Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 28
28 carried out. However, case examples that would not warrant in the bridge, there was no advantage. There appeared to be a "yes" were provided. For example, the discovery of typical little recognition of the significantly superior toughness of out-of-plane distortion cracks, which usually take years to HPS in decreasing the potential for fracture of a given become critical, would not warrant a response of "yes." Fur- member. In addition, most agencies (79%) did not feel that thermore, inspections that found fractured members would eliminating poor fatigue details, say less than a Category C also not warrant a response of "yes" because the inspection fatigue detail, would influence their decisions with respect to did not prevent the fracture and the bridge did not fail. inspection frequency or the level of detail in new bridges. Some states base their inspection frequency on the types of Interestingly, approximately 30% of the agencies that fatigue details that are on the bridge and use this (and other replied to this question answered "yes." However, in describ- data) to determine how often the bridge should be inspected ing the specific case where this had occurred, the fracture of and with what level of detail. the member had already taken place. Thus, the respondent should have answered "no." Note that these responses also With respect to fatigue, it is commonly observed that by indicate cases where FCMs had fractured, and bridges did not using simplified structural analysis methods the calculations collapse, and therefore would add to the few cases described for many bridges indicated no remaining fatigue life or even in chapter two (see Figures 16 and 17). "negative" fatigue life. These calculations imply that fatigue cracking should be observed presently or in the near future When one of the owners who completed the survey was on these bridges. However, such bridges typically show no asked why they answered "yes" to the question, they indi- signs of fatigue-related problems. (This does not include cated that although the fracture had occurred, the bridge did cracking from secondary stresses, such as web-gap cracking, not collapse. In addition, there was no indication of any sag- because this type of cracking is not explicitly considered in ging of the structure. Hence, because the inspection found the fatigue rating calculations.) When asked about their agency's fracture before any loss of structural integrity, the structure policy regarding cases when this inconsistency occurs, results could be repaired before there was any more significant dam- were that one of the owners responding had a formal policy age. (It should be noted that the fracture in question occurred regarding this common problem. Most indicated that they do in May 2003 under moderate temperatures. Had the frac- a more rigorous analysis or field instrumentation on a case- ture occurred in the winter during a cold period, it could by-case basis. Even then, these efforts are mostly limited to have been much worse and significant damage introduced. use on larger or critical bridges. However, other factors, such Furthermore, the forensic investigation revealed that the frac- as route, ADTT, existing condition, type of steel, and age ture had occurred less than a few days to a maximum of one also influence the decision to use the more advanced meth- week before being discovered. Therefore, it is almost purely ods. Weigh-in-motion or advanced three-dimensional analy- coincidental that the fracture was found before any additional ses are used by approximately 10% and 32% of the agencies, damage occurred. Ironically, the owner had developed retro- respectively, as needed. Interestingly, field instrumentation fit drawings and was about to let the construction contract to and load testing was used by approximately 45% of the agen- retrofit the bridge to prevent this type of problem at the time cies at one time or another. However, this is somewhat mis- the failure occurred.) leading, because an agency may have only used field instru- mentation once in 10 years. Thus, although nearly half of Nevertheless, when the responses were closely reviewed those responding have indicated they have used field instru- and adjusted to properly answer the question, the percentage mentation to improve fatigue life predictions at one time, it of "yes" respondents drops to 23%, which is still significant is not used very often. because failures were prevented. SUMMARY OF RESPONSES TO PART III--FAILURES Journal articles have reported on instances of fractures that were found during inspections; for example, the Edgewood This section collects information relating only to bridge fail- Road Bridge in Cedar Rapids, Iowa (31). This bridge was ures. Cracking associated with fatigue that did not result in being inspected by a private firm and a large fracture in the top fracture was not included as a failure. For example, out-of- flange of the twin-box structure was discovered at several loca- plane distortion cracks were not to be counted as failures. tions. Although contracting inspections out worries many However, fractures that resulted from a fatigue crack were to bridge owners, this firm was apparently doing a good job. Frac- be included. Respondents were asked to distinguish between tures in the Paseo Bridge on I-35 in Kansas City, Missouri, failures that occurred before and those that occurred after the were noticed only when an 8 in. (200 mm) gap opened at the implementation of the FCB inspection program. Failures in expansion joint (39). FCBs and non-FCBs were to be identified separately. Fur- thermore, owners were asked to distinguish between failures Only 5% of those responding indicated that the use of that were the result of impact, scour, and so forth, and those HPS would influence how they view FCBs. Most believed caused by fatigue or fracture. Unfortunately, the data col- that if a member fails, it does not matter what type of steel lected related to these questions were not as complete as was was used and that owing to the lack of redundancy inherent desired and were very difficult to quantify.