Michael E. Conner

Development of Swirling Flow in a Rod Bundle Subchannel

Experimental measurements of the axial development of swirling flow in a rod bundle subchannel are presented. Swirling flow was introduced in the subchannel from a split vane pair located on the downstream edge of the support grid. Particle image velocimetry using an optical borescope yielded full-field lateral velocity data. Lateral flow fields and axial vorticity fields at axial locations ranging from 4.2 to 25.5 hydraulic diameters downstream of the support grid were examined for a Reynolds number of 2.8×10 4

Mapping of the Lateral Flow Field in Typical Subchannels of a Support Grid With Vanes

Lateral flow fields in four subchannels of a model rod bundle fuel assembly are experimentally measured using particle image velocimetry. Vanes (split-vane pairs) are located on the downstream edge of the support grids in the rod bundle fuel assembly and generate swirling flow. Measurements are acquired at a nominal Reynolds number of 28,000 and for seven streamwise locations ranging from 1.4 to 17.0 hydraulic diameters downstream of the grid. The streamwise development of the lateral flow field is divided into two regions based on the lateral flow structure. In Region I, multiple vortices are present in the flow field and vortex interactions occur. Either a single circular vortex or a hairpin shaped flow structure is formed in Region II. Lateral kinetic energy, maximum lateral velocity, centroid of vorticity, radial profiles of azimuthal velocity, and angular momentum are employed as measures of the streamwise development of the lateral flow field. The particle image velocimetry measurements of the present study are compared with laser Doppler velocimetry measurements taken for the identical support grids and flow condition.

The Effect of Support Grid Design on Azimuthal Variation in Heat Transfer Coefficient for Rod Bundles

Support grids are an integral part of nuclear reactor fuel bundle design. Features, such as split-vane pairs. are located on the downstream edge of support grids to enhance head transfer and delay departure from nucleate boiling in the fuel bundle. The complex flow fields created by these features cause spatially varying hert transfer conditions on the surfaces of the rods. Azimuthal variations in heat transfer for three specific support grid designs, a standard gird, split-vane pair grid, and disc grid, are measured in the present study using a heated, thin film sensor. Normalized values of the azimuthal variations in Nusselt number are presented for the support grid designs at axial locations ranging from 2.2 to 36.7 D h . Two Reynolds numbers, Re = 28,000 and Re = 42,000 are tested

Benchmarking Computational Fluid Dynamics for Application to PWR Fuel

The present study demonstrates a process used to develop confidence in Computational Fluid Dynamics (CFD) as a tool to investigate flow and temperature distributions in a PWR fuel bundle. The velocity and temperature fields produced by a mixing spacer grid of a PWR fuel assembly are quite complex. Before using CFD to evaluate these flow fields, a rigorous benchmarking effort should be performed to ensure that reasonable results are obtained. Westinghouse has developed a method to quantitatively benchmark CFD tools against data at conditions representative of the PWR. Several measurements in a 5 x 5 rod bundle were performed. Lateral flow-field testing employed visualization techniques and Particle Image Velocimetry (PIV). Heat transfer testing involved measurements of the single-phase heat transfer coefficient downstream of the spacer grid. These test results were used to compare with CFD predictions. Among the parameters optimized in the CFD models based on this comparison with data include computational mesh, turbulence model, and boundary conditions. As an outcome of this effort, a methodology was developed for CFD modeling that provides confidence in the numerical results. (authors)

Investigation of Swirling Flow in Rod Bundle Subchannels Using Computational Fluid Dynamics

The fluid dynamics for turbulent flow through rod bundles representative of those used in pressurized water reactors is examined using computational fluid dynamics (CFD). The rod bundles of the pressurized water reactor examined in this study consist of a square array of parallel rods that are held on a constant pitch by support grids spaced axially along the rod bundle. Split-vane pair support grids are often used to create swirling flow in the rod bundle in an effort to improve the heat transfer characteristics for the rod bundle during both normal operating conditions and in accident condition scenarios. Computational fluid dynamics simulations for a two subchannel portion of the rod bundle were used to model the flow downstream of a split-vane pair support grid. A high quality computational mesh was used to investigate the choice of turbulence model appropriate for the complex swirling flow in the rod bundle subchannels. Results document a central swirling flow structure in each of the subchannels downstream of the split-vane pairs. Strong lateral flows along the surface of the rods, as well as impingement regions of lateral flow on the rods are documented. In addition, regions of lateral flow separation and low axial velocity are documented next to the rods. Results of the CFD are compared to experimental particle image velocimetry (PIV) measurements documenting the lateral flow structures downstream of the split-vane pairs. Good agreement is found between the computational simulation and experimental measurements for locations close to the support grid.Copyright

Perforated Plate Pressure Losses With Improved Inlet and Outlet Flow Hole Features

