The appraisal of concrete structures suffering rebar corrosion is one of the most urgent needs regarding the selection of the technical and economical optimum time for repair. Up to now this appraisal has been mainly based on empirical and subjective considerations. Among the different distressing consequences of rebar corrosion the best known is the cracking of concrete cover. However, very few data have been reported in the literature on the amount of corrosion needed to induce this cracking. In the present paper, some preliminary experiments are reported in which small reinforced beams are artificially corroded by an impressed current, and the amount of current (and loss of bar cross-section) needed to induce the crack at the surface are monitored, together with the evolution of crack width, by the use of strain gauges applied to the surface of the specimens. In a companion paper, a numerical model to relate the decrease in rebar cross-section to the cover cracking will be developed. That model is based on the orderly imposition of corrosion to finite elements of the rebar by a fictitious temperature increment that produces analogous effects, while concrete cracking is introduced by a standard smeared-crack model. The experimental results indicate that only a few micrometres of loss in rebar cross-section are needed to induce visible cover cracks (0.1 mm width) in the conditions of the test.
