1. Introduction
Application of seismic isolation technology has seen a rapid increase in Türkiye in recent years, particularly in the case of buildings. Following the completion of the Turkish national code for seismic design of buildings [1] in 2018, in which a chapter is devoted specifically to seismically isolated buildings, the General Directorate of Highways of Türkiye launched a similar effort to develop specifications for the seismic design of highway and railway bridges and other lifeline structures as well as for seismically isolated highway and railway bridges. This article describes the distinct features of the newly developed Turkish seismic isolated bridge design specifications, as part of the Turkish Bridge Design Standard [2] and compares them to the AASHTO Guide Specifications for Seismic Isolation Design of Bridges [3] and EN 15129 [4]: European Standard on Anti-Seismic Devices. These two documents were referenced in the development of the new standard, along with ASCE 7-16 [5] and other well-known national codes for seismic isolation of bridges and buildings.
Murat Dicleli
Professor and Chair
Department of Engineering Sciences, Middle East Technical University
Ankara, Türkiye
2. Classification of Bridges for Analysis and Design
Three distinct classifications of bridges are considered in the new Turkish Bridge Design Standard (TBDS) [2]. Based on typology, the bridges are categorized into two groups, namely, Standard bridges and Special bridges.
This categorization is intended to make the distinction between ordinary types of bridges and other, special bridges, such as cable-stayed, suspension, arch and truss bridges. Based on their usage, performance, and importance Standard bridges are further grouped into three types:
- Important bridges
- Regular bridges
- Other bridges
Bridges with strategic importance and those which are critical to transportation systems or which provide access to locations considered critical in emergency management and response are categorized as Important bridges. Single-span bridges and bridges with a maximum length of 100 m, having a maximum of 3 spans and maximum pier height of 10 m, without structural features that complicate their design, are categorized as Other bridges. All bridges outside these two categories are categorized as Regular bridges. Based on the degree of complexity in the analysis process, the bridges are categorized into three groups, namely:
- Complex bridges
- Single-span straight bridges
- Other bridges
Those bridges with features which include complexity in their analysis process are considered complex bridges. Such features include close proximity to an active fault (distance less than 25 km), large effective damping (≥ 30%), long-period bridges (≥ 1.0 sec), curved bridges and bridges with large skew angles (> 20 degrees).
3. Seismicity and Design Response Spectra
The proposed design response spectrum in the TBDS follows that of ASCE 7-16, both in form and the values proposed for site coefficients. The only difference is that in the TBDS the maximum direction effect factor is not considered in defining the mapped spectral response acceleration parameters. Rather, the TBDS proposes factors by which the mapped spectral response acceleration parameters should be multiplied, before using these parameters in the equations or to obtain site coefficient values. The proposed maximum direction factors are as follows. For bridges located less than 25 km from an active fault (LF ≥ 25 km), δs=1.2 and δ1 =1.3 are amplifiers for short-period and long-period acceleration parameters, respectively. For bridges located more than 25 km from an active fault (LF < 25 km), δs=1.2 is the amplifier for short-period acceleration parameter; and γF, which is a factor to consider the near field effect is used to amplify the long-period acceleration parameters. γF is given by the following equations:
The design response spectrum in the TBDS is based on a 2,475-year return period earthquake. In EN 1998-2 [6] it is based on a 475-year return period earthquake, and that standard also recognizes the use of an importance factor (γI), which can be as large as 1.3 for bridges with class III importance to allow for design for longer return period (lower probability) earthquakes. The value for γI is left to be determined on a national basis. There is no similar concept in the TBDS, however, the design displacement of the isolation system is amplified by two reliability factors as described in Section 6.1. The AASHTO Guide Specifications for Seismic Isolation Design of Bridges (GSSID) [3] bases the design on a 1,000-year return period earthquake. Additionally, the isolation system is required to have a displacement capacity of at least 1.1 times the total design displacement for the maximum considered earthquake. According to the commentary in the AASHTO GSSID, in the absence of a site-specific hazard study the maximum considered earthquake may be taken as one with a 2,500-year return period.
