Retrofitting Techniques for Bridges and Flyovers

Dr. S.K.Thakkar, Ex. Professor Railway Chair and Professor of Earthquake Engineering, Indian Institute of Technology, Roorkee.

Abstract

The extensive damage to bridges world over in earthquakes has generated considerable interest amongst the engineers and researchers on the seismic design of new bridges and retrofitting of existing ones. Prior to the development of modern codes, the bridges were designed for nominal seismic forces without provisions of ductility. Seismic codes began to incorporate ductility provisions only in 1970. Thus the bridges designed earlier may be deficient to withstand effect of future earthquakes. These structures may require seismic evaluation and retrofitting. The protection of bridges from damage in earthquakes has direct relevance in prevention of aggravation of disaster, as disruption of transportation routes hampers relief and rescue operations immediately following the earthquake. Several retrofitting techniques have been developed for bridges in recent years. Many of the existing bridges can be retrofitted by these techniques. Some seismic countries have undertaken the programmes of seismic assessment and retrofitting of bridges on a bigger scale. Such actions are proving to be effective as evidenced by performance of retrofitted bridges in recent earthquakes. There is need in India to investigate the likely behavior of existing bridge stock in future earthquakes and initiate seismic retrofitting. This paper highlights the important issues: structural deficiencies; retrofit philosophy, retrofit techniques, recent developments and effectiveness of retrofit techniques

Introduction

The extensive damage of bridges all over the world in recent earthquakes has been the motivation in significant advancement in the earthquake resistant design and retrofitting of bridges. The bridges constructed prior to 1970 were not designed for adequate seismic resistance as the ductility provisions were not incorporated in the seismic codes till then. As a result the bridges designed before this year lack in earthquake resistance and ductility and may be vulnerable to significant damage even from moderate earthquakes. The post earthquake damage surveys in recent earthquakes have confirmed this view. It has also been revealed from damage surveys that many of the damages that occurred in bridges and flyovers could be prevented by proactive measures of seismic retrofitting prior to earthquakes. Many of the reinforced concrete piers designed earlier had inadequate shear capacity due to lack of transverse steel and confinement, inadequately lapped longitudinal steel, and premature termination of longitudinal steel. The superstructures were vulnerable to falling down in the absence of restraining devices; bearings were deficient in accommodating large seismic displacement and bearing seat was inadequate. These deficiencies had an adverse impact on the performance of bridges. An existing bridge can be replaced by a newly designed bridge to meet earthquake demands or upgraded in its strength by appropriate retrofitting measures. The retrofitting is often an economical alternative than replacement. The retrofitting of bridges has received considerable attention in recent years because need of their safe operations in post-earthquake scenario for relief and rescue operations. There are two types of situations that require retrofitting in bridges, (i) the existing bridges that are deficient to meet requirements of current codes but are vulnerable to damage; these bridges have not yet experienced even moderate earthquakes, (ii) the existing bridges that are damaged in earthquakes. The bridges in the later category require both repair and retrofitting. There are many retrofitting techniques developed for upgrading earthquake resistance in bridges. The techniques are equally applicable to bridges damaged in earthquakes. The principal issues facing the retrofitting are: evaluation of seismic capacity of existing bridges, identifying structural deficiencies, ageing effects, decision on level of retrofitting, hazard levels for design, performance criteria, developing alternate strategies of retrofitting, choice of right strategy, re-evaluation of retrofitted structure, validating effective retrofitting measures and health monitoring of retrofitted bridges. There are not many codes and guidelines developed for the retrofitting of bridges. There are many issues that still need more research and development. The objective of the paper is to present an overview of retrofitting techniques and discussion on many of the issues highlighted above.

Earthquake Damages of Bridges

In order to design suitable retrofit measures, it is appropriate to highlight common bridge damage types and failures that has occurred in past earthquakes in India. These types of failures are to be often required to be dealt with in retrofitting of bridges. The common types of earthquake damages of bridges in India are briefly described below:

Falling down of bridge spans: The bridge spans can drop down from their supports. There are few examples of loss of span types of failures in past earthquakes. The Inchape Bridge over toppled from its supports in Bihar earthquake of 1934. A multispan masonry arch bridge collapsed in Koyna earthquake of 1967. A single span steel truss over toppled from its supports in Uttarkashi earthquake of 1991. There are many examples of dislodgement of spans on the bearings.

