SEISMIC DESIGN PHILOSOPHY FOR BUILDINGS

The Earthquake Problem

Severity of ground shaking at a given location during an earthquake can be minor, moderateand strong. Relatively speaking, minor shaking occurs frequently, moderate shaking occasionally and strong shaking rarely. For instance, on average annually about 800 earthquakes of magnitude 5.0-5.9 occur in the world while the number is only about 18 for magnitude range 7.0-7.9. So, should we design and construct a building to resist that rareearthquake shaking that may come only once in 500 years or even once in 2000 years at the chosen project site, even though the life of the building itself may be only 50 or 100 years? Since it costs money to provide additional earthquake safety in buildings, a conflict arises:Should we do away with the design of buildings for earthquake effects? Or should we design the buildings to be “earthquake proof” wherein there is no damage during the strong but rare earthquake shaking? Clearly, the former approach can lead to a major disaster, and the second approach is too expensive. Hence, the design philosophy should lie somewhere in between these two extremes.

Earthquake-Resistant Buildings

The engineers do not attempt to make earthquake-proof buildings that will not get damaged even during the rare but strong earthquake; such buildings will be too robust and also too expensive. Instead, the engineering intention is to make buildings earthquake-resistant; such buildings resist the effects of ground shaking, although they may get damaged severely but would not collapse during the strong earthquake. Thus, safety of people and contents is assured in earthquake-resistant buildings, and thereby a disaster is avoided. This is a major objective of seismic design codes throughout the world.

Earthquake Design Philosophy

The earthquake design philosophy may be summarized as follows (Figure 1):

(a) Under minor but frequent shaking, the main members of the building that carry vertical and horizontal forces should not be damaged; however building parts that do not carry load may sustain repairable damage.

(b) Under moderate but occasional shaking, the main members may sustain repairable damage, while the other parts of the building may be damaged such that they may even have to be replaced after the earthquake; and

(c) Under strong but rare shaking, the main members may sustain severe (even irreparable) damage, but the building should not collapse.

Figure 1: Performance objectives under different intensities of earthquake shaking – seeking low repairable damage under minor shaking and

collapse-prevention under strong shaking.

Thus, after minor shaking, the building will be fully operational within a short time and the repair costs will be small. And, after moderate shaking, the building will be operational once the repair and strengthening of the damaged main members is completed. But, after a strong earthquake, the building may become dysfunctional for further use, but will stand so that people can be evacuated and property recovered.

The consequences of damage have to be kept in view in the design philosophy. For example, important buildings, like hospitals and fire stations, play a critical role in post-earthquake activities and must remain functional immediately after the earthquake. These structures must sustain very little damage and should be designed for a higher level of earthquake protection. Collapse of dams during earthquakes can cause flooding in the downstream reaches, which itself can be a secondary disaster. Therefore, dams (and similarly, nuclear power plants) should be designed for still higher level of earthquake motion.

Damage in Buildings: Unavoidable

Design of buildings to resist earthquakes involves controlling the damage to acceptable levels at a reasonable cost. Contrary to the common thinking that any crack in the building after an earthquake means the building is unsafe for habitation, engineers designing earthquake-resistant buildings recognize that some damage is unavoidable. Different types of damage (mainly visualized though cracks; especially so in concrete and masonry buildings) occur in buildings during earthquakes. Some of these cracks are acceptable (in terms of both their sizeand location), while others are not. For instance, in a reinforced concrete frame building with masonry filler walls between columns, the cracks between vertical columns and masonry filler walls are acceptable, but diagonal cracks running through the columns are not (Figure 2). In general, qualified technical professionals are knowledgeable of the causes and severity of damage in earthquake-resistant buildings.

Figure 2: Diagonal cracks in columns jeopardize vertical load carrying capacity of buildings –unacceptable damage.

Earthquake-resistant design is therefore concerned about ensuring that the damages in buildings during earthquakes are of the acceptable variety, and also that they occur at the right places and in right amounts. This approach of earthquake-resistant design is much like the use of electrical fuses in houses: to protect the entire electrical wiring and appliances in the house, you sacrifice some small parts of the electrical circuit, called fuses; these fuses are easily replaced after the electrical over-current. Likewise, to save the building from collapsing, you need to allow some pre-determined parts to undergo the acceptable type and level of damage.