4. Equivalent Linear and Time History Analyses
The TBDS has two different damping reduction equations for equivalent linear analysis, making a distinction between bridges located in the near-field and those far from an active fault. The equation for far-field bridges is identical to that of the AASHTO GSSID:
For bridges less than 25 km from an active fault, the following equation is proposed, which is based on the research by Dicleli and Kara [7]:
where ξe denotes effective damping, g is gravitational acceleration, SDS is the spectral acceleration response parameter at short periods, W is the weight of the superstructure, TB is the corner period at the end of the velocity sensitive region of the design spectrum, Td is the post-elastic period of the isolation system calculated based on the post-elastic stiffness, and ∑Qd is the overall characteristic strength of the isolation system.
In the study by Dicleli and Kara, a new damping reduction equation is proposed to obtain reasonable estimates of the actual nonlinear response of seismic isolated structures subjected to near fault ground motions with forward-rupture-directivity effect, using the equivalent linear analysis (ELA) procedure. A comparison between Equation (3) and that suggested by the AASHTO GSSID and EN 1998-2 (or EN15129) is presented in Figure 1. The chart shows the ratio of displacements obtained from equivalent linear analysis using the relevant damping reduction equation to those obtained from nonlinear time history analysis.
Fig. 1. Ratio of displacements obtained from equivalent linear analysis using the relevant damping reduction equation to those obtained from nonlinear time history analysis.
5. Types of Seismic Isolation and Energy Dissipation Devices Covered in the Design Specifications
The types of seismic isolation and energy dissipation devices covered in the TBDS include low and high-damping elastomeric bearings, lead-rubber bearings, flat and curved-surface sliders and metallic and viscous dampers. Specifically, the code recognizes double-surface friction pendulum isolators that may include unequal upper and lower surface friction coefficients, which require a restraining ring at the upper or lower surface (at the surface where friction coefficient smaller), as shown in Figure 2. The code also addresses metallic dampers having a typical force-displacement curve as shown in Figure 3.
Fig. 2. Double-surface friction pendulum isolators with restraining rings and unequal upper and lower surface friction coefficients.
Fig. 3. Force-displacement curve of metallic dampers with adaptive behavior.
In the case of metallic dampers, the TBDS offers a closed-form equation for calculation of the cyclic dissipated energy, as follows:
where Fy denotes the yield force, Ki initial elastic stiffness, Kd secondary (post-elastic) stiffness, d0 displacement, dy yield displacement, ns number of similar dampers, and ηs is a factor, which is equal to 1.0 for metallic dampers whose hysteresis behavior follows a kinematic hardening rule.
TBDS only addresses elastomeric and lead-rubber bearings with circular cross-sections sections and prohibits rectangular sections due to the possibility of high strain concentrations at corners.
6. Types of Seismic Isolation and Energy Dissipation Devices Covered in the Design Specifications
The TBDS recognizes three methods for analysis of isolated bridges namely, simplified analysis (SA), modal combination analysis (MCA) and non-linear time history analysis (NLTHA). The analysis method to be used is based on the bridge category, as laid out in Table 1.
Table 1. Selection of appropriate analysis method based on bridge category in TBDS.
