Abstract
Exposure testing of concrete block or slab specimens by cyclic chloride ponding has been widely utilized for ranking or qualifying different reinforcement types and concrete mix designs and for projecting service performance. However, experimental issues such as inadequate bar end isolation, ponding bath leakage, and inappropriate specimen design and testing methodology can result in data that misrepresents long-term service performance. Further, the fact that most testing protocols employ cyclic chloride solution ponding can lead to erroneous ranking of different corrosion resistant reinforcements and false projection of service performance, specifically because surface chloride concentration becomes elevated compared to that for actual service. The present paper discusses these issues and provides options whereby accelerated test data can be acquired that more correctly rank different reinforcements and thereby more realistically represent service performance.
