Validation of CFD analysis for rod bundle flow test with vaned spacer grids

Abstract Spacer grids with mixing vanes are generally used in fuel assemblies of Pressurized Water Reactor (PWR), which is because that mixing vanes could enhance the lateral turbulent mixing in subchannels. Thus, heat exchangements are more efficient, and the value of departure from nucleate boiling (DNB) is greatly increased. This paper presents the CFD simulation and validation of the turbulent mixing induced by spacer grid with mixing vanes in rod bundles. Experiment data used for validation came from 5 × 5 rod bundle test with Laser Doppler anemometry (LDA) technology, which is organized by Science and Technology on Reactor System Design Technology Laboratory. A 5 × 5 rod bundle with two spacer grids were used. Mean axial velocities and turbulent intensities (W rms ) were measured in the test as well as the pressure drop of spacer grids. This simulation employed the ANSYS code CFX 14.5. RANS models such as K-epsilon, RNG K-epsilon, Shear-Stress Transport (SST) K-omega and BSL Reynolds-stress turbulence model were chosen for validation. Validation results showed that RANS models were nearly adequate for prediction of mean velocities, while K-epsilon and RNG K-epsilon are more accurate under low Re condition, and Shear-Stress Transport (SST) K-omega and BSL RSM have better performance under high Re condition; as to turbulent intensities, all RANS models underestimate them; as to the pressure drop comparison results, RANS models predict well under high Re condition, especially for RNG K-epsilon, but there are large deviations under low Re condition.