Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4The research approach followed the task list given in the RFP by NCHRP. The work was split between the team members from BergerABAM and the University of Washington, the for- mer focusing on the deployment and implementation side of ABC and the latter on the seismic performance substantia- tion side. Literature and Practice Review The literature and practice review started with the devel- opment of a list of bridge owners, researchers, organiza- tions, contractors, and suppliers who might provide direct input or indirect leads to the state of the art of current ABC technologies. Parallel to this effort, a set of questionnaires was developed for each contact group. The questionnaires focused on the ABC work done by the institutions and they requested access to design specifications, design guidelines, reports, standard plans or details, special construction pro- visions, or design examples that relate to the use of ABC, seismic or non-seismic. For bridge designers and owners, the questionnaire was extended to obtain information on specific design procedures that are typically employed for seismic design for ABC and identification of the roadblocks to employing ABC. Copies of the questionnaires are pro- vided in Appendix I. In total, 43 U.S. and 13 international organizations were contacted. Twenty-three state DOTs were contacted, including Alaska, Arkansas, California, Florida, Georgia, Idaho, Illinois, Indiana, Louisiana, Massachusetts, Minnesota, Missouri, Montana, Nevada, New York, North Carolina, Oregon, Rhode Island, South Carolina, Tennessee, Texas, Utah, and Washington. Research institutions that were contacted included Univer- sity of California, San Diego; California State University, Sacra- mento; Multidisciplinary Center for Earthquake Engineering Research (MCEER); University of California, Berkeley; Iowa State University; University of Texas at Austin; University of Nevada, Reno; Utah State University; University of Min- nesota; University of Tennessee at Knoxville; Stanford Uni- versity; University of Washington; and the FHWA. The 13 international contacts included researchers from the University of Nottingham, University of Rome, University of Kyoto, Tokyo Tech, University of Canterbury, National Autonomous University of Mexico (UNAM), Technical Uni- versity Frederico Santa Maria, University of Patras, and Uni- versity of Pavia. From the construction side, eight contractors and precast producers were asked to share their experience on ABC work, including, Concrete Technology, Inc.; C.C. Myers, Inc.; Encon United Companies; Flatiron Construction Corp.; Mammoet USA South, Inc.; Mowat Construction; Kiewit Construction; and PCL Construction. Information material on ABC was also gathered from libraries, research databases, and the internet. The material included standard guidelines or surveys from FHWA, AASHTO, the Precast/Prestressed Concrete Institute (PCI) Manual, as well as individual research papers and product information. The collected information was compiled and used by the research team for this project. Definition of Seismic Connections and Performance Strategies The use of precast element technology for bridges in moderate-to-high seismic regions must consider the location and seismic resistance function of the bridge elements being connected. In the United States, bridges are designed for seis- mic resistance by permitting some inelastic deformation of the structure. Such inelastic response is typically restricted to the substructure between the ground level and the soffit of the superstructure. An example of such behavior is a reinforced concrete column that may be designed to experience inelas- tic action in the form of plastic hinges that form at points of high moment, which are often at the bottom and top of the column depending on continuity of the connections between C H A P T E R 2 Research Approach
foundation and superstructure. In the recently adopted AASHTO Guide Specification for LRFD Seismic Bridge Design (2009), the overall concept of seismic behavior is related by identifying an earthquake resisting system (ERS), which is made up of earthquake resisting elements (ERE). In general, seismic forces are limited by the formation of a plastic mech- anism within the structure when subjected to large infrequent earthquakes. There are two seismic design procedures, a force-based procedure in the AASHTO LRFD Bridge Design Specifications (2010) and a displacement-based procedure in the AASHTO Guide Specifications for LRFD Seismic Bridge Design (2009). Both procedures are predicated on the use of inelastic action to resist large earthquakes. The design earthquake in both AASHTO specifications has approximately a 1,000-year recur- rence interval. The use of inelastic action and, specifically, the formation of a plastic mechanism, limits the internal forces that the bridge will experience and provides energy dissipa- tion to limit seismic response. However, the locations of inelastic action are typically at the areas of connection between two members (for example, column and cap beam) because these are the locations of maximum moment. Figure 2 illustrates a pier of a bridge where ABC techniques have been used to connect both superstructure and substruc- ture elements. The connections are shown as lines. Addition- ally, the plastic hinge locations for this bridge are indicated. It can be seen that some of the connection interfaces are adjacent to or in the plastic hinge zones and some are away from such zones. This is an important distinction for the use of ABC tech- niques in seismic regions of the country. If a connection is made in a plastic hinging location, then the connection must be capable of sustaining inelastic deformations and dissipating kinetic energy input to the bridge