Acceptable Damage: Ductility

So, the task now is to identify acceptable forms of damage and desirable building behaviour during earthquakes. To do this, let us first understand how different materials behave. Consider white chalk used to write on blackboards and steel pins with solid heads used to hold sheets of paper together. Yes… a chalk breaks easily!! On the contrary, a steel pin allows it to be bent back-and-forth. Engineers define the property that allows steel pins to bend back-and-forth by large amounts, as ductility; chalk is a brittle material.

Earthquake-resistant buildings, particularly their main elements, need to be built with ductility in them.

Such buildings have the ability to sway back-and-forth during an earthquake, and to withstand earthquake effects with some damage, but without collapse (Figure 3). Ductility is one of the most important factors affecting the building performance. Thus, earthquake-resistant design strives to predetermine the locations where damage takes place and then to provide good detailing at these locations to ensure ductile behaviour of the building.

(a) Building performances during earthquakes: two extremes – the ductile and the brittle.

(b) Brittle failure of a reinforced concrete column

Figure 3: Ductile and brittle structures – seismic design attempts to avoid structures of the latter kind.

HOW BUILDINGS TWIST DURING EARTHQUAKES?

Why a Building Twists

In your childhood, you must have sat on a rope swing – a wooden cradle tied with coir ropes to the sturdy branch of an old tree. The more modern versions of these swings can be seen today in the children’s parks in urban areas; they have a plastic cradle tied with steel chains to a steel framework. Consider a rope swing that is tied identically with two equal ropes. It swings equally, when you sit in the middle of the cradle. Buildings too are like these rope swings; just that they are inverted swings (Figure 1). The vertical walls and columns are like the ropes, and the floor is like the cradle. Buildings vibrate back and forth during earthquakes. Buildings with more than one storey are like rope swings with more than one cradle.

Figure 1: Rope swings and buildings, both swing back-and-forth when shaken horizontally. The former are hung from the top, while the latter are raised from the ground.

Thus, if you see from sky, a building with identical vertical members and that are uniformly placed in the two horizontal directions, when shaken at its base in a certain direction, swings back and forth such that all points on the floor move horizontally by the same amount in the direction in which it is shaken (Figure 2).

Figure 2: Identical vertical members placed uniformly in plan of building cause all points on the floor to move by same amount.

Again, let us go back to the rope swings on the tree: if you sit at one end of the cradle, ittwists (i.e., moves more on the side you are sitting). This also happens sometimes when more of your friends bunch together and sit on one side of the swing. Likewise, if the mass on the floor of a building is more on one side (for instance, one side of a building may have a storage or a library), then that side of the building moves more under ground movement (Figure 3). This building moves such that its floors displace horizontally as well as rotate.

Figure 3: Even if vertical members are placed uniformly in plan of building, more mass on one side causes the floors to twist.

Once more, let us consider the rope swing on the tree. This time let the two ropes with which the cradle is tied to the branch of the tree be different in length. Such a swing alsotwists even if you sit in the middle (Figure 4a). Similarly, in buildings with unequal vertical members (i.e., columns and/or walls) also the floors twist about a vertical axis (Figure 4b) and displace horizontally. Likewise, buildings, which have walls only on two sides (or one side) and thin columns along the other, twist when shaken at the ground level (Figure 4c).

Figure 4: Buildings have unequal vertical members; they cause the building to twist about a vertical axis.

Buildings that are irregular shapes in plan tend to twist under earthquake shaking. For example, in a propped overhanging building (Figure 5), the overhanging portion swings on the relatively slender columns under it. The floors twist and displace horizontally.

Figure 5: One-side open ground storey building twists during earthquake shaking.