Bearing failure: There are many examples of bearing failures in bridges in India. To mention a few, the holding down bolts of fixed bearings sheared off in one fixed bearing in a railway bridge in Broach earthquake of 1970. There was failure of bearings in Silcher in Cachar earthquake of 1984 (Arya and Thakkar, 1992). There was failure of elastomeric bearings in Surajbari Bridge in Bhuj earthquake of 2001.

Expansion joint failure: There are examples of expansion joint failure in Kaliabhomora Bridge at Tejpur in Assam earthquake of 1988. There are examples of expansion joint failures in Bhuj earthquake of 2001 (Thakkar et al, 2001).

Substructure damage: There are many examples of substructure damage in earthquakes in India. To mention a few, the twin RC piers of a bridge became inclined and cracked in Cachar earthquake of 1984. There are examples of damage to substructures of bridges in the form of cracking and spalling of concrete in Bhuj earthquake of 2001.

Liquefaction failure: There are examples of liquefaction of sand in Bihar earthquake of 1934, Bihar-Nepal earthquake of 1988 and in river beds in Bhuj earthquake of 2001. The effect on bridges has not led to catastrophic failures.

Structural Deficiencies

The observation of performance of bridges in past earthquakes world over has highlighted following deficiencies in bridges:

Superstructure: Traditionally there is no linkage provided between two adjacent spans in the case of multi–span simply supported bridges as a result spans are dislodged from supports due to out of phase motion in piers or bearing failures. The bearing failure occurs in fixed bearings when these are unable to withstand the seismic force generated in the superstructure. In many cases, spans fall down from their supports resulting into irrepairable damage. Superstructure deficiency is also associated with inadequate seat length at expansion joints on the supports or at the abutments resulting into unseating of span.

Bearings: The traditional rocker and roller bearings have not shown satisfactory performance in earthquakes. There have been problems of jumping, inadequacy of bearings in accommodating displacements.

Inadequate seat width on supports: The inadequacy of seat width on bearing supports or at expansion joints have caused unseating of span.

Substructure: The various types of deficiencies observed in RC columns and piers are (i) Lack of flexural strength, (ii) Lack of shear strength (iii) Insufficient transverse reinforcement and confinement (iv) In adequate lap splicing of longitudinal steel (iv) In adequate ductile detailing in plastic hinge region of columns (iv) Premature termination of longitudinal steel in piers (v) Insufficient strength of joints between pile cap beams.

Inadequacy of foundation and soil strength: Liquefaction of soil often results in damage due to unequal settlements and loss of span types of failures. Abutment failure can occur due to increased earth pressure; abutment slumping is known to occur in soft soils. Inadequate strength of footings, wells and/or piles can result in foundation failures.

Retrofit Philosophy

Design Philosophy: The design philosophy for retrofitting should normally conform to that for new bridge design. This is the minimum design performance expected from a retrofitted structure. The design philosophy for new structures is as follows:

1. The structure is designed to resist Design Basis Earthquake (DBE) with only minor damage, which should be repairable.
2. The structure is designed to resist Maximum Considered Earthquake (MCE) with some structural damage but controlled so as to prevent collapse.

Higher Level of Performance: A higher degree of performance can however be specified to control damage such as the availability of vital communication route immediately after the earthquake. This will restrict the occurrence of category of damage specified, such as loss of girder supports, collapse of substructure and liquefaction failure of foundation soil. The purpose of this level of retrofit is to prevent collapse, as well as to provide serviceability after a major earthquake. The performance level is to be specified prior to retrofit design.

Decision to Retrofit: There are two basic decisions required at the beginning of under taking retrofitting of bridges (i) whether to retrofit or not (ii) level to which bridge should be retrofitted. The first decision should be based on results of detailed seismic evaluation and level of risk. The second decision may involve reduction of seismic risk by incorporating retrofitting measures such as extending seating width and/or providing restrainers. The additional reduction in risk is possible by providing other types of seismic devices such as energy dissipating devices/dampers, and base isolation bearings between super and substructure. In case of substructures jacketing for increase in strength and ductility and detailing of plastic hinge region shall be specially required.

