U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
This magazine is an archived publication and may contain dated technical, contact, and link information.
|Publication Number: Date: March/April 1999|
Issue No: Vol. 62 No. 5
Date: March/April 1999
Recent major earthquakes in California and Japan have again demonstrated their potential for damage to highway bridges and loss of life. However, many of the bridges that failed were designed and built before the adoption of modern codes specifying earthquake-resistant design.
In the United States, it was after the devastating 1971 San Fernando Earthquake and at the direction of the Federal Highway Administration (FHWA) that the Applied Technology Council (ATC) formed a group to look at seismic bridge-design specifications and to recommend changes to improve bridge-design practices. These recommendations, which were published in report FHWA/RD-81/081 (commonly referred to as ATC6) were later adopted by the American Association of State Highway and Transportation Officials (AASHTO) as part of its "Standard Specifications for Highway Bridges" and by the California Department of Transportation (Caltrans). Since then, the bridge-design specifications have been amended several times as the lessons learned from new earthquakes were incorporated into design procedures.
In the United States, earthquakes are usually associated with California and Alaska. However, seismological data indicates that earthquakes can occur over broad areas east of the Rocky Mountains. The very intense New Madrid (Mo.) Earthquake in 1811 is an example of major seismic events in the central United States. According to seismologists, moderate earthquakes may reoccur in this region within the next 40 years.
Conventional Design Practices
During the past two decades, bridge engineers have gained a better understanding of the behavior of structures during an earthquake. Today, the concept of building stiffer bridges is no longer the preferred solution because added stiffness in bridge elements attracts more force in a seismic event. In general, the current design procedures rely not only on the strength but also on the ductility of piers to improve the seismic performance of new bridges.
Today, bridge piers are often built integrally with the superstructure. In this approach, engineers rely on plastic hinging in a controlled region of columns to dissipate seismic energy. This is achieved by good detailing and provision of sufficient shear or confinement reinforcement to increase ductility and to minimize seismic-related damage. This design philosophy also calls for preventing the failure of non-ductile elements and elements not easily inspected or repaired (e.g., piles). The ductility-design philosophy is the mainframe of modern seismic-design guidelines in the United States and most countries abroad.
While this design philosophy provides safety against collapse, it tends to be costly due to the damage induced in plastic-hinge zones and to the severe lateral displacements that can occur even in a moderate earthquake. Therefore, in recent years, design engineers have sought an alternative design philosophy that avoids or limits damage to a bridge to maintain post-earthquake serviceability. One approach that has been shown to be promising is the use of seismic isolation.
Retrofitting Existing Bridges
It is clear that existing bridges, designed without consideration of seismic forces, must be upgraded to current standards if public safety is to be improved. To achieve this goal, a number of retrofit measures have been used successfully in recent years. Cable restrainers, steel-column shell jacketing, pipe seat-extenders, and seismic isolation are just a few.
The first significant bridge seismic retrofit program was initiated by Caltrans after the 1971 San Fernando Earthquake. The so-called "Phase 1" program concentrated on preventing the superstructure from falling off its supports. A "Phase 2" program to improve the strength and ductility of existing bridge columns was launched in the late 1980s. In general, implementation of both phases provided good protection for bridges during the 1989 Loma Prieta and the 1994 Northridge earthquakes.
In 1995, FHWA published its Seismic Retrofitting Manual for Highway Bridges (FHWA-RD-94-052), replacing a 1983 report. This manual reflects the state-of-the-art knowledge that has been acquired through research and earthquake studies conducted since 1983. The retrofit manual is comprehensive in nature and nationally applicable to bridges in different seismic zones.
Lessons Learned From Recent Events
The 1994 Northridge Earthquake validated Caltrans' bridge-design philosophy of relying more on ductility rather than strength to resist large earthquakes. It also helped engineers to better understand the concept of isolation to reduce structural vulnerability.
