Experimental measurements of turbulent flows in a rod bundle with a 3-D printed channel-type spacer grid

Pressurized water reactors (PWRs) are the most common types of electricity generating nuclear reactors. Within their cores, fuel rods generate heat by fission and the high-pressure water is used as a coolant. The heated water is then used as a heat source in a steam generator that boils water in the secondary loop. The steam is used to spin a turbine and generate electricity. Spacer grids are key components of a PWR's core. Their objectives are to maintain the fuel rods at their positions and enhance the coolant mixing and heat exchange. Over the past decades, numerous experimental and numerical studies have been performed, to characterize the flow induced by different types of spacer grids. The recent advancements of the computational description of the above-mentioned flows, led to a shift to higher resolution models for turbulence like Large Eddy Simulations. Therefore, highly spatially resolved experimental data is needed for code validation. The comparison and usage of the data about the flow in rod bundles with spacer grids obtained by different research groups is challenging. One major reason comes from the fact that the majority of the spacer grids studied are proprietary, and their geometries cannot be disclosed. The data required for some purposes like numerical code validation do not necessarily have to originate from one specific type of spacer grid. It is important to provide a detailed overview of the geometry tested and the experimental conditions. Equally crucial is to possess high spatial resolution and well-quantified and low relative uncertainty. In this study, a new, non-proprietary spacer grid is designed and 3D-printed. The flow induced by the spacer grid is characterized in a 5 x 5 rod bundle facility with Matching of Index of Refraction (MIR) at a Reynolds number Re = 27,390 using particle image velocimetry techniques. A complete uncertainty analysis is performed accounting for a variety of uncertainty sources. A low relative uncertainty is obtained. Eight middle-of-subchannel planes are selected as the domain of interest for this paper. Highly spatially resolved statistical results for the velocities, root-mean-square (RMS) fluctuating velocities, and Reynolds stresses are obtained along the span-wise direction of the flow, starting at the edge of the spacer grid up to 4.3 hydraulic diameters. The cross-flow between subchannels induced by the spacer grid is sustained for all the domain assessed. The turbulence induced by the spacer grid is assessed with the use of two-points cross-correlations and with the comparison of the RMS fluctuating velocities at different elevations. An increase of the size of the turbulent vortices along the span-wise direction is related to the decrease in the rms fluctuating velocities to evidence the turbulent dissipation. Differences between the size of the turbulent structures found with integral length scale calculations in the stream-wise and span-wise directions are attributed to the turbulence anisotropy. The full dataset and spacer grid geometry design will be available on the research group website of the authors. The data obtained can be used by other groups according to their interests.