What Twist does to Building Members

Twist in buildings, called torsion by engineers, makes different portions at the same floor level to move horizontally by different amounts. This induces more damage in the columns and walls on the side that moves more (Figure 6). Many buildings have been severely affected by this excessive torsional behaviour during past earthquakes. It is best to minimize (if not completely avoid) this twist by ensuring that buildings have symmetry in plan (i.e., uniformly distributed mass and uniformly placed vertical members). If this twist cannot be avoided, special calculations need to be done to account for this additional shear forces in the design of buildings; the Indian seismic code (IS 1893, 2002) has provisions for such calculations. But, for sure, buildings with twist will perform poorly during strong earthquake shaking.

Figure 6: Vertical members of buildings that move more horizontally sustain more damage.

EARTHQUAKES EFFECTS ON REINFORCED CONCRETE BUILDINGS

bends with the beam but moves all columns at that level together.

After columns and floors in a RC building are cast and the concrete hardens, vertical spaces between columns and floors are usually filled-in with masonry walls to demarcate a floor area into functional spaces (rooms). Normally, these masonry walls, also called infill walls, are not connected to surrounding RC columns and beams. When columns receive horizontal forces at floor levels, they try to move in the horizontal direction, but masonry walls tend to resist this movement. Due to their heavy weight and thickness, these walls attract rather large horizontal forces (Figure 3). However, since masonry is a brittle material, these walls develop cracks once their ability to carry horizontal load is exceeded. Thus, infill walls act like sacrificial fuses in buildings; they develop cracks under severe ground shaking but help share the load of the beams and columns until cracking. Earthquake performance of infill walls is enhanced by mortars of good strength, making proper masonry courses, and proper packing of gaps between RC frame and masonry infill walls. However, an infill wall that is unduly tall or long in comparison to its thickness can fall out-of-plane (i.e., along its thin direction), which can be life threatening. Also, placing infills irregularly in the building causes ill effects likeshort-column effect and torsion.

Figure 3: Infill walls move together with the columns under earthquake shaking.

Horizontal Earthquake Effects are Different

Gravity loading (due to self weight and contents) on buildings causes RC frames to bend resulting in stretching and shortening at various locations. Tension is generated at surfaces that stretch and compression at those that shorten (Figure 4b). Under gravity loads, tension in the beams is at the bottom surface of the beam in the central location and is at the top surface at the ends. On the other hand, earthquake loading causes tension on beam and column faces at locations different from those under gravity loading (Figure 4c); the relative levels of this tension (in technical terms, bending moment) generated in members are shown in Figure 4d. The level of bending moment due to earthquake loading depends on severity of shaking and can exceed that due to gravity loading. Thus, under strong earthquake shaking, the beam ends can develop tension on either of the top and bottom faces. Since concrete cannot carry this tension, steel bars are required on both faces of beams to resist reversals of bending moment. Similarly, steel bars are required on all faces of columns too.

Strength Hierarchy

For a building to remain safe during earthquake shaking, columns (which receive forces from beams) should be stronger than beams, and foundations

Figure 4: Earthquake shaking reverses tension and compression in members – reinforcement is required on both faces of members.

(which receive forces from columns) should be stronger than columns. Further, connections between beams & columns and columns & foundations should not fail so that beams can safely transfer forces to columns and columns to foundations.

When this strategy is adopted in design, damage is likely to occur first in beams (Figure 5a). When beams are detailed properly to have large ductility, the building as a whole can deform by large amounts despite progressive damage caused due to consequent yielding of beams. In contrast, if columns are made weaker, they suffer severe local damage, at the top and bottom of a particular storey (Figure 5b). This localized damage can lead to collapse of a building, although columns at storeys above remain almost undamaged.

Figure 5: Two distinct designs of buildings that result in different earthquake performances –columns should be stronger than beams.

Relevant Indian Standards

The Bureau of Indian Standards, New Delhi, published the following Indian standards pertaining to design of RC frame buildings: (a) Indian Seismic Code (IS 1893 (Part 1), 2002) – for calculating earthquake forces, (b) Indian Concrete Code (IS 456, 2000) – for design of RC members, and (c) Ductile Detailing Code for RC Structures (IS 13920, 1993) – for detailing requirements in seismic regions.