After the 1995 Hanshin/Awaji (Kobe) Earthquake occurred in Japan, relatively little applicability to California bridges was found. This is due to differences in the design approach and the types of bridges that were damaged. The findings from this earthquake are mostly applicable to bridges in the central and eastern United States. Post-earthquake investigations showed that most damage was due to the presence of non-ductile columns and an earthquake force level that was much larger than anticipated, a situation that could well occur in the central and eastern United States.
Another important observation from the Kobe Earthquake is the similarity of the types of bridges damaged in Kobe to those of the central and eastern United States. These bridges are mostly steel stringer superstructures supported by steel bearings, reinforced concrete piers, and foundations that were designed with no consideration of seismic loads.
The sudden failure of steel bearings caused most of the damage to the superstructure. On the other hand, reports from Kobe also indicated that the premature failure of the steel bearings allowed the piers to move freely under the superstructure, thus significantly reducing the forces in the piers. This failure of the bearings acted as a safety valve to save the columns from shear and flexural failure and a number of spans from complete collapse. This fuse-type action points out the potential effectiveness of seismic isolation.
During the past 20 years, seismic isolation has emerged as one of the most promising retrofitting strategies for improving the seismic performance of existing bridges. It is also an attractive approach for new construction when conventional design is not suitable or economical.
Japan, New Zealand, and a number of European countries pioneered the use of seismic isolation in civil engineering structures. More recently, the United States has begun to implement isolation technology in bridges. In 1985, Caltrans became the first transportation agency in the United States to use seismic isolation on a bridge when it retrofitted the Sierra Point Overpass. The bridge's steel bearings were replaced with seismic isolation bearings (a lead-rubber isolation system) on the existing piers. Since then, approximately 90 U.S. bridges, many of them in the East, have incorporated this technology.
In the seismic isolation approach, the superstructure mass is uncoupled from seismic ground motions. This is also referred to as "superstructure" isolation. It uses special types of bearings called "seismic isolation bearings," which are placed below the superstructure and on top of the substructure (piers and/or abutments). Under normal conditions, these bearings behave like conventional bearings. However, in the event of a strong earthquake, they add flexibility to the bridge by elongating its period and dissipating energy. This permits the superstructure to oscillate at a lower frequency than the piers. It could also give rise to large relative displacements across the isolator interface; this can be controlled by incorporating damping elements in the bearing or by adding supplemental dampers.
This widespread use of seismic isolation strategy in the United States can be credited to the development of the AASHTO Guide Specifications for Seismic Isolation Design in 1991, as well as a partnership among academic researchers, seismic design engineers, and industrial entrepreneurs. However, there have been a few shortcomings in the guide specifications. One was its failure to incorporate systems other than elastomeric bearings because, at the time of its development, elastomeric systems were the only readily available isolation bearings on the U.S. market. This made it difficult for systems other than elastomeric lead-rubber bearings to be competitive. There was also the lack of sufficient data on the dynamic characteristics and the longevity of isolation systems then in use.
Also, most studies emphasized applications for buildings rather than bridges and included the testing of scale models rather than full-size bearings. There is a fundamental difference between bridges and buildings as bridges require multiple supports, the spacing of which range from a few to tens of meters. Recent earthquakes have demonstrated that even relatively close points on the soil surface can experience significant relative displacement. In addition, bridge bearings must withstand daily live loads and ambient environments that are much more severe than those in buildings. These concerns gave a sense of urgency to refining and updating the 1991 AASHTO Guide Specifications for Seismic Isolation Design, as well as to developing a testing program for evaluation and dynamic characterization of the different available systems.
In 1998, AASHTO approved the new Guide Specifications for Seismic Isolation Design, which, in addition to numerous modifications to the 1991 guide specifications, provides recommendations for the design of sliding isolation bearings as well as elastomeric bearings.
