~~The Public Roads Administration in Norway has been using coating systems consisting of thermally sprayed zinc (TSZ) and a paint system since about 1965. Desired lifetime of the coating system is at least 35 years. During the past decade TSZ with paint has also been included in NORSOK M-501 (System 2B) and applied on offshore platforms particularly on areas with difficult access for maintenance e.g. between cellar deck and splash zone.Lately a number of bridges over costal straits have shown corrosion in the zinc coating after about 20 years exposure. The corrosion attacks are mainly found as individual spots randomly distributed on horizontal surfaces under the bridges. Coating thickness was in most cases found to be as specified. These are the surfaces on a coastal bridge that are most susceptible to corrosion. They are facing directly down to the seawater and rain will never wash deposited salts away. Hence these surfaces will always be covered by a thin layer of salt.Since the attacks were localized in spots and the film thickness was as specified it is reasonable to assume that the attacks were caused by pinholes in the paint. Pinholes are a well-known problem in painting of TSZ. At the corrosion spots it is no longer possible to find evidence of pinholes since the paint has degraded by the corrosion. However investigation of the same coating system applied on an inland bridge in a much less corrosive environment revealed pinholes without corrosion. Pinholes therefore seem to be a plausible explanation for the problem.A laboratory investigation of application of two sealers on TSZ showed that pinholes were easily formed when applying sealers with as little as 40% solids. Higher sealer film thickness increased tendency to formation of pinholes. The pinholes were gas bubbles formed in the wet sealer film during or immediately after either from solvent evaporation or trapped air in the film. The bubbles were covered by a thin layer of binder and some of them would remain covered by this layer even after full curing of the sealer. This will then effectively prevent subsequent coats to fill the pinholes and the coating system will have a permanent weak spot. When the sealers were diluted to 30% solids all the bubbles collapsed and no pinholes were formed. Hence sufficient dilution seems to solve the problem.