Inspection and Management of Bridges with Fracture-Critical Details (2005)

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.

This report is focused on inspection and maintenance of bridges with fracture-critical members (FCMs). The AASHTO LRFD Bridge Design Specifications (LRFD Specifications) defines an FCM as a “component in tension whose failure is expected to result in the collapse of the bridge or the inability of the bridge to perform its function.” Note that the FCM can refer to a compo- nent such as a flange of a girder and does not necessarily include the whole “member.” Approx- imately 11% of the steel bridges in the United States have FCMs. Most of these (83%) are two- girder bridges and two-line trusses, and 43% of the FCMs are built-up riveted members. The objectives of this synthesis project were to: • Survey the extent of and identify gaps in the literature; • Determine best practices and problems with how bridge owners define, identify, docu- ment, inspect, and manage bridges with fracture-critical details; and • Identify research needs. The information was obtained from the literature, from a survey of bridge owners and consultant inspectors, and from targeted interviews. Thirty-four states and three Canadian provinces responded to the survey. Information was gathered regarding how bridge owners define, identify, document, inspect, and manage bridges with fracture-critical details. Specific inquiries were made about the following issues: • Inspection frequencies and procedures; • Methods for calculating remaining fatigue life; • Qualification and training of inspectors; • Available and needed training; • Experience with FCM fractures and problem details; • Examples of where the inspection program prevented failures; • Cost of inspection programs; • Retrofit techniques, including emerging technologies; • Nondestructive evaluation/nondestructive testing; • International experience and practice; • Fabrication methods and fabrication inspection; and • Needed research related to fracture-critical bridges (FCBs). This report will be useful to owners and consulting engineers engaged in the design, inspection, and management of bridges with fracture-critical details as a guide to present specifications and typical engineering judgment. During the 1970s, the material, design, fabrication, shop inspection, and in-service inspec- tion requirements were improved for steel bridges in general. In 1978, special provisions were implemented for FCMs, primarily in reaction to bridge collapses. These requirements were successful in transforming the industry and the design of modern bridges, so that fatigue and fracture are very rare in bridges built in the last 20 to 25 years (i.e., since 1980). Unfor- tunately, approximately 76% of all FCMs presently in inventory were fabricated before 1978. SUMMARY INSPECTION AND MANAGEMENT OF BRIDGES WITH FRACTURE-CRITICAL DETAILS

The focus on inspection and maintenance is appropriate, because it was found that the extra fabrication and material costs were small in comparison with the additional costs of “hands-on” fracture-critical inspection mandated since 1988 by the FHWA’s National Bridge Inspection Standards. The approximate initial cost premium for new bridges with FCMs is on the order of 8% of the cost of fabricated steel. There is also a hidden initial cost in some cases where more expensive superstructure designs have been used than are necessary to maintain an acceptable reliability level because of restrictions or more subtle prejudice against bridges with FCMs. The major impact on life-cycle costs is the additional mandate for hands-on, in-service inspection of FCMs. The cost of the fracture-critical inspection is typically two to five times greater than inspections for bridges without FCMs. Inspections of closed sections such as tub girders and tie members are extremely expensive, because inspectors must get inside. Inter- estingly, what constitutes a fracture-critical inspection is often subject to interpretation and disagreement. The frequency of fracture-critical inspections is actually not currently speci- fied in the National Bridge Inspection Standards and varies up to every 5 years, is typically every 2 years, but often is every year or more frequently if there are specific problems. These hands-on inspections have revealed numerous fatigue and corrosion problems that otherwise might have escaped notice. Twenty-three percent of survey respondents indicated that they found significant cracks that could have become much worse, possibly averting col- lapses. Similar experiences may be found in the literature. The information from these inspections is useful in bridge management; that is, in prioritizing bridges for rehabilitation and replacement. Furthermore, states that have centralized teams that do all of the fracture- critical inspections report numerous advantages with this approach. FCMs are nonredundant. The LRFD Specifications define redundancy as “the quality of a bridge that enables it to perform its design function in the damaged state.” Note that these defi- nitions are not clear about what type of damage, load type, magnitude, distribution on the bridge, dynamic amplification, and load factors are supposed to be resisted by the damaged structure. Nonredundant is a broader term than FCM because nonredundant also includes: • Substructures; • Members that may be inherently not susceptible to fracture, such as compression mem- bers, but still could lead to collapse if damaged by overloading, earthquakes, corrosion, fire, terrorism, ship or vehicle collisions, etc.; and • Members made of materials other than steel. Substructures such as piers are often nonredundant and have contributed to most of the major collapses of both steel and concrete bridges. In addition to prevention of collapse in the event of fracture, redundancy of the super- structure is important for several other reasons of varying levels of importance. The first is the need to more easily redeck the bridge, which can often be readily overcome by adding an additional line of stringers. It must be recognized that events other than fracture can also damage and completely destroy members of the superstructure. These are compelling rea- sons to have redundancy. These reasons for redundancy (other than fracture) could be used to encourage redundancy outright instead of indirectly by penalizing FCMs. For example, redundancy is encouraged in Sections 1.3.2 and 1.3.4 of the LRFD Specifications. Load factors are modified based on the level of redundancy, and it is stated that multiple-load-path and continuous structures could be used unless there are compelling reasons to do otherwise. In the AASHTO Manual for Condition Evaluation and Load and Resistance Factor Rat- ing (LRFR) of Highway Bridges (LRFR Manual), redundancy is reflected in system factors 2