| Bridge importance category (KÖS) | Bridge analysis category (KAS) | ||
| K (complex) | D (other) | T (Single-span straight bridges) | |
| 1 | NLTHA | NLTHA | SA |
| 2 | NLTHA | MCA | SA |
| 3 | MCA | MCA | SA |
6.1. Displacement Capacity of Isolation System
Similar to EN 1998-2, the TBDS requires consideration of increased reliability in design displacement of the isolation system, through application of reliability factors. EN 1998-2 suggests a value of 1.2 for buildings and a value of 1.5 for bridges. In the TBDS the reliability factor is defined as a product of two reliability factors: one reflecting uncertainty in the analysis (γg1) and the other reflecting the consequences of failure (γg2). The reliability factor, γg1, varies between 1.0 to 1.1 and is related to the bridge analysis category (KAS) as shown in Table 2. The reliability factor, γg2 , varies between 1.0 to 1.1 and is related to the bridge importance category (KÖS) as shown in Table 3. For example, for an important bridge (KÖS-1) falling in the complex analysis category (KAS = K), the overall displacement amplification factor will be:
Table 2. Values of reliability factor, γg1, based on bridge analysis category (KAS).
| Bridge analysis category (KAS) | ||
| K (Complex bridges) | D (Other bridges) | T (Single-span straight bridges) |
| 1.10 | 1.05 | 1.00 |
Table 3. Values of reliability factor, γg2, based on bridge importance category (KÖS).
| Bridge importance category (KÖS) | ||
| KÖS-1 (Important bridges) | KÖS-2 (Normal bridges) | KÖS-3 (Other bridges) |
| 1.10 | 1.05 | 1.00 |
6.2. Minimum Testing Requirements
Prototype testing of isolators is required by the TBDS, and is composed of eight tests, as summarized in Table 4. In comparison to the AASHTO GSSID, the minimum testing requirements in the TBDS are more severe, considering that the full earthquake test sequence is repeated in step 5, whereas in the AASHTO GSSID the repetition of the earthquake test involves application of three cycles at the maximum displacement.
Table 4. 2020 TBDS seismic isolator prototype test requirements.
| Test No. | Purpose of the test | Test load | Description |
| 1 | Maximum thermal displacement | DL+0.20×LL | 20 Cycles @ Maximum thermal Displacement at a speed < 5 mm/sec |
| 2 | Wind load test | Average DL (Average among all bearings) | 20 Cycles @ Maximum expected tributary wind load; Test duration 40 sec, minimum |
| 3 | Braking load test | Average DL (Average among all bearings) | 20 Cycles @ Maximum expected tributary braking load; Test duration 40 sec, minimum Maximum expected tributary braking load hold for 1 min |
| 4 | Earthquake test | DL | 3 Cycles @ 0.25d1 3 Cycles @ 0.50d1 3 Cycles @ 0.75d1 3 Cycles @ 1.0d1 |
| 5 | Repetition of Earthquake test No. 4 | ||
| 6 | Repetition of Wind load test No. 2 | ||
| 7 | Repetition of Braking load test No. 3 | ||
| 8 | Stability test | 0.9×DL | Performed at displacement of d1 |
| DL+LL |
6.3. System Property Adjustment Factors
Property modification factors account for effects on isolator properties from multiple effects. In determining these factors, in order to consider the remote possibility that all contributing effects assume their maximum value simultaneously, the AASHTO GSSID also uses a system property adjustment factor. Partial property modification factors due to different effects which are to be multiplied to obtain the overall property modification factor are multiplied by the property adjustment factor in order to more realistically moderate the total, final factor. Property adjustment factors are: 1.0 for critical bridges, 0.75 for essential bridges and 0.66 for all other bridges.
The concept of the property adjustment factor has been expanded in the TBDS to allow for adoption of different values for different effects. According to the TBDS, the upper-bound property modification factor is calculated as:
where β factors are coefficients to account for the likelihood of an extreme state for a specific effect simultaneously with the extreme state of other contributing effects. These β factors are: βtest for velocity and heating effects, βürt for consideration of manufacturing variations, βsc for the effect of temperature, βy for the effects of aging, βaş for effect of travel and wearing, and βk for the effect of contamination. β factors suggested by the TBDS for elastomeric-based and sliding-based isolators are given in Tables 5 and 6, respectively.
