Journal of Performance of Constructed Facilities

An overflow pond dam at a gold mine in South Carolina failed in October 1990 approximately 2 weeks after an exceptionally large rainstorm produced more than 280 mm (11 in.) of rain in 1 day (>>2,500-year event). The pond contained a cyanide leaching solution used to extract gold from low-grade ore. The pond was lined with two geomembranes separated by a geodrain intended to serve as a leak detection system. A detailed investigation attributed the failure to internal erosion within the dam that resulted from uncontrolled natural ground-water seepage occurring below the dam. The dam was constructed of gap-graded rockfill containing significant portions of highly erodible fines. Shortcomings in the design and in the construction procedures that contributed to the failure are identified and discussed.

During 2 days in October 1994, 4-month-old concrete on a runway developed extensive spalling along transverse joints. Damage was extensive, limited to transverse joints; was located in trafficked areas only; and developed only after the beginning of freezing weather. An investigation into possible causes of this spalling concluded that the concrete probably developed a network of microcracks from premature loading before the concrete had gained adequate strength. The concrete continued to gain strength with time and these cracks were not sufficient to cause spalling. Later, moisture accumulated in the cracks froze, which propagated the original cracking to free the spall fragments. The most likely cause of the original inadequate strength gain was probably poor curing of the concrete in the vicinity of the transverse joints caused by damage to the curing compound membrane during initial sawing of transverse contraction joints. Alternate explanations of the damage were unable to account for all of the observed characteristics of the damage. The remaining concrete is essentially sound, and permanent repair of the runway requires only patching of the spalled areas.

The objective of this study is to evaluate the American Association of State Highway and Transportation Officials (AASHTO) rating procedure for reinforced concrete (RC) slab bridges by using nondestructive load testing. Three RC slab bridges, each with three continuous spans, were selected for this study. These bridges had minor to severe deterioration. For each bridge, the rating factors were calculated using AASHTO specifications. Nondestructive load tests were carried out on all three bridges. Four dump trucks, each filled with gravel, were used for loading. The trucks were placed in three different load positions to cause maximum positive moment in the middle span, maximum positive moment in the end span, and maximum negative moment at the intermediate supports. The analytical models were updated using the results from load testing and then were used to calculate new rating factors for each bridge. The current AASHTO rating procedure was found to be too conservative.

Many multigirder steel bridges built before 1985 are experiencing distortion-induced fatigue cracking at diaphragm-girder connections. In bridges in Birmingham, Ala. fatigue cracks in the welds and base metal in hundreds of diaphragm-girder connections have been discovered. In addition, one repair scheme, a bolted-angle retrofit, has not performed satisfactorily because of fatigue cracking in the connection angle. Field tests at six bridge sites in the Birmingham, Ala. area were conducted to evaluate the potential for further cracking at previously uncracked connections, the appropriateness of continuing a drilled hole repair of girder web cracks, and the performance of a proposed new bolted-angle retrofit. Primarily, surface strain measurements resulting from traffic and calibration trucks crossing the bridge were recorded to evaluate the performance of the connections. Results indicate cracking will continue at welded diaphragm-girder connections, hole-drilling crack tips is not a reliable fix of web cracking, and the proposed new bolted-angle retrofit is an improvement of the original bolted connection.

Vibration measurements were made on a high-rise condominium structure in Singapore to obtain its dynamic characteristics. Subsequently the response and wind characteristics were measured over a 15-month period. Based on combination of the measurements it is possible to show how the structure responds to wind and other lateral loading. The main conclusions are that the dynamic component of response to wind is low, mainly due to the low wind speeds encountered in Singapore. Even so, although the mean wind speeds are low, the wind has high gust factors. The measured response of the building to a tremor arising from an earthquake in nearby Sumatra was an order of magnitude larger than the maximum measured dynamic response due to wind. It appears that the code provisions for lateral load would be adequate for this building even in strongest (design) winds, but the validity of code provisions for large far-field tremors is not proven.

In contrast to design of structures for their service life, very little information is provided to the engineer for selecting appropriate design loads for construction. Before new design loads or load combinations can be recommended or existing recommended values modified, the adequacy of current design loads for construction must be evaluated. This can be accomplished only by comparing actual load data with design loads prescribed by available standards. This paper presents the results of one such study to investigate the adequacy of current construction load provisions in the context of vertical loads on shores during reinforced concrete construction. Construction load provisions from ACI 347, the OSHA standard, ANSI A10.9, and the proposed ASCE construction load standard are first reviewed. Design loads recommended by available standards are evaluated and compared with actual (measured) axial shore loads. Shore load data from six sites are considered, including two recently instrumented projects and four multistory buildings studied previously.