system by an earthquake. Such connections are energy-dissipating (ED) connections. Connections that are not located where inelastic action is expected would typically be designed using capacity protection principles where the element and adjacent connection are not permitted to experience inelastic action. Such elements and connections are termed capacity-protected (CP) connections. Figure 3 illustrates two examples of locations of connec- tions relative to the plastic hinging zones. The figure on the left has ED connections that are in the plastic hinging zones. These are commonly encountered because the preferred loca- tions for connecting precast elements are also the preferred locations for plastic hinge zones. The preference for connec- tions at the ends of members is related to the desire to trans- port elements that are compact and do not have pieces that protrude. The figure on the right shows an option where the precast connections are kept away for the plastic hinging zones and are, therefore, CP connections. This concept would be ideal for seismic use of ABC techniques, but is not consis- tent with the realities of handling and transporting heavy pre- cast elements. A third type of element and connection may be used, typi- cally when seeking to provide internal articulation and permit displacements with minimal force induced. An example of this type of element would be an internal pin connection. Such connections are termed deformation elements (DE). Another type of DE element is the seismic isolation bearing. Such bearings provide both internal system articulation, typically between the super- and substructure, and they may also pro- vide significant energy dissipation. Seismic isolation is a rel- atively mature technology and, thus, is not focused on in this report. That is not to imply that seismic isolation cannot be effective with ABC, it certainly can. The three types of seismic performance strategies are summarized in Table 1. The extent to which the connection possesses strength and deformation capabilities determines its usefulness for different types of seismic resistance and the seismic zone for which it could be considered. In many cases, the strength and deforma- bility are associated with rotational behavior and, thus, refer to moment and rotation. 5 Figure 2. Potential precast element connection locations relative to plastic hinge locations. Figure 3. Energy-dissipating (ED) versus capacity- protected (CP) connections.
Ideally, the strength would not degrade with cycling, although in reality some degradation is almost inevitable. The strength of the connection is evaluated relative to that of the adjacent members because their relative strengths control the location of the damage. The cyclic deformation capacity measures the ability of the connection to undergo cycles of deformation without jeo- pardizing the strength or performance in some other degree of freedom. For example, a connection with good rotational deformability could undergo many cycles of rotation without loss of shear strength. The deformability is evaluated relative to the expected deformation demand in a high seismic zone, assuming rigid behavior in the connected elements. Figure 4 presents the three seismic performance strategies from Table 1 in terms of their cyclic strength and deformation capabilities and how they are applicable to different seismic zones. A CP connection may be used in any seismic zone, including high seismic zones, as long as deformation capacity and energy dissipation are provided somewhere else within the bridge system. Note that such capacity protection may occur naturally by virtue of the dimensions of the members. For example, in a bridge bent that includes a dropped cap beam and cast-in-place (CIP) diaphragm, the combined cap beam- diaphragm is typically much larger than the column. Thus, the inelastic deformation is likely forced into the column. If the lower stage of the beam and the column are precast, and then connected by grouting bars into sleeves or ducts within the lower stage, the connection between the two stages of cap beam occurs in a beam-column joint region, which must be designed as capacity protected. For energy-dissipating and deformable connections, the required deformation capacity depends on the seismic zone. The distinction between an energy-dissipating connection and a deformable one is not precise, and depends on the level of energy dissipation that is considered necessary to limit peak displacements during an earthquake. The word âconnectionâ has been used numerous times, but it has not been defined. A clear definition is required to under- stand how a connection, particularly an ED or deformable- elements connection, relates to the seismic demands placed on the system of connected elements. Figure 5 illustrates the 6 Figure 4. Seismic performance of connection elements in relation to cyclic strength and deformability and their application for moderate and high seismic zones. Table 1. Seismic performance strategies for connections. Seismic Performance Strategy Performance Behavior Capacity-Protected (CP) Connections CP elements provide a cyclic strength that is higher than the strength of the adjacent bridge members, allowing the connection to remain essentially elastic with minimal or no damage. As a result, the inelastic deformations are forced to occur in the adjacent elements. Energy-Dissipating (ED) Connections ED elements provide a cyclic strength that is lower than that of the adjacent members, thereby causing the inelastic deformation to occur in the connection, but high enough to dissipate enough energy to contribute usefully to the system damping. The deformation capacity is high enough to satisfy the demands associated with the seismic zone in which the bridge is built. The connection may suffer damage, but