A cyclic immersion apparatus has been constructed for the evaluation of edge corrosion of coil-coated cladding. In this method, edge corrosion is accelerated by coupling each coil-coated specimen to an external stainless steel cathode in order to increase the cathode to anode area ratio, thereby increasing the rate of undercutting of the metal coating. Comparisons with outdoor results reveal that the cyclic immersion test produces realistic edge corrosion and blistering morphologies and has significantly less variability than conventional cabinet tests. Furthermore, it has the distinctive ability amongst accelerated tests to potentially control the acceleration of undercutting at the cut edge by varying the size of the cathode. A limitation of the test at present is the relatively small acceleration factor for the simulation of industrial environments. This limitation is explained with the aid of polarisation test results and methods for increasing the acceleration of the test are discussed. (C) 1999 Elsevier Science S.A. All rights reserved.