Numerical Calculation of Shear Stress Distribution on the Inner Wall Surface of CANDU Reactor Feeder Pipe Conveying Two-Phase Coolant

Abstract

Two-phase flow fields inside feeder pipes of a CANada Deuterium Uranium (CANDU) reactor have been simulated numerically using a computational fluid dynamics (CFD) code to calculate the shear stress distribution, which is the most important factor to be considered in predicting the local areas of feeder pipes highly susceptible to flow-accelerated corrosion (FAC)-induced wall thinning. The CFD approach with schemes used in this study to simulate the turbulent flow situations inside the CANDU feeder pipes has been verified by showing a good agreement between the investigation results for the failed feedwater pipe at Surry Unit 2 plant in the U.S. and the CFD calculation. Sensitivity studies of the three geometrical parameters such as angle of the first and second bends, length of the first span between the grayloc hub and the first bend, and length of the second span between the first and second bends had already been performed. In this study, the effects of void fraction of the primary coolant coming out from the exit of pressure tubes containing nuclear fuel on the fluid shear stress distribution at the inner surface of the feeder pipe wall have been investigated to find out the local areas of feeder pipes conveying a two-phase coolant, which are highly susceptible to FAC-induced wall thinning. From the results of the CFD analysis, it is seen that the local regions of feeder pipes of the operating CANDU reactors in Korea, on which the wall thickness measurements have been performed so far, do not coincide with the worst regions predicted by the present CFD analysis, which is the connection region of straight and bend pipes near the inlet part of the bend intrados. Finally, based on the results of the present CFD analysis, a guide to the selection of the weakest local positions where the measurement of wall thickness should be performed with higher priority has been provided.