the consequent strength loss must be acceptable in all degrees of freedom, including both the primary one in which the inelastic deformation occurs and others in which minimal deformation is expected. Deformation Elements (DE) DE have little or no strength in the degree of freedom in which the deformation occurs. The deformation capacity is high enough to satisfy the demands associated with the seismic zone in which the bridge is built. The connection protects the adjacent bridge members by concentrating seismic deformation within the connection region but typically provides negligible energy dissipation. The deformation may be free (e.g., a pin), elastic (e.g., an elastomeric pad), or inelastic. Seismic isolation bearings may provide both large deformation capabilities and significant energy dissipation. St re n gt h R el at iv e to A dja ce nt M em be rs High Capacity Protected Moderate Not Permitted Energy Dissipating (Moderate Seismic) Energy Dissipating (High Seismic) Low Not Permitted Deformable (Low Seismic) Deformable (High Seismic) Low Moderate High Deformability
various pieces of a connection that must be considered when evaluating its suitability for seismic use. The connection shown is between a column and a cap beam, and the seismic plastic hinging zone is indicated. The various reinforcing ele- ments that constitute the connection are also indicated. On either side of the connection interface, or location where the members actually touch, the reinforcement must be devel- oped or anchored to provide continuity of internal force flow for both seismic and permanent loads. Because the reinforce- ment may need to extend well away from the interface and plastic hinging zone, portions of an ED connection may actually be in capacity-protected zones. In fact, they must be because only selected portions of the connection will typically be capable of sustaining the inelastic demands without deteri- oration. This, of course, is highly dependent on the connec- tion configuration and concept. With respect to seismic performance, connections generally can be classified as âemulativeâ of CIP reinforced concrete or ânon-emulative.â The connection illustrated in Figure 5 might be an emulative connection because the configuration of con- nection hardware, shown as the white boxes above the inter- face, is kept away from the plastic hinging zone and only conventional reinforcement is used in the plastic hinging zone. The non-emulative types are typically unique to each concept. Emulative behavior is desirable on one level because confi- dence in the connection performance, be it seismic or durabil- ity, is generally high due to the vast experience that exists with CIP construction. The use of ABC techniques with bridges designed for seismic loading (seismic accelerated building construction [SABC]) will generally follow a building-block approach. This means that various types of connections may be used to assem- ble a bridge that is completely or partially built with ABC tech- niques. The combination of connections, however, must result in a rational seismic load resisting system. This is generally a simple thing to achieve with emulative connections. However, with the more unusual non-emulative types, the connection technologyâs impact on the seismic system and perform- ance must be considered by the designer and kept consistent throughout. The building-block approach does have the advan- tage of permitting several types of connections to be used in a structure to solve various constructability or other problems, so long as the overall ERS is rational. Rational in this sense means capable of providing the expected performance of the entire bridge. Classification of Connection Types The information materials provided in the questionnaire responses and obtained from other sources were screened for details of connection intended to transfer seismic forces between bridge members. These details were classified in terms of their location within the bridge structure, their force trans- fer mechanism, seismic performance, and method of instal- lation. The classifications are explained in more detail in the following subsections. Classification by Location The connections were categorized according to the location for which they might be suitable. For the purpose of this proj- ect, only locations that are important for the seismic behavior of the bridge were considered. Starting from the ground up, the following location categories were defined: ⢠Pile to Pile Cap Connections are typically completely below grade and are very difficult to access for inspection or repair. The connection might join the pile cap to straight or batter piles, driven or drilled piles, and concrete or steel piles. Dur- ing an earthquake, this location experiences a high shear, moment, tension, and compression demand. ⢠Foundation to Substructure Connections are at grade and may or may not be covered by overburden soil. The founda- tion may be a spread footing foundation, pile cap, or drilled shaft. One or more elements of the substructure can be con- nected to a foundation; typical elements include columns, piers, and walls. The location could be obscured by other structural elements, such as barriers, pavement, and build- ings, or submerged in water, thereby making inspection difficult. The location could also be exposed to a harsh environment or be susceptible to damage from accidental impact. Under a seismic event, this location typically expe- riences a high moment and shear demand in multiple direc- tions and may also be subjected to tension/compression in the case of multiple-column bents. ⢠Connections between Column Segments are generally splices of prefabricated columns, piers, or walls. The con- nections may be obscured in ways similar to those of the 7 Figure 5. Connection definition.