Table 5. β factors for elastomeric-based isolators.
| Bridge importance category (KÖS) | βaş (travel and wear) | βk (contamination) | βsc (temperature) | βtest (velocity and heating) | βürt (manufacturing variations) | βy (aging) |
| KÖS-1 | 1.0 | 1.0 | 0.90 | 1.0 | 1.0 | 0.95 |
| KÖS-2 | 1.0 | 1.0 | 0.75 | 1.0 | 1.0 | 0.85 |
| KÖS-3 | 1.0 | 1.0 | 0.65 | 1.0 | 1.0 | 0.75 |
Table 6. β factors for sliding-based isolators.
| Bridge importance category (KÖS) | βaş (travel and wear) | βk (contamination) | βsc (temperature) | βtest (velocity and heating) | βürt (manufacturing variations) | βy (aging) |
| KÖS-1 | 0.95 | 0.97 | 0.95 | 1.0 | 1.0 | 0.97 |
| KÖS-2 | 0.87 | 0.92 | 0.87 | 1.0 | 1.0 | 0.92 |
| KÖS-3 | 0.80 | 0.87 | 0.80 | 1.0 | 1.0 | 0.87 |
7. Minimum Requirements for Seismic Isolation and Energy Dissipation Devices to Ensure Continued Functionality and Serviceability of Bridges
For highway bridges, the TBDS requires that under wind and breaking loads the horizontal displacement of the isolation units be limited to 1.5% of the maximum displacement of the isolator and not exceeding 3 mm or 5 mm for piers and abutments, respectively. For railway bridges with continuous rails, the above-mentioned displacements are limited to 2 mm and 5 mm for piers and abutments, respectively. Should the limits be exceeded, the TBDS suggests use of mechanical restraints (locking devices) to limit displacements under service (non-seismic) loads.
For railway bridges, the TBDS also places a limit of 3 mm on vertical displacement of the isolator under live load.
In addition to the recentering requirements of the AASHTO GSSID, the TBDS includes an additional requirement for bridges less than 25 km from an active fault, by requiring the post-elastic period (pendulum period) satisfy the following inequality:
where d is the shortest distance to the active fault, which controls the seismic hazard.
8. Summary and Conclusions
The distinctive features of the newly developed Turkish seismic isolation bridge design specifications, (TBDS), and comparisons with the AASHTO GSSID and EN 15129 are presented. Different sections of the design specifications are compared with these documents, and the comparative study shows that the TBDS differs in many ways compared to the AASHTO GSSID and EN 15129, including: the approaches related to seismicity and design spectra with particular emphasis on near-field effects, minimum requirements for the serviceability of bridges, property modification factors used for seismic isolation and energy dissipation devices, and also device testing procedures.
References:
1. Turkish Seismic Code (Türkiye Bina Deprem Yönetmeliği) (2018). Afet ve Acil Durum Yönetimi Başkanlığı, Ankara (in Turkish).
2. Turkish Bridge Design Standard (Türkiye Köprü Deprem Yönetmeliği) (2020). Afet ve Acil Durum Yönetimi Başkanlığı, (2020) Ankara (in Turkish).
3.Guide Specifications for Seismic Isolation Design (2014). American Association of State Highway and Transportation Officials, 4th Edition, Washington, D.C.
4. Anti-seismic Devices (2009). European Standard EN 15129, European Committee for Standardization,. Brussels.
5. Minimum Design Loads for Buildings and Other Structures (2016). ASCE/SEI 7-16, American Society of Civil Engi-neers (ASCE), Reston, Virginia.
6. Design of structures for earthquake resistance. Part 2: Bridges (2011). Eurocode 8, EN 1998-2, European Committee for Standardization, Brussels.
7. Dicleli, M and Kara, E (2020). Damping Reduction Equation for the Equivalent Linear Analysis of Seismic Isolated Structures Subjected to Near Fault Ground Motions, Engineering Structures, Vol. 220, pp 1-.17, Elsevier Science.