The Need for Testing
Bridge owners around the country were interested in opening the market to different manufacturers that offer a variety of isolation and energy dissipation systems, including elastomeric bearings, sliding systems, and others. There are certain advantages to each system; however, the performance of any device depends greatly on the proper design of all components and on the manufacturing quality. Due to the lack of an industry group that brings those suppliers together, there has been no uniformity of design standards, performance criteria, or even an agreement on prototype testing.
In addition, while several of the available systems have undergone static and dynamic testing, the testing often consisted of full-scale pseudo-static tests or reduced-scale shake-table tests. Some dynamic characteristics can be established from static behavior, but a full-scale dynamic test is the only true test of the response of any of these systems to an earthquake. Full-scale tests are essential to develop the higher confidence levels among bridge designers that will lead to a wider application of seismic isolation in the United States.
The HITEC Evaluation Program
In January 1994, FHWA, Caltrans, and the Highway Innovative Technology Evaluation Center (HITEC) cooperated in launching a full-scale, dynamic testing and evaluation program of readily available seismic isolation and energy dissipation systems. FHWA provided financial assistance through the Applied Research and Technology (ART) program; Caltrans' engineering staff provided technical support; and HITEC administered the program.
HITEC is a non-profit organization established under an agreement between FHWA and the Civil Engineering Research Foundation (CERF), a subsidiary of the American Society of Civil Engineers (ASCE). The missions of HITEC are to evaluate new products, materials, and equipment for which industry standards or specifications do not exist and to work towards overcoming barriers to innovation. In fulfilling its missions, HITEC facilitates the conduct of national consensus-based evaluations using public highway agencies and other organizations.
A total of 11 domestic and international manufacturers participated in and completed the HITEC testing program. A test plan was developed with the guidance of a panel appointed by HITEC. The panel was comprised of experts representing several state departments of transportation, FHWA, university researchers, and private industry.
The objectives of the evaluation program were to:
This program examined the important properties and performance characteristics of all systems — stability, range, capacity, resilience, resistance to service and dynamic loads, energy dissipation, functionality in extreme environments, resistance to aging and creep, predictability of response, fatigue and wear, and size effects.
These properties provide the engineer with critical information on the suitability of these devices for specific applications. Furthermore, the program addresses the ability of the vendor or manufacturer to provide a quality system and to understand and predict system response.
The isolation and energy dissipation systems submitted for testing were all passive. This means they require no energy input or mechanical interaction with an outside source.
Table 1 presents the systems' categorization and their respective manufacturers.
Table 1 — Systems' Categorization
|Dynamic Isolation Systems Inc. Scougal Rubber Corp.
Skellerup Industries Ltd.
|1- Lead-Rubber Isolation Bearings
2- High Damping Rubber Bearings
3- Lead-Rubber Isolation Bearing
4- Steel Rubber Bearing
R.J. Watson, Inc.
|5- Slider Bearings
6- Eradiquake (Slider Bearings)
Earthquake Protective System Inc.
|7- Roller Bearing
8- Friction Pendulum Bearing
|9- Viscous Shear Type Damper
10- Hydraulic Damper
11- Hydraulic Damper
Each manufacturer submitted five test articles for testing and evaluation. Each article had predetermined design parameters specified by the HITEC plan. (See tables 2 and 3.)
Table 2 — Isolator Requirements
|Design Comprehensive Load (+10%)(DCL)||Number of Devices||Test Article Number||Lateral Design Disp. (DD)||Minimum Movement Rating (MR)|
|150 kips*||1||1||6 in**||2.0 in|
|500 kips 3||3||2,3,4||9 in||3.0 in|
|750 kips||1||5||12 in||4.0 in|
|* 1 kip = 4450 newtons ** 1 in = 25.4 millimeters|
Table 3 — Energy Dissipator Requirements
|Design Rating (+10%)(DR)||Number of Devices||Test Article Number||Axial Design Disp. (DD)||Movement Rating (MR)|
|50 kips*||1||1||<12.0 in**||2.0 in|
|150 kips||3||2,3,4||<12.0 in||3.0 in|
|240 kips||1||5||12.0 in||4.0 in|
|* 1 kip = 4450 newtons ** 1 in = 25.4 millimeters|
The testing for the various systems included a combination of the following tests:
Testing Facility and Its Capacity
HITEC's full-scale testing program required testing of systems at frequencies of up to 2.0 Hz with large displacements. This demanded a major source of energy, which was not available to most laboratories. The laboratories of the Energy Technology Evaluation Center (ETEC) had the necessary capabilities to perform the proposed testing and were selected. ETEC is owned by the Rocketdyne Division, which is a subsidiary of Boeing North America Corp.