Retrofitting Steps

The major steps in retrofitting of bridges are as follows:

Preliminary Screening: The seismically deficient bridges are identified by preliminary screening. The screening procedure is mainly based on (a) Seismicity (b) Vulnerability and (c) Importance. The prioritization of bridges to be retrofitted can be made on the basis of rating procedures based on above factors.

Detailed Seismic Assessment: The detailed seismic evaluation of expected performance of existing bridge is necessary to determine seismic capacity, weaker sections and mode of failure. The strength evaluation can be made according to codes following dynamic methods of analysis and push over analysis techniques. It is normally desired that assessment procedure should be more precise than code. A number of prioritization methods have been developed, ATC (1983), and Kawashima (1991).

Selection and Design of Retrofit Measures: On the basis of detailed seismic assessment, it should be determined whether individual component level retrofit such as extending seating width, providing restrainer or a global retrofit of complete bridge is to be undertaken. The global retrofit may include: jacketing of bridge piers, replacement of bearings and retrofit of foundations.

Re-analysis of Retrofitted Structure: The retrofitted structure should be re-analyzed using dynamic analysis. The checking of design of retrofitted structure should be based on current design codes. The retrofit techniques can also be tested on model/prototype components in laboratory under cyclic loading tests.

Retrofitting Techniques

The retrofitting of bridge may be required in each of the components: superstructure, bearings, substructure and foundations. The extent of retrofitting will be based on detailed seismic assessment following dynamic methods of analysis. The retrofitting techniques for various portions of bridge are described below:

Superstructure

Superstructures will not normally need strengthening due to their high rigidity and large flexural capacity. Superstructure deficiency is associated with their unseating at expansion joints or on bearing supports due to relative displacements. The most commonly observed type of failure in superstructure of simple supported bridges is a girder falling off the supports due to longitudinal response. To prevent such type of failure, the adjoining spans should be interconnected by connection rods/restrainers/linkage bolts (Park, 1993).

A typical example of linkage and tying of spans by restrainer, ATC (1983) is given in Fig. 1 and 2. An example of vertical restrainer retrofit is shown is Fig. 3. The unseating prevention for steel girders is often made by tying webs together with steel plates over the supports. (Keady, Alameddine and Sardo, 2000).

Bearings

Bridge bearings are one of the most vulnerable components in resisting earthquake. Bearing deficiencies are associated with (i) inadequate seat width and (ii) inadequacy to accommodate displacements in earthquakes. The possible retrofit solutions are: (i) replacing steel bearings, by (a) elastomeric bearings (b) base isolation bearings, (ii) bearing seat extension (iii) provision of stoppers and devices to prevent jumping of girders. Some devices have been recommended by ATC (1983). A typical example of replacement of bearing is shown in Fig. 4.

Substructures

The RC columns in substructures are commonly deficient in flexural strength, shear strength and ductility. In addition, these deficiencies could be due to lack of transverse reinforcement in the plastic hinge region leading to inadequate confinement. There could be other detailing deficiencies such as premature termination of longitudinal steel in cantilever piers, inadequate anchorage and inadequate splicing of longitudinal reinforcement. The common retrofitting techniques for such deficiencies are: steel jacketing, concrete jacketing and composite material jackets:fiberglass, fiber reinforced plastic (FRP), pre-stressing steel and carbon fiber. The jacketing normally increases flexural strength of piers. The increase of flexural strength of piers tends to increase the seismic force transmitted from piers to the foundation. Therefore, there should be controlled increase in flexural strength of piers when jacketing technique is adopted. The jacketing techniques are briefly described.

