Reliability performance in tension governed reinforced concrete reservoirs as a function of target crack width

Abstract

The limit state that typically governs the design of circular reinforced concrete water retaining structures is the limitation of leakage through cracks. Current design codes prescribe target crack widths in the range of 0.05–0.2 mm, depending on the design situation, to limit leakage to acceptable levels. Fairly high model uncertainty exists in the prediction of load‐induced crack widths and the beneficial effect of autogenous self‐sealing has not effectively been quantified, so that the achieved levels of reliability using current codified procedures are unknown. This research uses a reliability analysis previously developed by the authors to assess the level of SLS reliability achieved in tension‐governed reinforced concrete reservoirs covering much of the scope of practice. The SLS reliability performance is evaluated using a limit state of allowable versus predicted leakage, including a probabilistic quantification of the effect of autogenous self‐sealing, and the uncertainty present in the prediction of load‐induced crack widths. Many permutations of practical reservoir geometries are considered for four combinations of stabilization period and water tightness test time (leakage regimes) typically used in water tightness tests. The target crack width recommendations of EN 1992‐3 and Model Code 2010 are evaluated across the range of geometries and leakage regimes. The commonly used target crack width of , recommended by Model Code 2010 and many other codes, was found to be inadequate to achieve a target SLS reliability index of . The values from EN 1992‐3 showed better performance but were found to result in reliability indices frequently in excess of , indicating that the values are perhaps too stringent. Results confirm that smaller target crack widths and longer water tightness test times increase the achieved reliability. However, even for a given target crack width and leakage regime, a substantial range of achieved reliabilities is observed depending on reservoir configuration. This shows that design to meet current target crack widths is not effective at meeting target reliability across the range of practical application. A suggestion is made to refine reservoir specific crack width targets to achieve target reliability more consistently.