EARTHQUAKE RESISTANT DESIGN TECHNIQUES

The conventional approach to earthquake resistant design of buildings depends upon providing the building with strength, stiffness and inelastic deformation capacity which are great enough to withstand a given level of earthquake-generated force. This is generally accomplished through the selection of an appropriate structural configuration and the careful detailing of structural members, such as beams and columns, and the connections between them.

But more advanced techniques for earthquake resistance is not to strengthen the building, but to reduce the earthquake-generated forces acting upon it.

Among the most important advanced techniques of earthquake resistant design and construction are:

• Base Isolation
• Energy Dissipation Devices

Base Isolation

A base isolated structure is supported by a series of bearing pads which are placed between the building and the building’s foundation. (See Figure 1.) A variety of different types of base isolation bearing pads have now been developed.

The bearing is very stiff and strong in the vertical direction, but flexible in the horizontal direction.

Figure 1: Base-Isolated and Fixed-Base Buildings

Earthquake Generated Forces

To get a basic idea of how base isolation works, examine Figure 2. This shows an earthquake acting on both a base isolated building and a conventional, fixed-base, building. As a result of an earthquake, the ground beneath each building begins to move. In Figure 2, it is shown moving to the left. Each building responds with movement which tends toward the right. The building undergoes displacement towards the right. The building’s displacement in the direction opposite the ground motion is actually due to inertia. The inertial forces acting on a building are the most important of all those generated during an earthquake.

It is important to know that the inertial forces which the building undergoes are proportional to the building’s acceleration during ground motion. It is also important to realize that buildings don’t actually shift in only one direction. Because of the complex nature of earthquake ground motion, the building actually tends to vibrate back and forth in varying directions.

Figure 2: Base-Isolated, Fixed-Base Buildings

Deformation and Damages

In addition to displacing toward the right, the un-isolated building is also shown to be changing its shape-from a rectangle to a parallelogram. It is deforming. The primary cause of earthquake damage to buildings is the deformation which the building undergoes as a result of the inertial forces acting upon it.

Response of Base Isolated Building

By contrast, even though it too is displacing, the base-isolated building retains its original, rectangular shape. It is the lead-rubber bearings supporting the building that are deformed. The base-isolated building itself escapes the deformation and damage–which implies that the inertial forces acting on the base-isolated building have been reduced. Experiments and observations of base-isolated buildings in earthquakes have been shown to reduce building accelerations to as little as 1/4 of the acceleration of comparable fixed-base buildings, which each building undergoes as a percentage of gravity. As we noted above, inertial forces increase, and decrease, proportionally as acceleration increases or decreases.

Acceleration is decreased because the base isolation system lengthens a building’s period of vibration, the time it takes for the building to rock back and forth and then back again. And in general, structures with longer periods of vibration tend to reduce acceleration, while those with shorter periods tend to increase or amplify acceleration.

Finally, since they are highly elastic, the rubber isolation bearings don’t suffer any damage. But the lead plug in the middle of our example bearing experiences the same deformation as the rubber. However, it generates heat. In other words, the lead plug reduces, ordissipates, the energy of motion–i.e., kinetic energy–by converting that energy into heat. And by reducing the energy entering the building, it helps to slow and eventually stop the building’s vibrations sooner than would otherwise be the case–in other words, it damps the building’s vibrations.

Energy Dissipation Devices

The second of the major new techniques for improving the earthquake resistance of buildings also relies upon damping and energy dissipation, but it greatly extends the damping and energy dissipation provided by lead-rubber bearings.

As we’ve said, a certain amount of vibration energy is transferred to the building by earthquake ground motion. Buildings themselves do possess an inherent ability to dissipate, or damp, this energy. However, the capacity of buildings to dissipate energy before they begin to suffer deformation and damage is quite limited. The building will dissipate energy either by undergoing large scale movement or sustaining increased internal strains in elements such as the building’s columns and beams. Both of these eventually result in varying degrees of damage.

So, by equipping a building with additional devices which have high damping capacity, we can greatly decrease the seismic energy entering the building, and thus decrease building damage.