foundation to substructure connections, but are generally more accessible for inspection and repair. Under a seismic event, these connections can experience loading that is sim- ilar in nature to, but less intense than, the loads experienced by the foundation to substructure connection. In some cases, connections between column segments may be specif- ically designed to accommodate earthquake displacements. ⢠Substructure to Superstructure Connections join piles, columns, piers, or walls to a cap beam or diaphragm. The location is relatively easily accessible and is typically pro- tected from environmental exposure by the bridge super- structure above. In the longitudinal direction, this type of connection can experience high seismic demands for both moment and shear, and deformation demand may be high. Depending on how the connection is integrated with the bridge diaphragm, the girder moments can have a signifi- cant effect on the seismic behavior of the connection. In multiple-column bents, the moment and shear demands may also exist in the transverse direction. ⢠Connections between Precast Girders and Pier Diaphragms can be accessible in the same way as substruc- ture to superstructure connections. Seismic loading can sub- ject this connection to reversing moments and high shear loads. The deformation demand on these connections is typically small by virtue of the capacity protection provided by the system geometry. Classification by Force Transfer Mechanism The connections were classified according to their force transfer mechanisms, which range from highly localized, such as bar couplers, to more global mechanisms involving large volumes of site-cast concrete. The categories used for this project are as follows: ⢠Bar Couplers can be used to butt-splice reinforcing bars, allowing a continuous force flow in them across the inter- face between the adjacent members. The coupler type most commonly used in bridges is a steel sleeve that is filled with a high-strength grout after the members have been erected. Several proprietary versions are available. These couplers allow for some tolerance in field placement, but they are inevitably larger than the bar itself, especially if oversize couplers are selected to provide extra placement tolerance. Bar couplers allow the connection reinforcement details to resemble those of CIP construction as long as there is enough space to physically fit the coupler. ⢠Grouted Ducts do not directly splice reinforcement bars from adjacent bridge members, but rather allow individual reinforcement bars to be fully developed within the adjacent member. A separate duct is provided for each connection bar. The length of the duct is defined by the length needed to fully develop the capacity of the bar or the length needed to transfer the bar force to the adjacent bars in the segment. As the duct provides some confinement, the bar develop- ment length can be shorter than for typical CIP concrete design. However, the duct length can govern the size of a connection. Grouted ducts are nonproprietary and provide larger construction tolerances than bar couplers, but they require more space within the adjacent memberâs reinforce- ment cage. Nevertheless, the general joint reinforcement lay- out can still be similar to that of a CIP system. ⢠Pocket Connections involve forming a large opening, or pocket, in one bridge member, such as a cap beam. Rein- forcement projecting from another member, such as a col- umn, can be inserted into it, after which the pocket is filled with CIP concrete. The connection reinforcement is fully developed in the CIP concrete within the pocket. The con- nection allows for ample construction tolerances as long as the joint region is not heavily reinforced. The pocket requires that all or part of the joint reinforcement be cast into the precast member. For example, if the longitudinal cap beam reinforcement in a column-to-cap-beam connec- tion penetrates the pocket, then the joint shear reinforce- ment must exist in the precast cap beam, as it cannot be post-installed with the longitudinal column reinforcement. On the other hand, if the column connection reinforcement includes longitudinal and spiral reinforcement, then the longitudinal cap beam reinforcement has to be placed out- side the pocket. The relocation of reinforcement plus the requirement to provide full development length for the connection bars often leads to an increase of the member size compared with a CIP member. ⢠Member Socket Connections provide a socket in which an entire precast member can be inserted and grouted. A socket connection differs from a pocket connection in that no bare reinforcement crosses the interface between the two mem- bers; the connection bars are completely encased in the pre- cast member. The inserted precast member is anchored by the bond provided by the grout and by prying action. Both interface surfaces are roughened to increase the bond resist- ance. The connection offers ample installation tolerances, particularly if the member with the socket is cast-in-place, as may be the case with a footing. If the member is precast, it needs to be large enough to accommodate the socket with enough strength to resist the expected prying action. ⢠Hybrid Connections are connections that contain un- bonded post-tensioning through the joint, which remains elastic and renders the connection self-centering under lat- eral cyclic loading. The hybrid connections also contain bonded bar reinforcement that is either spliced by bar cou- plers or anchored in grouted ducts. It yields alternately in tension and compression to dissipate energy under cyclic loading. 8
⢠Integral Connections typically provide stay-in-place form- work in which two adjacent members can be monolithically connected. An example of an integral connection is a steel diaphragm/cap beam of a composite bridge that is filled with site-cast concrete to connect a concrete column. The inte- gral connection requires the largest in-situ concrete pour of all connections previously described. ⢠Emerging Technologies, Deformation Elements, and Mis- cellaneous Types are in a class of connections that includes the various improvements proposed to either existing con- nection types, such as the hybrid type, or entirely new types of connections, that use new materials or advanced materi- als in new ways. Examples are the use of elastomeric bearings incorporated with columns and the use of shape-memory alloys (SMAs). The class also includes connections that might be used to relieve internal forces and, thus, primarily accommodate deformations. These types of connections are called deformation elements. They are included in the same type class as emerging technologies, because those connections are often used similarly to accommodate deformations. ⢠Mechanical Connections are bolted, welded, or provide mechanical devices to connect two adjacent members such as a bridge bearing. In the connections that have been evaluated, proprietary hardware is commonly found. In some cases, there are multi- ple vendors of similar products so that sole source procure- ment of the hardware is not an issue. However, for some types of connection hardware, this is not the case. The user should be aware of this when considering a connection for potential use on a project. In this report, no effort has been made to iden- tify the connection hardware by manufacturer, and in fact, the opposite is true. The report sought to avoid identification of specific manufacturers. In the categories above, the ones that typically have the potential for sole-source proprietary hard- ware that could affect the detailing are bar couplers, emerging technologies, and mechanical connections. Evaluation