Steel Jacketing

The circular columns are retrofitted by circular steel jackets while rectangular columns are retrofitted by elliptic or rectangular jackets, the former is more effective. In the case of circular columns, two half shells of steel plate are rolled to a radius of 12.5 to 25 mm bigger than the column radius, Fig 5 (a) and (b), (Priestley, 1996), and are site welded up the vertical seams. The steel shall provide a continuous tube around the concrete column with an annular gap. This gap is grouted with cement grout or epoxy resin. A small gap of about 50 mm is provided at the bottom of piers between steel jacket and the top of the footing. This gap enables jacket to function as a passive confinement and prevents excessive increase in the flexural strength. This jacket primarily increases shear strength and confinement. The increase of thickness of column after retrofit is very small.

Concrete Jacketing

This type of jacket uses a thick layer of reinforced concrete around the column. The RC jacket increases flexural strength, shear strength and ductility of column. The longitudinal reinforcement should be dowelled with adequate anchorage in the footing to develop flexural strength. The flexural strength of column is increased (Unjoh, 2000) but it should be accompanied by footing retrofit so as to ensure that plastic hinge forms in column and not in footing. A typical method of retrofitting using RC jacket is shown in Fig. 6 (Priestley, 1996).

The confinement of circular columns can be achieved by using closely spaced hoops or spiral reinforcement. In the case of rectangular column, circular or elliptic jackets are more effective than rectangular jackets. In the later form of jacket longitudinal bars are susceptible to buckling in the middle region with the type of reinforcing shown in Fig 6b (Priestley, 1996). The concrete jacketing has an advantage of cost for construction. This method increases thickness of column by 0.50 m. This type of jacketing increases the flexural strength of column also.

Composite Material Jackets

The use of Advanced Composite Materials (ACM) is receiving prominence in retrofitting because it involves lesser effort; these materials have favorable mechanical properties, low weight, high strength and easier in construction. The procedure involves wrapping layers of thin, flexible straps or sheets of fiber composites around the column in the plastic hinge zone or along the entire height of column, (Saadatmanesh, 1997), Fig. 7.

The composite materials in use are fiberglass, fiber reinforced plastic (FRP), carbon fiber sheets and aramid fiber sheets. These techniques are used where enough construction space is not available (Priestly, 1996, Kawashima 2000, Unjoh, 2000). The composite jackets are effective in improving strength and ductility of columns. Techniques are found to be effective as evidenced by observed in laboratory testing. Design methods are not fully developed. There is a lack of knowledge on performance of retrofitted bridge with these techniques in major earthquakes.

Foundation

The retrofit of foundation, may involve high costs, is normally avoided. The rocking and uplift of foundation though undesirable is often considered as a form of isolation and may reduce seismic forces in the bridge superstructure and substructure. While ascertaining retrofitting needs in foundation, detailed seismic analysis should be carried out. However, if analysis and re-design shows deficiency in foundation, retrofitting of foundation is carried out. The retrofitting option available for footing is the enlargement of footing size. The retrofitting option for pile foundation may consist in addition of piles and then integrating these with existing piles by extending pile caps.

Recent Advances in Retrofitting Techniques

There have been developments in retrofitting using newer materials and innovative technology some developments include: seismic base isolation, structural control, smart materials and viscous dampers. These techniques are briefly described below:

Seismic Base Isolation

The seismic base isolation (SBI) method of retrofitting consists in replacement of conventional bearings by base isolation device. The isolation bearings elongate the natural period of bridge from a typical value of less than one second to 3 second or more. This will result in significant reduction of earthquake-induced response. The force reduction could be of the order of 3 to 8 from those of conventional fixed bearings. The base-isolation devices will result in increase in displacement response. To counter this effect some kind of damping device is provided externally or internally within the bearing. The devices should be designed according to principles of isolation. The available options of isolation system are: elastomeric bearing with external damper, high damping rubber bearing, lead rubber bearing and friction pendulum system (FPS). A typical seismic base isolation system of bridge using lead rubber bearings is shown in Fig. 8. A judicious choice may be necessary based on the benefits of isolation and additional costs of such devices.

