During the course of plugging leaking tubes in the No. 12 steam generator at Indian Point #1, fragments of a thermal sleeve were found in the primary side of the steam generator water box. The thermal sleeve came from the 4-in. makeup line to the No. 12 24-in. primary coolant pipe. Fragments of the thermal sleeve were found on May 20, 1970. The last radiographs of weld repairs were taken on Dec. 30, 1970. During the period of 7 months and 10 days, nuclear engineering determined the reasons for failure through physical examinations, and analytical and experimental studies. It was concluded that the failure of the sleeve and the internal cracking of the primary coolant pipe were both caused by thermal fatigue. No evidence of deficiency in the material properties of the sleeve or the pipe was disclosed. Repairs were made by installing a sweepolet fitting and a thermal sleeve of improved design. Thermal sleeves in other lines were examined to ensure their integrity. During this repair health physics was extensively involved because of the large number of mechanics engaged, many of whom were unfamiliar with radiation safety practices, the long periods of work in high radiation areas, and the need to keep accurate up-to-date exposure records on each man. The special tools to do the many intricate operations required were not commercially available, except on a long lead time basis, including development and manufacture. Shops personnel of Consolidated Edison Company of New York were ingenious in adapting available tools and developing and building several tools to accomplish the job. To be certain that the full extent of the problem was understood, and that the repairs completed were satisfactory, a very detailed quality assurance program was instituted and successfully implemented. This program included certification of all materials and NDT personnel, the qualification of personnel such as welders and of all procedures including welding, NDT, and special plant operations associated with the repairs.