Methodology Connection details were assembled from the questionnaire responses and from other sources in the literature review phase. Each connection detail was evaluated to appraise its fit- ness for use and its potential performance during construc- tion, service, and a seismic event. A set of evaluation criteria was developed that included the following: practical experi- ence with an individual connection, measures for the evalua- tion of the technology, readiness of the connections for seismic application, the type of seismic performance, and the practi- cal characteristics of the connections, such as constructability, durability, inspectability, and reparability. The characteristics, classification, and evaluation of each individual connection are presented in a standardized form and are summarized for each connection type in this report. The details for each con- nection are provided in the appendices. The three measures then indicate three different, but necessary, characteristics of the connection types: ⢠Technology Readiness Levelâthe level of development. ⢠Performance Potentialâthe potential merits and disadvan- tages in terms of performance. ⢠Time Savings Potentialâthe potential for accelerating the construction schedule. Furthermore, an evaluation procedure was developed for the assessment of the performance of entire bridge systems composed of suitable combinations of the individual connec- tion types analyzed. The following paragraphs explain the development of the various measures. Technology Readiness All the connections were assessed according to their readi- ness for implementation in accelerated bridge construction in seismic areas. In particular, if connections used newly devel- oped technologies, the assessment provided insight into the requirements for complete deployment in the field or, failing that, described the activities required to move the technology to the next level. The U.S. Department of Energy (DOE) provides a process for this type of evaluation and calls it Technology Readiness Assessment. The process was originally developed by the National Aeronautics and Space Administration (NASA) and has since found its application in the U.S. Departments of Defense and Energy, as well as in related industries. The process is defined by a scale of nine steps, referred to as technology readiness levels (TRL), that describe the readiness state of a new or newly applied technology, from TRL 1 (an observation) to TRL 9 (a fully implemented and functional technology). Typ- ically, the process focuses on the assessment of a particular technology in a particular environment. A guideline can be found in the U.S. Department of Energyâs Technology Readi- ness Assessment Guide (2009). The TRL levels defined by the DOE are given in Table 2. An attempt was made to apply these nine levels of readiness directly to SABC. However, the DOE definitions were found to be a poor fit for the important characteristics of ABC, despite their apparent generality. Thus, a new scale, similar in intent but different in detail, was developed. The new scale is shown in Table 3. A TRL value of 1 is applicable to all connection types in the database because their presence there indicates that a concept has at least been formulated. 9
In TRL 2, either analysis or testing is deemed acceptable for verification of static strength. Analysis is deemed acceptable for this purpose because member design for static response, such as bending of beams, is conducted without question using analysis alone. TRL 3 provides a basic verification of constructability. If the connection concept is used in a seismic region, the bar conges- tion may be more severe and the connection may be less read- ily constructible than its non-seismic counterpart. However, a TRL of 3 indicates that there are no fundamental impediments to construction. TRL 4 indicates that the connection type has been evaluated analytically for seismic loading. The analysis should include the effects of cyclic loading and inelasticity, including ductility demand and capacity. It is also deemed to satisfy the require- ments of TRL 2. TRL 5 demonstrates by test the viability of the critical components. For example, in a connection that depends on mechanical couplers to transfer tension between bars, TRL 5 might apply to an individual coupler. The testing must, of course, be successful for TRL 5 to apply. TRL 6 refers to testing of a complete connection, such as a column-to-cap beam moment connection, rather than to a component, such as a bar coupler. TRL 7 represents an important step towards field imple- mentation. The existence of guidelines implies more than a single test and significant analysis, including studies of the influence of the important parameters in the design. TRL 8 provides verification of constructability when seismic details are used. It is also deemed to satisfy the requirements of TRL 3. TRL 9 provides verification that the connection satisfies the twin requirements of constructability and seismic perform- ance. Few, if any, connections can be expected to achieve TRL 9 because ABC connections have been used in bridges only in recent years and have rarely been built in high-seismic regions. A full-scale dynamic shaking table test does not qual- ify as a TRL 9. The technology readiness assessment procedure was applied to each identified ABC connection to evaluate its readiness for application in seismic regions. As part of the summary, the individual connection types were also evaluated for the extent the connections met the TRL in terms of percentage completed for each level. The evaluation table can be seen in Table 4. The evaluation allows identifying gaps in the current knowledge of a connection type. Ideally, a TRL is met 100% before the next level is started. However, in reality, some levels are skipped or deemed satisfactory after partial completion. A skipped or par- tially completed level does not necessarily mean that a knowl- edge gap exists, as later development steps might fill such gaps. But gaps in the TRL indicate risks that problems might show up later in the development that should have been addressed 10 Table 2. Original definition of technology readiness level (DOE). Table 3. Definition of technology readiness level for seismic accelerated bridge construction (SABC). Level Definition per DOE 1 Basic principles observed and reported 2 Technology concept and/or application formulated 3 Analytical and experimental critical function and/or characteristic proof of concept 4 Component and/or system validation in laboratory environment 5 Laboratory scale, similar system validation in relevant environment 6 Engineering/pilot-scale, similar (prototypical) system validation in a relevant environment 7 Full-scale, similar (prototypical) system demonstrated in a relevant environment 8 Actual system completed and qualified through test and demonstration 9 Actual system operated over the full range of expected conditions Level Definition for Accelerated Bridge Construction 1 A design concept has been formulated. 2 The connection type has been analyzed or tested for static strength. 3 The connection type has been successfully deployed in a low seismic region. 4 The connection type has been analyzed for response to inelastic cyclic loading. 5 The critical connection components have been tested under inelastic cyclic loading. 6 A connection subassembly has been tested under inelastic cyclic loading. 7 Seismic design guidelines for the connection type have been formulated and published. 8 The connection has been used in a bridge constructed in a high seismic region. 9 The connection type has performed adequately during a design-level seismic event in the field.
