical Research Abstracts, Vol. 7, 01849
Comartin, C., Brzev, S., Naeim, F., Greene, M., Blondet, M., Cherry, S., D’Ayala, D., Farsi,
M., Jain, S., Pantelic, J., Samant, L., and Sassu, M. (2004). "A challenge to earthquake
engineering professionals." Earthquake Spectra, Volume 20, No. 4, pages 1049–1056
D’Ayala, D., Bostenaru, M., Goretti, D., Yakut, A., Tomazevic M., (2005). "Application of the
‘World housing encyclopedia’." Geophysical Research Abstracts, Vol. 7, 01649
Eshghi, S., and Naserasadi , K. (2005). "Performance of essential buildings in the 2003 Bam,
Iran, earthquake." Earthquake Spectra, Vol. 21, No. S1, S375–S393
Gibu, P. and Zavala, C. "The current state-of -the art of masonry and adobe buildings in Peru."
Japan Peru Center for Earthquake Engineering Research and Disaster Mitigation
Jain, S.K., Murty, C.V.R., Arlekar, J.N., Rajendran, C.P., Rajendran, K. ,and Ravi Sinha, T.
(1999). "Chamoli (Himalaya, India) earthquake of 29 March 1999." EERI Newsletter Vol.33,
No.7,
http://www.nicee.org/Chamoli.php
Jain, S.K., Murty, C.V.R., Chandak, N., Seeber, L., and Jain, N.K. (1994). "The September 29,
1993, M6.4 Killari, Maharashtra earthquake in Central India." EERI Newsletter, Vol. 28, No. 1,
http://www.nicee.org/Killari.php
Jain, S.K., Murty C.R.V., Dayal, U., Arlekar J.N., and Chaubey, S.K. (2001). "Learning from
Earthquakes: A field report on structural and geotechnical damages sustained during the 26
January 2001 M7.9 Bhuj Earthquake in Western India." Department of civil engineering Indian
Institute of Technology Kanpur,Kanpur, India,
http://www.nicee.org/Bhuj.php
Jain, S.K, Singh, R.P., Gupta, V.K., and Nagar, A. (1992). "Garhwal earthquake of Oct. 20,
1991." EERI Special Earthquake Report, EERI Newsletter, Vol.26, No.2,
http://www.nicee.org/Uttarkashi.php
Learning from earthquakes: The Northern Sumatra earthquake of March 28, 2005. EERI Special
Earthquake Report, 2005,
http://www.eeri.org/lfe/indonesia_sumatra3.html
Meli, R., and Alcocer, S.M. (2004). "Implementation of structural earthquake-disaster mitigation
programs in developing countries." Natural hazards review, Vol. 5, No. 1
89
Marinilli, A.,and Castilla, E (2004). "Experimental evaluation of confined masonry walls with
several confining-columns."13th world conference on earthquake engineering, Vancouver, B.C.,
Canada, Accession No. 2129
Murtry C.V.R. (2003). "Earthquake tips: Tip 12, 13, 14 and 15", Indian institute of technology
Kanpur and building materials and technology promotion council, India
http://www.iitk.ac.in/nicee/EQTips/EQTip22.pdf
Naderzadeh, A., and Khademi, M.H.(1998). "A summery of "A preliminary report on the
Ardekul ,Iran Earthquake of 10 May 1997." Earthquake hazard centre newsletter, Vol. 1, No. 3
Nasrollahzade Nesheli, K. (2004). "Structural damage to buildings due to Bam earthquake of
December 26, 2003, Iran." ICUS/INCEDE newsletter Vol. 3, No.4
Nasrollahzadeh Nesheli, K.,and Meguro, M. (2006). "Damage to Non-engineered structures due
to past earthquakes in Iran." Proceedings of the 8th U.S. National Conference on Earthquake
Engineering, San Francisco, California, USA, No.2117
Patil, P.B., Holihosur, S.N., and Kallapur, V.L. (2006). "Effects of the 2005 Muzaffarabad
(Kashmir) earthquake on built environment." CURRENT SCIENCE, VOL. 90, NO. 8
Ramirez, J. (1995). "The October 9, 1995 magnitude 7.6 Manzanillo, Mexico Earthquake."EERI
special earthquake report,
http://www.eeri.org/lfe/mexico_manzanillo.html
Rodriguez, M., and Blondet, M. (2004) “Evaluation of housing losses in recent earthquakes in
Latin America." 13th world conference on earthquake engineering, Vancouver, B.C., Canada,
No. 2203
"Seismic Evaluation Procedures for Existing Civil Works Powerhouses." United States Army
Corps of Engineers (USACE), Washington, DC, United States, ETL 1110-2-568, 2006.
http://www.usace.army.mil/publications/eng-tech-ltrs/etl1110-2-568/entire.pdf
Tobriner, S., Comerio, M., and Green, M. (1997). "Learning from earthquakes: Reconnaissance
report on Umberia- Marrche, Italy, Earthquakes of 1997." EERI special earthquake reports
http://www.eeri.org/lfe/italy_umbria_marche.html
Tomazevic, M., and Klemence, I. (1997b). "Verification of seismic resistance of confined
masonry buildings." Earthquake engineering and structural dynamics, vol. 26, 1073-1088
Yoshimura, K., Kikuchi, K., Kuroki, M., Nonaka, H., Tae Kim, K., Wangdi, R. and Oshikata, A.,
(2004b)." Experimental study for developing higher seismic performance of brick masonry
walls." 13th world conference on earthquake engineering, Vancouver, B.C., Canada, No. 1597
Appendix B
QUALITATIVELY EVALUATING MASONRY UNIT ADEQUACY:
THE BUILD CHANGE APPROACH
Source – Earthquake-Resistant Design and Construction Guideline – For Single Story
Reinforced Concrete Confined Masonry Houses Built in the Aceh Permanent Housing
Reconstruction Program, Build Change, Indonesia, 2006.
Applicability – Solid clay brick masonry units.
Procedure – Check that units adhere to the following:
-
No cracks or chips
-
No visible unmixed portions or divots
-
Units are square, not warped or curved
-
Unit dimensions are consistent among a sample of 10-20 units; dimensions do not vary
by more than 10mm in length and 5mm in height and width
-
When two units are hit together, they make a metallic clink rather than a dull thud
-
When soaked in water for 24 hours, units do not crumble
-
When subjected to the 3-point bending test illustrated in Figures B.1 and B.2, 80% of
units pass; note that the test has been calibrated to an individual of 60kg applying static
pressure to the unit (no bouncing)
Figure B.1 – 3-Point Bending Test: Passing Test Figure B.2 – 3-Point Bending Test: Failing Test