Accordingly, a wide range of energy dissipation devices have been developed and are now being installed in real buildings. Energy dissipation devices are also often calleddamping devices. The large number of damping devices that have been developed can be grouped into three broad categories:

• Friction Dampers: these utilize frictional forces to dissipate energy
• Metallic Dampers : utilize the deformation of metal elements within the damper
• Viscoelastic Dampers : utilize the controlled shearing of solids
• Viscous Dampers: utilized the forced movement (orificing) of fluids within the damper

Fluid Viscous Dampers

General principles of damping devices are illustrated through Fluid Viscous damper. Following section, describes the basic characteristics of fluid viscous dampers, the process of developing and testing them, and the installation of fluid viscous dampers in an actual building to make it more earthquake resistant.

Damping Devices and Bracing Systems

Damping devices are usually installed as part of bracing systems. Figure 3 shows one type of damper-brace arrangement, with one end attached to a column and one end attached to a floor beam. Primarily, this arrangement provides the column with additional support. Most earthquake ground motion is in a horizontal direction; so, it is a building’s columns which normally undergo the most displacement relative to the motion of the ground. Figure 3 also shows the damping device installed as part of the bracing system and gives some idea of its action.

Figure 3: Damping Device Installed with Brace

Seismic Design Of Earth-Retaining Structures And Foundations

DESIGNS of retaining walls, earth dam, abutments, foundations, etc. are very important problems related to the geo-technical engineering. To minimize the devastating effect of earthquake on retaining structures, computation of earth pressures and the point of applications carries attention of the researchers. The most commonly used methods to design the retaining structures under seismic conditions are the force equilibrium based pseudo-static analysis, pseudo-dynamic analysis, displacement based sliding block methods. In the design of these retaining structures, knowledge of earth pressures under both active and passive conditions should be cleared to compute the earth forces and their point of applications. Design of earth dam under seismic condition is a challenge to the geo-technical engineers. Slope stability method is basically used for the analysis of earth dam using pseudo static approach for seismic forces. Shallow foundations are also designed under earthquake condition by computing the bearing capacity using pseudo-static methods available in literature. This paper reviews the different methodologies available to compute the active and passive earth pressures under seismic conditions for retaining wall, design of earth dam and shallow foundations and the mitigation techniques to the earthquake hazards are also discussed.

Economical Design Of Earthquake Resistant Bridges

With the occurrence of every major earthquake, there has been in the past, almost a world-wide tendency to increase the capacity demand of the structure to counteract such events. It is only in the last decade that new strategies have been successfully developed to handle this problem economically. The current international practice has shifted towards a performance-based engineering design, wherein the accent is on serviceability and safety under different levels of magnitude of earthquakes. Also there is an increasing realization that apart from techniques for improving ductility, the structural engineer’s tool-box should include energy-dissipating and energy-sharing devices and those that can control the response of the system. There have also been further advances on appropriate methods and devices of preventing ‘dislodgement’ or ‘unseating’ of the superstructure in the event of severe ground shaking. How these ideas can be used in economical earthquake resistant design of bridges is the subject of this paper.

PLASTIC HINGING AND DURABILITY

There is a marked difference in seismic design aspects of bridges and buildings. The reduced degree of indeterminacy of bridge structures leads to reduced potential of dissipating energy and load redistribution. In bridges, the superstructures (piers and abutments) are the main structural elements which provide resistance to seismic action. For energy dissipation, ductile behaviour is necessary during flexure of these structural elements under lateral seismic loads. This essentially means that the formation of plastic hinges or flexural yielding is allowed to occur in these elements during severe shaking to bring down the lateral design forces to acceptable levels. Since yielding would lead to damage, plastic hinging are localized by design at points accessible for inspection and repair, i.e., parts of the substructure that lie from foundation upwards (see Figure 1). No plastic hinges are, of course, allowed to occur in the foundations or in the bridge deck.

SOURCES OF EARTHQUAKE DAMAGE

Vibration of the structure in response to ground shaking at its foundation is the concern of the structural engineer, and which is taken into account by codal provisions of the different seismic-resistant design codes. However, these codes do not include any provision due to other effects, which may even exceed that due to vibration, as the procedure of their estimation and the needed steps for the design are outside the scope of the structural engineering discipline. Even then, it is essential that the structural engineer should be aware of the different seismic hazards in order to advise the client of potential damage involved in selecting sites in such zones. Hence, the first step in the design procedure of a future structure should be the analysis of the suitability of the site selected with proper consideration for the potential of any one of the following types of damage.