in earlier stages. For example, for the development of a seismic connection, it is unlikely that a non-seismic deployment per TRL 3 is conducted. Hence, constructability issues in the field cannot be recognized until deployment of the connection in a seismic area at TRL 8. This seems to be a bearable risk. On the other hand, a risk might be perceived higher if a connection has been deployed to a seismic area without TRL 6, seismic testing. Evaluation of Potential Time Savings and Performance of Connections To identify connection systems that merit further invest- ment in testing and analysis, two other measures were seen to be necessary. In this report, they are called the âtime savings potentialâ and âperformance potential.â The first is a simple measure that indicates the time savings that appear possible if the connection is used. The real time savings depends on the construction system in which the connection is incorporated, so the evaluation is necessarily subjective and approximate. However, it is a measure of the possible time advantage rela- tive to traditional CIP construction for the connection type. The scale is given in Table 5. To establish the time savings for each connection type, the researchers convened a meeting attended by a bridge contrac- tor, a Washington State DOT construction engineer, two Washington State DOT design engineers, and two researchers. A detailed description of this is included in Appendix H. The goal was to generate time estimates for each step of the con- struction of a CIP bridge bent, and then to do the same for four of the precast groups (socket, pocket, bar couplers, and grouted ducts) used to construct a typical bent. The predicted time sav- ings were then computed as the difference between the time needed for the precast system in question and the CIP system. The connections were evaluated by connection type, rather than including every possible variant found in the appendices. The estimates were necessarily subjective and were influ- enced by the prevailing building culture in the state (drop cap beams and precast, prestressed concrete girders). However, efforts were made to minimize the subjectivity, first by arrang- ing the presence of both design and construction expertise at the meeting, and second by discussing each construction step in detail, including possible adverse circumstances, before assigning an approximate time. Despite those efforts, slight regional variations must be expected for the estimated time savings computed by this process. The second measure is a composite measure of the perform- ance potential of the connection type and is defined in terms of the connectionâs construction risk, seismic performance, dura- bility, and post-earthquake inspectability. The scale is given in Table 6. To show good performance potential, a connection must be expected to score at least adequately in all four categories. A âmuch worseâ score in any one indicates a potentially unsatis- factory connection type. If connections have âslightly worseâ scores in one or two categories, those scores could be offset by better performance in other categories. However, outstanding performance in one category would not necessarily make the connection type more attractive than performance that is merely satisfactory. Thus, the scales should saturate. For exam- ple, if most structures might be expected to experience 1.5 to 2% drift demand during a major seismic event and drift capac- ity of 5% was considered âadequate,â then a connection drift capacity of 12% adds little to the value of the system because it is very unlikely to be used. This argument also applies to all four categories. The construction risk evaluates the possibil- ity that something might go wrong during construction and 11 Table 4. Evaluation of technology readiness level of connection types and identification of knowledge gaps. Technology Readiness Level (TRL) % of Development Complete 001-5757-0505-5252-0noitpircseDLRT stsixetpecnoC1 2 Static strength predictable 3 Low seismic deployment 4 Analyzed for seismic loading 5 Seismic testing of components 6 Seismic testing of subassemblies 7 Design and construction guidelines 8 Deployment in high seismic area 9 Adequate performance in earthquake Time Savings Potential Definition Relative to CIP Value +2 Much better +1 Slightly better 0 lauqE -1 Slightly worse -2 Much worse Table 5. Connection time savings potential.