Structural Control

The structural control method consists in reduction of vibrations in the system by design employing passive or active methods. The passive methods include installing isolators to isolate ground motions (SBI) and/or adding supplemental damping or passive energy dissipating devices (PED) for reduction of dynamic response. The active method consists in applying external forces to the system by actuators based on the feedback of the system. The active control system has better control and adaptivity than passive system, thus offers greater promise in bridge engineering (Zaiguaing Wu, 2000)

The passive methods have their own merits and limitations. The limitations are mainly for nearfield earthquake motions and for structures with longer natural periods of vibration. In active control system, external power is required for control actuator(s) to apply forces to counter the effect of dynamic loads. The control forces can both add or dissipate energy in the structure. The feed back signals of the system response received from physical sensors mounted on the system are sent to control actuators, which can generate necessary forces for reduction of response. The active control system has better control but it suffers from the disadvantage of requiring significant power to run actuators at the time of earthquake. Active mass dampers are one of the applications of active control method that can be used for bridges. The semi-active devices, which require much less external power as compared to active control, offer promise for applications in bridge structures.

Smart Materials

The smart materials have unusual thermo mechanical properties that have been explored for purpose of earthquake protection and retrofitting measures in structures. These materials also called intelligent materials have selfrepairable and self-diagnosis characteristics. Shape memory alloys (SMAs) are one kind of materials, which has found application in bridge structures. The SMA materials differ from ordinary materials in that in the former, heating the material can change the crystal structure. These materials can have phase transformation above a transformation temperature range that is specific to various alloys. One such material is use is Nitinol, which has shape memory effect and super elastic effect. Shape memory effect permits having pseudo yield and pseudo elastic deformation. The pseudo-elastic deformation remains as residual strain upon unloading but can be removed upon heating. This is called shape memory effect. SMAs also possess super elastic characteristics. After pseudo yield the deformation can be recovered simply by unloading. This is called super elastic effect. The high elastic strain capabilities are of the order of 6-8%. The shape memory and super elastic effect depend on the temperature: In the lower range they show shape memory effect but in 20 to 300C higher than transformation temperature these materials show super elastic effect. The transformation temperature is a function of alloy compound. SMAs can act both as restrainers and dampers. SMAs can dissipate significant amount of energy due to shape memory effect (Andrawes and Desroches, 2004). One application of SMAs is in the use of restrainers. These devices are installed between adjoining spans to limit excessive relative displacements. Traditionally steel restraining cables have been used for this purpose. The disadvantages of traditional devices are, (i) these have no re-centering capability (ii) the devices increase ductility demand of structure. Both the drawbacks of traditional devices are overcome in restrainers made of SMA alloys.

Viscous Dampers

The viscous dampers are installed between deck and pier and function as an energy dissipating devices. These devices have been employed as retrofitting devices in bridges. Twelve bridges have been retrofitted in South Korea using this technology. A viscous fluid damper generally consists of a piston in a damper housing filled with a compound of silicone or oil. It dissipates energy through movement of the piston in the highly viscous fluid. If the fluid is purely viscous then damping force is directly proportional to the velocity of the piston. The device can cause significant dissipation of energy. The viscous dampers are reliable, easy in installation and efficient in operation.

Effectiveness of Retrofit Techniques

The success of retrofit scheme depends on how effective it is in enhancing strength and ductility of the member/structure. The real test of retrofit schemes is the observation of performance of such structures in earthquakes. The effectiveness of retrofit schemes is studied by conducting cyclic test on model/prototype components in the laboratory. The study of hysteretic behavior under cyclic test of both original and retrofitted structure reveals the increase in stiffness, strength and ductility. The stable hysterisis loops show overall improvement in the performance. Such tests are performed in quasi-static testing laboratory. Shake table testing of original and retrofitted structure is more rational method of testing effectiveness of retrofit measures.

