A Computational Fluid Dynamics based model that predicts wall shear stress in CANDU outlet feeder pipes
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Wall thinning of carbon steel in CANDU reactor outlet feeder pipes due to Flow Accelerated Corrosion (FAC) is identified as one of the challenges for CANDU reactors since it would force them to shut down due to safety reasons. Several models have been developed over time to predict the corrosion rate (i.e. the rate of wall thinning) of CANDU outlet feeders. These models are developed based on the corrosion chemistry and the mass transfer theories on growth and removal of the protective magnetite layer on the outlet feeder pipe surface. The magnetite layer is acting as a protective layer for the carbon steel feeder pipes by avoiding further corrosion. However, due to the wall shear stress that exerts on the feeder pipe wall, this protective layer is flushed away with the primary heat transport fluid. Wall shear stress is identified as one of the crucial factors behind FAC. Other parameters such as Fe ion concentration, fluid temperature, and pressure would remain within a certain range for a typical CANDU reactor. Still, the distribution of wall shear stress highly depends on the physical arrangement of the outlet feeder pipes. Therefore, wall shear stress would change drastically from one feeder pipe to another resulting in a higher degree of impact on the rate of wall thinning due to FAC. The model developed in this study predicts the maximum wall shear stress on the first bend of a particular feeder pipe considering the fluid Reynolds number, the bend angle and the linear length from the grayloc hub to the first bend. The model is developed using the wall shear stress distribution results generated by Computational Fluid Dynamics (CFD) studies using Siemens NX. The wall shear stress results from the model is then compared against the rate of wall thinning data available for the reactor 01 of the Darlington Nuclear Generating Station as well as some other models available in the literature. iii The model shows a good trend of predicted wall shear stress values against the rate of wall thinning data available. At this stage, the model can be used to identify the feeder pipe with the highest rate of wall thinning due to FAC among a set of given feeder pipes with 2” or 2.5” nominal diameters. This model can be used to identify the optimum feeder pipes for wall thickness measurements during routine maintenance and hence replace the required feeder pipes to avoid any unplanned shut down due to safety reasons.