detract from the quality or schedule. However, it does not measure the potential time savings so it is also measured using a scale that saturates. The performance evaluation is somewhat subjective, as it is based on the knowledge of the individual research team mem- bers in combination with the information gathered as part of this project. No additional studies or calculations have been performed to support any individual evaluation. However, where the need for such further studies is recognized, sugges- tions are given in the report. Because the performance poten- tial is a qualitative measure, the scale was deliberately kept simple. Each characteristic was defined relative to the corre- sponding measure for CIP reinforced concrete construction. In most cases, little information was provided by the respon- dents about durability and inspectability. Therefore, a default value of 0 (equal to CIP concrete) was used unless specific information was available. It should be noted that some interaction exists between the TRL and the performance potential. If the connection type has already been developed to a high level, much will be known about its performance. The performance potential value can, therefore, be based on objective facts rather than subjective estimates. The measures of potential time savings and performance could be combined to simplify the evaluation procedure. How- ever, doing so would fail to display the relationship between risk and reward. For example, the same value might be assigned to one connection type that offered the potential for large time savings (if everything went right) but carried a high risk of something going wrong and to a second connection type with little potential time savings and little risk. To preserve this important distinction, the two characteristics were evaluated separately. The individual performance evaluation criteria are defined in the following paragraphs. Definition of Construction Risk Rating Criteria The evaluation of the construction risks of a connection col- lectively assesses the difficulty to fabricate and install a con- nection and the associated quality, cost, and schedule risk for typical ABC practice. The ratings are presented in Table 7. The following issues are considered: ⢠Complexity of detailing and number of parts ⢠Required construction tolerances during component fabrication and field installation ⢠Handling, lifting, and shoring equipment needed for field installation ⢠Difficulty of labor access, work environment, and work condition ⢠Complexity of installation procedure and number of steps ⢠Vulnerability to construction mishaps, such as component damage during handling and noncompliant procedures, and availability of inspection and mitigation methods ⢠Sensitivity of installation schedule to individual operations, such as grouting and the time for grout to set ⢠Dependence on specialty trades or parts ⢠Repetitiveness of work and learning curve ⢠Risks associated with subcontracting part of the work Definition of Seismic Performance Rating Criteria The seismic performance rating evaluates how an ABC con- nection for a specific seismic performance strategy, per Table 1 (CP, ED, and DE), performs compared with a CIP connec- tion for a specific seismic zone (low, moderate, and high). The criteria are defined in Table 8. The following issues are considered: ⢠Are there experimental data available and do they demon- strate a cyclic loading behavior that is comparable to CIP construction? ⢠Is the connection emulating CIP construction with proper seismic detailing? ⢠Is it possible to develop bar strength, as necessary, for cyclic loading? ⢠Is it possible to prevent excessive concrete spalling? ⢠Is it possible to prevent bar buckling? ⢠Does the connection allow strain penetration under inelas- tic loading? ⢠Are bars spliced outside the plastic hinge zone? 12 Performance Potential Definition Relative to CIP Construction Risk Value Seismic Performance Value Durability Value Inspectability Value +2 Much better +1 Slightly better 0 Equal -1 Slightly worse -2 Much worse Table 6. Connection performance potential.
⢠Is it possible to retain axial and shear capacity during cyclic loading? ⢠Is the connection self-centering after cyclic loading? ⢠Can damage be contained to the plastic hinge region? ⢠Does the connection have a potential for energy dissipation similar to CIP? ⢠Does the connection provide adequate deformability and strength for its intended seismic performance strategy? Definition of Inspectability Rating Criteria The inspectability rating focuses on post-earthquake inspec- tion of the seismic connection elements. It considers the abil- ity to recognize damage by visual inspection and whether there are methods available for damage assessment of the critical structural components, such as reinforcement and post- tensioning. The rating directly compares to how difficult it would be to inspect and assess damage of the same connection built in CIP or CIP-emulative precast concrete. The rating is presented in Table 9. The following issues are considered: ⢠Can an inspector conclude that there is no damage if no damage is observed by visual inspection? 13 Risk Potential Definition ksiRnoitcurtsnoCkroWdleiFfonoitpircseDPICotevitaleR +2 Much Better Detailing is simple and can be done by a reduced construction crew with minimum need for large construction equipment. There is a very high likelihood that a connection will meet the required quality standard, cost, and installation schedule. +1 Slightly Better Detailing is simple and fabrication and installation of components can be performed by typically skilled construction workers under predictable conditions using conventional construction equipment. There is a high likelihood that a connection will meet the required quality standard, cost, and installation schedule. 