that reduce the capacity of each member in nonredundant systems. The system factors are cal- ibrated so that nonredundant systems are rated more conservatively at approximately the same level of reliability associated with new bridges designed by the LRFD Specifications, called the “inventory” level in former rating procedures. Redundancy is related to system behavior rather than individual component behavior. Redundancy is often discussed in terms of three types: • Internal redundancy, also called member redundancy, can occur when a nonwelded member is comprised of multiple elements, and a fracture that formed in one element cannot propagate directly into the adjacent elements. • Structural redundancy is external static indeterminacy and can occur in a two or more span continuous girder or truss. • Load-path redundancy is internal static indeterminacy resulting from having three or more girders or redundant truss members. One can argue (and show analytically) that the transverse members such as diaphragms between girders can also provide load-path redundancy. In preparing this synthesis, it was found that internal and structural redundancy are often neglected by agencies and designers. Retrofits are described, which have been used to add all three forms of redundancy to bridges with FCMs. Ultimately, it is the target level of reliability that designers and raters of bridges should strive to achieve, rather than focusing exclusively on redundancy. Redundancy has a major impact on the risk of collapse and this impact is accounted for appropriately for all types of structures in both the LRFD Specifications and the LRFR Manual. Using the LRFD and LRFR procedures, it is possible to achieve the target level of reliability without redundancy in a bridge that is more conservatively designed by only approximately 17%. For example, a nonredundant bridge designed for an HS-25 loading would have greater reliability than a redundant bridge designed for HS-20 loading. There are various classifications of superstructure types having FCMs and consequently there is wide disagreement. For example, twin box girders would be expected to perform even better than twin I-girders owing to the torsional capacity of the intact box girder and the alter- nate load paths available within each box girder; however, most agencies consider these as FCMs. Unfortunately, the LRFD Specifications are not clear about what types of superstruc- tures have FCMs. Common assumptions that fractures in certain superstructure types will lead to collapse are generally too simplistic. Many bridges have had a full-depth fracture of an FCM girder and did not collapse, usually owing to the alternative-load-carrying mechanism of catenary action of the deck under large rotations at the fracture. It is apparent that other elements of these two-girder bridges, particularly the deck, are sometimes able to carry the loads and prevent collapse in these fracture-critical bridges (FCBs). Indeed, these alternate load paths were so robust in some of these failures that there was little or no perceptible deformation of the structure. The capacity of damaged superstructures (with the FCM “damaged” or removed from the analysis) may be predicted using refined three-dimensional analysis. However, there is a strong need to clarify the assumptions, extent of damage, load cases and factors, and dynamic effects in these analyses. The commentary of the LRFD Specifications states that The criteria for refined analysis used to demonstrate that part of a structure is not fracture critical has not yet been codified. Therefore, the loading cases to be studied, location of potential cracks, degree to which the dynamic effects associated with a fracture are included in the analysis, and fine- ness of the models and choice of element type should all be agreed upon by the owner and the engi- 3

neer. The ability of a particular software product to adequately capture the complexity of the prob- lem should also be considered and the choice of software should be mutually agreed upon by the owner and the engineer. Relief from full factored loads associated with the Strength I Load Combi- nations of Table 3.4.1-1 should be considered as should the number of loaded design lanes versus the number of striped traffic lanes. NCHRP Report 406: Redundancy in Highway Bridge Superstructures, also gives practical requirements for estimating the residual capacity of the damaged superstructure. This same type of analysis is being used to evaluate older structures, as well to better direct resources for maintenance and replacement; for example, if a bridge is not really fracture- critical then it may not be necessary to replace it as soon. Interestingly, other countries do not make a distinction between FCM and non-FCM. Scan- ning tours have noted that three-dimensional refined analysis is more often used in the design and evaluation of bridges in Europe and that the inspection interval is often based on risk. There are some in the United States who believe that fracture-critical inspection should be less frequent for more modern bridges as a result of the much better detailing, materials, and fab- rication used in their construction. It has been suggested that the inspection interval should be based on the level of truck traffic, fatigue details, and other risk factors. Overall, training for inspectors appears to be adequately available through existing courses provided by the National Highway Institute. None of the agencies responding to the survey identified any problems with current educations strategies. However, one area that could require additional effort is related to the documentation and archiving of previous failures and problems. Failures known to have occurred in certain states were not reported in some of the survey responses. Hence, a better method of tracking, archiving, and disseminating such information appears to be needed. Although training seems adequate related to inspection, many engineers have noted how the education and training related to fatigue and fracture is not. Engineers are reportedly not learning lessons from fatigue and fracture incidents because of inadequate understanding. There is concern that they are not able to predict future problem details. Better education in this area in engineering programs, as well as short courses for practicing engineers, could lead to improved new bridge designs and better educated engineers available to participate in maintenance and inspection programs. 4