The isolator test rig was originally developed for the dynamic testing of seismic isolation bearings for nuclear power plants. The test rig has the following capacities:
Program StatusThe HITEC testing program was completed earlier this year, and 14 reports will be published. The results for each device tested under the seismic evaluation process will be covered in separate reports, and together these will account for 12 of the 14 reports. One report is a summary of the evaluation findings; the summary synthesizes the performance of all systems submitted for evaluation and provides basic knowledge of seismic isolation and energy dissipation. A test system overview report describes the test system and methodologies used to obtain isolator characterization.
The eight individual reports for isolators have been published by HITEC. The drafts of the three damper reports, the summary report, and the overview report have been completed, and the published reports are expected to be available by April 1999.
Table 4 — Dynamic Performance Characteristics at Temperature Extremes (2nd Cycle) for the Skellerup Bearing (500-Kip Unit)
|Performance Parameters||Cold Temperature
49 hrs @ -20
70 degrees F*
23 hrs @ 120 degrees F*
|17.0 (+56%)**||10.9||10.4 (-5%)**|
|EDC [in-kips]||2900 (+44%)**||2004||1777.0 (-11%)**|
|*-20 degrees F = -29 degrees C 70 degrees F = 21 degrees C 120 degrees F = 49 degrees C
** indicates percent change from ambient temperature test results
Table 5 — Dynamic Performance Characteristics at Temperature Extremes (2nd cycle) for the EPS Bearing (500-Kip Unit)
|Performance Parameters||Cold Temperature
22 hrs @ -40 degrees F
|Ambient Temperature 70 degrees F||Hot Temperature 24 hrs @ 120 degrees F|
|Stiffness (kips/in)||7.9 (+0%)**||7.8||7.1 (-9%)**|
|23.9 (-6%)**||25.5||22.9 (-10%)**|
|EDC (in-kips)||1044 (+8%)**||968.8||917 (-5%)**|
|*-40 F = -40 degrees C 70 F = 21 degrees C 120 F = 49 degrees C
** indicates percent change from ambient temperature test results
The HITEC testing program is expected to have a major impact on the use of seismic isolation systems nationwide. It will increase the confidence level of bridge owners and will lead them to consider and use isolation/damping technology cost-effectively to protect otherwise vulnerable structures from more severe earthquake damage. In addition, it will provide an invaluable resource and useful data to academic researchers.
Several states are just beginning to retrofit structures against seismic forces, and seismic isolation may very well be a viable strategy for them. It is certain that seismic isolation has reached a level of maturity in the United States that can provide earthquake hazard mitigation for the entire nation.
Dr. Hamid Ghasemi is a research structural engineer in FHWA's Office of Research, Development, and Technology. He was instrumental in developing the HITEC evaluation of the seismic isolation and energy dissipation test plan and in analyzing the test results. In addition, he was responsible for writing the HITEC technical summary report of the findings and a portion of the individual reports. His research interests are in the area of seismic isolation, modal analysis, and testing. slide (use as title art or somewhere on first page of article): the hanshin/awaji earthquake on jan. 17, 1995, near kobe, japan, caused the collapse of an 18-span viaduct section of the hanshin expressway. this section was reconstructed with reinforced concrete piers using seismic isolation bearings to distribute forces and dissipate energy.