0 Equal Detailing is simple but requires attention to fit-up and appropriate use of materials. Reasonably common supervision is required. Fabrication and installation of components might require a specialty contractor or specialized equipment. Most contractors will be able to successfully construct the project. There is a high likelihood that a connection will meet the required quality standard, but there is a minimal risk for not meeting installation cost or schedule. â1 Slightly Worse Detailing is somewhat complex, but skilled construction workers can execute the construction. The work, while complex, is not out of the experience range of a skilled crew but might lead to a slow learning curve, with attendant mistakes, for an inexperienced one. The work might involve specialty contractors or specialized equipment. Close control will be needed to ensure appropriate quality and final acceptance. There is a minimum likelihood that a connection will not meet the required quality standard without repairs after initial construction and there is a moderate risk for not meeting installation cost or schedule. â2 Much Worse Detailing is complex and skilled construction workers under close supervision will be required to properly execute the construction. Specialty contractors or specialized equipment will most likely be required for installation work. Tolerances may be close, materials may be difficult to use in the construction, and tight controls over the work must be worked out in advance and specifically for the particular project. Mock-ups would typically be beneficial and potentially required. Only the most experienced contractors will be successful with execution of the work. There is a moderate likelihood that some repairs may be required after initial construction to satisfy the acceptance criteria for the work and there is a high likelihood that the connection will not meet the cost or installation schedule. Table 7. Rating for construction risk. Seismic Performance Potential Definition Relative to CIP +2 Much Better +1 Slightly Better 0 Equal â1 Slightly Worse â2 Much Worse Table 8. Definition of seismic performance potential.
⢠Can visual inspection recognize that there is a failure of a crit- ical structural component that needs immediate mitigation? ⢠Can damage be assessed with nondestructive evaluation tools? ⢠Can damage be assessed with minimum need of decon- struction? Note that the inspectability during construction is evalu- ated as part of the construction risk rating and is not consid- ered here. Definition of Durability Rating Criteria The durability rating evaluates how the durability of an ABC connection would compare with a connection built of CIP or CIP-emulative precast concrete under similar typical envi- ronmental exposure. The criteria are defined in Table 10. The following issues are considered: ⢠Does the connection provide adequate protection of its structural components? ⢠Does the connection avoid ingress paths for contaminants to structural components? ⢠Is the durability of the connection affected by the quality of construction? ⢠How easy is it to detect deterioration during routine bridge inspections? Bridge Systems This report focuses primarily on individual connection or pier system technologies and not directly on full bridge systems. The nature of the ABC technologies, as they have been devel- oped to date, lend themselves to consideration on the element (connection) or subsystem (column) levels. The seismic per- formance and efficacy of such elements or systems can be inferred through a building-block approach. Conceivably, dif- ferent connection types could be used within a single bridge and the overall seismic performance could be made to conform to the objectives of the AASHTO design specifications. The per- formance and success of the design depends on where, within the bridge, ABC technology is used; specifically, whether the technology is used within a region where inelastic response is expected or within a region that is capacity protected. Where ABC connections or systems are used in the columns where inelastic action is expected, then adequate seismic per- formance of the bridge will depend on the ability of the con- nections to tolerate cyclic inelastic deformations or on the ability to locate the connections where such inelastic action can be avoided. Where inelastic action is expected of the ABC con- nections, then proof testing must be conducted to demonstrate that sufficient toughness is incorporated into the connections. This type of connection is the focus of this report. In other parts of a bridge system, (for example, the super- structure), ABC connections may be designed as capacity- protected elements; thus, they do not need to be designed to accommodate inelastic cyclic actions, and non-seismic con- nection technologies may be used. In abutments, typically the seismic loads are either carried elastically without damage or are transferred through shear keys or other fusible detailing to prevent damage to the superstructure and the abutment itself. Technologies that lend themselves to inclusion of fusible ele- ments between the superstructure and substructure may also find application in ABC projects. For example, seismic isola- tion may be used to provide such fusing, while also providing an interface for assembling large prefabricated systems, such as complete superstructures. 14 Potential Definition Relative to CIP Description +2 Much Better Damage of critical structural components is easily assessed by visual inspection or nondestructive testing. +1 Slightly Better 0 Equal No cracking indicates no damage, large cracks indicate yielding of reinforcement, and spalling of concrete indicates excessive deformations and potential bar failure. Damage of critical structural components can typically be assessed with nondestructive testing. â1 Slightly Worse The absence of visual signs of damage might not necessarily guarantee the integrity of all the critical structural components. Damage assessment of critical structural components is difficult and cannot be done without dismantling a portion of the connection. â2 Much Worse Table 9. Definition of inspectability evaluation criteria. Table 10. Definition of durability evaluation criteria. Potential Definition Relative to CIP +2 Much Better +1 Slightly Better 0 Equal â1 Slightly Worse â2 Much Worse