Research and Development Issues

There are significant advances in seismic retrofitting of bridges, yet there are issues that need further research for effective use of retrofitting methods. The following issues need further research:

1. Seismic assessment: Development of seismic assessment techniques of existing bridges, particularly those bridges, which have been damaged in earthquake.
2. Level of retrofitting: The level to which retrofitting should be done for DBE and MCE seismic conditions. This is related with retrofit philosophy adopted for bridges.
3. Validation techniques: The validation techniques could be analytical and/or experimental. The experimental techniques include: quasi-static cyclic load testing techniques, shake table testing and pseudo dynamic testing and observation of performance in earthquakes.
4. Structural control techniques: The passive methods such as base isolation, passive energy dissipating devices like dampers and active methods are to be explored further for their application in retrofitting.
5. Smart materials: Application of smart materials such as SMAs is to be studied further for their greater application in retrofitting as restrainers and dampers. Research in structural health monitoring and damage detection in hidden regions of foundations and superstructures relevant to bridges is necessary as its benefits have direct bearing in maintenance and retrofitting. Research on Non Destructive Testing (NDT) for status determination of structures is important.
6. Design procedures: Design procedures of retrofitting using advanced composite materials are required to be developed.

Conclusion

The bridges need retrofitting primarily because of two reasons (i) these were designed for smaller forces than that can occur, and (ii) these may be lacking in ductility in the absence of ductile detailing of reinforcement. The structural deficiencies in many of the existing bridges have been observed from seismic behavior in recent earthquakes. Several bridges constructed prior to existence of modern seismic codes fail to meet requirement of safety. Seismic retrofitting is required for protection of such bridges in future earthquakes. It is possible to retrofit many of the existing bridges against falling of spans and other distress by simple retrofitting measures. The key issues in retrofitting are: retrofit philosophy, seismic assessment, retrofit techniques, validating effectiveness of retrofit measures, application of composite materials and smart materials, in retrofitting. The bridge retrofit program is a necessity for the country and should be undertaken for vulnerable bridges that have not yet experienced earthquakes. The retrofitting of existing bridges has direct relevance towards mitigating disaster caused by earthquakes.

References

• Andrawes, Bassem and Desroches, Reginald (2004) Comparison of Different Methods for Seismic Retrofit of Bridges Using Smart Materials, 13 WCEE, Vancouver, B.C. Canada, Aug. 1-6, paper no. 274.
• Arya, A.S., Thakkar, S.K., and Bakir, S.M. (1992), Retrofitting of an Earthquake Damaged Reinforced Concrete Bridge, 10WCEE, July 19- 24, Madrid, Spain.
• ATC (1983) Seismic Retrofitting Guidelines for Highway Bridge, Report ATC-6-2, Applied Technology Council, Palo Alto, California, August.
• Kawashima, K., Ichimasu, H. and Ohuchi, H. (1991), Retrofitting of Bridges, Proceedings, InternationalWorkshop on Seismic Design and Retrofitting of Reinforced Concrete Bridge, Bormio, Italy, April, pp. 471-501.
• Keady, Kevin, I., Alameddine, Fadel,and Sardo, Thomas, E. (2000), Seismic Retrofit Technology, Chapter 11, Bridge Engineering, Seismic design, Edited by Wai-Fah Chen and Lian Duan, CRC Press.
• Park, R. and Rodriguez, M.E. (1993), Assessment and Retrofit of a Reinforced Concrete Bridge Pier for Seismic Resistance, Earthquake Spectra, Vol. 9, No. 4.
• Priestley, M.J.N., Seible, F, and Calvi, G.M. (1996), Seismic Design and Retrofit of Bridges, John Wiley & Sons.
• Saadatmanesh, H., Ehsani, M.R., and Jin, L., (1997), Seismic Retrofitting of Rectangular Bridge Columns with Composite Straps, Earthquake Spectra, Vol. 13, No. 2, May.
• Thakkar, S.K., Dubey, R.N. and Agarwal, Pankaj (2001), Behavior of Buildings, Bridges and Dams in Bhuj Earthquake of January 26, 2001, Proceedings of 17th U.S. Japan Bridge Engineering Workshop. Nov.12-14, Tsukuba City, Japan.
• Unjoh, Shigeki (2000), Seismic Design Practice in Japan, Chapter 12, Bridge Engineering, Seismic design, Edited by Wai-Fah Chen and Lian Duan, CRC Press
• Wu, Ziaguang (2000), Active Control in Bridge Engineering, Chapter 13, Bridge Engineering, Seismic Design, Edited by Wai-Fah Chen and Lian Duan, CRC Press.