As shown in fig. 3 the corrosion attacks were covered by tubercles. Sulphide was qualitatively detected in the crevices at corrosion attacks by an iodazide test, but not directly in the tubercles. The sulphide could originate from the presence of SRB, but no microbiological analysis was performed.
In this case the primary critical parameter was defects in welds, which were found by metallographical examinations. To avoid this type of failure better weld quality must be specified (root defects, crater pores, lack of fusion) and then the normal pickling procedure will ensure resistant fittings without sensitive areas. However, it is likely that a high degree of microbiological activity has enhanced the critical situation and helped build tubercles under which corrosion attacks continued. According to the Danish Code of Practice for domestic water installations [14] no place in the system should the hot water temperature be lower than 50°C, and 45 °C at peak load. This is to reduce the risk of bacterial growth of especially pathogen bacteria such as Legionella, but it would also help to reduce the risk of MIC. Furthermore the use of circulating lukewarm water is unwise from both a hygienic and a corrosion point of view.