The different ‘Direct’ and ‘Indirect’ seismic effects are as follows

Direct effects:

• Ground failures, which include Surface faulting, Vibration of soil (or effects of seismic waves), Ground cracking, Liquefaction, Ground lurching, Differential settlement, Lateral spreading and Landslides.

(A) Damage due to surface faulting:

These damages to buildings and facilities along the fault scarps vary widely from completely demolished houses to rupture of the foundations, tilting of the foundation slabs and walls. Sometimes houses also have minor damage.

(B) Damage due to liquefaction:

The instability of the soil in the area affected by internal seismic waves can cause significant damage. The mechanical characteristics of the soil layers, the depth of the water table and the intensities and duration of the ground shaking influence the soil response. Deposits of loose granular materials if present in the site may be compacted by the ground vibrations induced by the earthquake. This will cause large settlement and differential settlements of the ground surface. Further, the compaction of the soil may result in the development of excess hydrostatic pore water pressures of sufficient magnitude to cause liquefaction of the soil, resulting in settlement, tilting and rupture of structures. The seismic-resistant design provisions of most codes only assure an effective design and construction of structures against damage due to the possible vibratory response of the structure to the shaking introduced at their foundation by the ground. However, it may not be possible to have success in all such cases. The only option remains in those areas is to prohibit the construction of building structures there.

(C) Damage due to ground shaking:

Integrated field inspection and post-earthquake analyses of structural damage due to earthquake shaking is one of the most effective means of having expertise knowledge on seismic response with a view to improving the state of the art and of the practice in seismic-resistant design and construction. Such integrated inspection and analyses revealed that in addition to the soil conditions mentioned above, the seismic performance of a structure is very sensitive to type of foundation; configuration of the structure; structural material; and design and construction detailing.

(D) Damage due to sliding of superstructure on its foundation:

One of the basic guidelines in the seismic-resistant design and construction of structures is that the whole structure-foundation system should work as a unit, and that the superstructure be tied or anchored properly to the foundation.

• Vibrations transmitted from the ground to the structure.

(E) Damage due to Structural Vibration:

I. Wood-Frame Houses:

The inertia forces develop during the vibratory response of a structure to earthquake ground shaking whose intensity depends on the product of the mass and acceleration. Hence, it is of the utmost importance to reduce the mass of the structure to a minimum. It is obvious that timber is the most efficient earthquake-resistant material for low-rise buildings among the traditional structural materials – timber, masonry, concrete, steel and aluminum. However, provision of proper lateral bracing and tying of all components together from the roof down to the foundation are to be followed.

II. Concrete structures:

Concrete is a comparatively heavy material and have a very good compressive strength. Due to its very small tensile and flexural strengths steel reinforcement is provided when used in structures. Such reinforced concrete can be used effectively in seismic-resistant construction. To overcome its relatively low strength per unit weight when normal weight aggregates are used, the use of lightweight aggregate concrete offers a significant advantage in seismic regions. members carefully: the proper amount and correct detailing of the reinforcing steel plays an important role in the seismic response of a reinforced concrete structure.

III. Steel structures:

Steel comes out of steel plants having excellent quality control. The stiffness per unit weight of steel is practically the same as of any other traditional constructional material. However, its strength, ductility and toughness per unit weight are significantly higher than concrete and masonry materials. Accordingly, the slenderness of steel structural members usually exceeds significantly that of other structural members. Hence, buckling becomes a serious problem. The danger of buckling becomes much more at higher yielding strength of the steel. Further, the plate elements used extensively to form the structural shapes are prone to local buckling, particularly when strained in the inelastic range. Accordingly, the compactness requirements for the cross section of the critical regions of structural members are more stringent in earthquake-resistant design than that of normal condition. Moreover, the field-connection of the structural members is another problem in achieving efficient seismic-resistant construction.

Indirect effects (or Consequential Phenomena):

i. Tsunamis

ii. Seiches

iii. Landslides

iv. Floods

v. Fires