Evaluation of a numeric procedure for flow simulation of a 5 × 5 PWR rod bundle with a mixing vane spacer

Abstract

The fuel assemblies of the Pressurized Water Reactors (PWR) are constituted of rod bundles arranged in a regular square configuration by spacer grids placed along its length. The presence of the spacer grids promote two antagonist effects on the core: a desirable increase of the local heat transfer downstream the grids and an adverse increase of the pressure drop due to the constriction on the coolant flow area. Most spacer grids are designed with mixing vanes which cause a cross and swirl flow between and within the subchannels, enhancing even more the heat transfer performance in the grid vicinity. The improvement of the heat transfer increases the departure from the nucleate boiling ratio, allowing higher operating power in the reactor. Due to these important thermal and fluid dynamic features, experimental and theoretical investigations have been carried out in the past years for the development of spacer grid design. More recently, the Computational Fluid Dynamics (CFD) using three dimensional Reynolds Averaged Navier Stokes (RANS) analysis has been used efficiently for this purpose. Many computational works have been performed, but the appropriate numerical procedure for the flow in rod bundle simulations is not yet a consensus. This work presents results of flow simulations performed with the commercial code CFX 11.0 in a PWR 5 × 5 rod bundle segment with a split vane spacer grid. The geometrical configuration and flow conditions used in the experimental studies performed by Karoutas et al. were assumed in the simulations. To make the simulation possible with a limited computational capacity and acceptable mesh refinement, the computational domain was divided in 7 sub-domains. The sub-domains were simulated sequentially applying the outlet results of a previous sub-domain as inlet condition for the next. In this study the k– ε turbulence model was used. The simulations were also compared with those performed by Karoutas et al. in half a subchannel and In et al. in one subchannel computational domains. Comparison between numerical and experimental results of lateral and axial velocities along of the rod bundle show good agreement for all evaluated heights downstream the spacer grid. The present numerical procedure shows better predictions than Karoutas et al. model especially further from the spacer grid where the peripheral subchannels have more influence in the average flow.