Desalination, 46 (1983) 345-366 ElsevierScience Publishers B.V.,Amsterdam-F?intedin TheNetherlands 345 PROBLEMS ENCOUNTERED DURING COMMISSIONING AND SUBSEQUENT PERFORMANCE OF AN MSF DESALINATION PLANT IN SAUDI ARABIA. F. H. BUTT, A. G. MAADHAH, F. RAHMAN and H. M. Al-ZAHRANI Research Institute, University of Petroleum & Minerals, Dhahran, Saudi Arabia ABSTRACT A 7550 m3/d 3-unit MSF plant suffered corrosion of alloy 20 acid injection lines and 316L SS vacuum system when commissioned during Aug.-Oct. 1981. Cor- rosion of acid lines was believed to have been caused by the chloride ions of seawater which flowed back into the acid lines during repeated shutdowns. Often high levels of residual Cl2 in the make-up, coupled with low pH levels before and after the decarbonator (DC), were established to have favored the generation of Br:! gas which had caused corrosion of the vat. system. Measures to rectify corrosion of acid lines and the vat. system are discussed. Performance of DC and DA is discussed. The plant performance, both during commissioning and in the subsequent period, is discussed in terms of distillate output, heater FF, GOR, etc. INTRODUCTION In February 1981, we were requested by the appropriate officials of a Naval Base to study some of the problems that had been encountered during commission- ing (Aug.-Oct. 1981) of the three acid-dosed units of the MSF plant, located on the Arabian Gulf coast of Saudi Arabia. We found the plant to be generally well designed and the materials of construction quite appropriate for the intended service; all water boxes, including those of ejector condensers, are lined with 90/10 cupronickel while the flash chambers of heat gain and heat reject sections are lined with 90/10 and 316L SS, respectively. The tubes in the heater and heat gain section are 90/10 and those in the reject section and ejector conden- sers are titanium. The vacuum system's interconnecting piping, ejectors and condenser shells are made out of 316L SS. Other features of the plant are given in Table 1. During the commissioning period, the plant performed well and produced better than the guaranteed amount of quality product. But unexpectedly, the acid injec- tion lines and the vacuum system developed some corrosion problems. Following the required maintenance work and modifications by the contractor, the units were re-started; Unit 1 successfully completed the performance test in Feb., Unit 2 in March and Unit 3 in April 1982. Between Feb. and Oct. 1982, Units 1, 2 & 3 were run for 33, 54 and 28 days, respectively. OOll-9164/83/$03.00 0 1983ElsevierSciencePublisheraB.V.