Four D-12 coil nests have failed since the start-up of the Redox Separations plant. Due to the high contamination and radiation levels of these vessels, the cost of decontamination has prevented the examination of all but one of the four. It was hoped that the examination of D-12 pot #2 would lead to an explanation of all four failures. This report will deal primarily with the metallurgical examination of the failure of D-12 pot #2.

The following is an attempt to explain the method by which the precision associated with an observed volume reading in a process vessel (E-12, C-1) should be calculated based upon a regression analysis of cumulative data. There are two types of volume measurement consists of estimating the total volume in a vessel at some inventory time. A transfer measurement consists of estimating the volume between two levels within a tank.