This paper describes the tests and studies performed to better understand the geometric factors affecting pressure loss in a perforated plate. In this study, the impact of a specific perforated plate flow hole geometry on pressure drop was investigated. The methodology established in this paper to investigate this hole geometry can be extended to other components with orifice type perforated plates. To reduce the pressure drop of the perforated plate, various fundamental hole geometries, including edge chamfers and edge radii, were considered. Results from various edge treatments are provided in this study, including separate effects for inlet and outlet hole geometries. Specific trends, such as the effect of increasing edge geometries on the hydraulic losses, are presented. Additionally, a correlation between small-scale and full-scale pressure loss coefficients was found and is defined.Copyright

Flow Induced Vibration and Fretting Wear in PWR Fuel

Fuel rod fretting wear is due to complex combinations of factors. Excitation of the fuel rod and motion relative to its supports can be caused by coolant flow and mechanical forces. The amplitude of response for a given set of flow and mechanical forces is dependent on the fuel rod and its support system. The fuel rod mass and stiffness and location of supports along its length dictate the mode shapes and natural frequencies. The contact geometry between the support system and fuel rod is also an important factor. The two primary areas covered in this paper are 1) the types of flow conditions that may exist in a reactor core and how they would result in excitation of the fuel rod and 2) how the relative motion of the rod to its supports can result in fretting wear. Numerous testing techniques, ranging from microscopic, to single grid cell tests, to high temperature and flow full scale multiple fuel assembly tests, are discussed.Copyright

CFD Analysis of PWR Core Top Region: Top Fuel Assembly and Top Nozzle Regions

This paper aims to study the pressure distribution and flow patterns in the top fuel region of the AP1000™ reactor using CFD. This study is being performed as part of a CFD evaluation of the flow in the top fuel and upper plenum regions of a PWR reactor vessel. The flow patterns, including cross flows in the top fuel region, are inter-related with the flow distribution and pressure forces in the reactor vessel upper plenum region. Before detailed computations of the flow in the whole top fuel and upper plenum region are performed, conducting local computations for segments of the domain can provide information about physical aspects of the flow as well as mesh sensitivities. The domain of interest in this paper is the top fuel region including the upper part of the fuel assembly (top grid, fuel rods, top nozzle), upper core plate, and core component hold-down device. The commercial CFD computer code STAR-CCM+ is employed to generate the computational mesh, to solve the Reynolds-averaged Navier-Stokes equations for incompressible flow with a Realizable k-epsilon turbulence model, and to post-process the results. The complicated geometry of the top fuel region needs to be simplified so that the mesh size for the CFD model of the whole upper plenum and top fuel region does not exceed current software and hardware capabilities. In this study, several different trimmed meshes have been generated to study the effects of the geometries of the hold-down device and the lateral flows. Mesh sensitivity studies have been conducted for each individual part, i.e., the top grid, top nozzle, upper core plate, and hold-down device, in order to determine the proper geometrical simplifications. Pressure drop measurement data are compared with the predicted CFD results and act as a guideline for the mesh selection. These studies support the applicability of the geometrically simplified models and chosen mesh size for the CFD model of the full upper plenum and top fuel regions.Copyright

Two and Three-Dimensional Simulations of Enhanced Heat Transfer in Nuclear Fuel Rod Bundles

The effects of artificial roughness for the purpose of thermal performance improvement in pressurized water nuclear reactors are investigated. The artificial roughness consists of two-dimensional ribs parallel to the turbulent flow. The fuel rod bundle subchannel is preliminarily modeled as an annulus using the finite element method in ANSYS/FLOTRAN. The Navier-Stokes equations are solved from the SST (Shear Stress Transport) turbulence model for the simulated annulus thermal-flow. The analyses are performed for ribs dimensions and pitch provided by published previous work. It is found that, heat transfer and differential pressure have similar behavior with highest heat transfer occurring at the reattachment point. The finite element model describes well the characteristics of turbulent flow in smooth and rough rod when compared to previous semi-empirical models. Next paper extends the analysis by comparing numerical results with experimental test data and sensitivity analyses for different roughness configurations.Copyright

The Effect of Support Grid Design on Azimuthal Variations in Heat Transfer Coefficient for Rod Bundles

Support grids are an integral part of nuclear reactor fuel bundle design. Features, such as vane pairs, are placed on the downstream edge of support grids to enhance heat transfer and delay departure from nucleate boiling. The complex flow fields created by these features cause spatially varying heat transfer conditions on the surfaces of the rods. Azimuthal variations in heat transfer for a standard grid, split-vane pair grid, and disc grid are measured in the present study using a heated, thin film sensor. Normalized values of the variations in Nusselt number are presented for the support grid designs at axial locations ranging from 2.2 to 36.7 Dh . Two Reynolds numbers, Re = 28,000 and Re = 42,000 are tested. Results identify distinctive azimuthal variations in Nusselt number for all three of the support grid designs tested. The split-vane pair grid exhibits the largest variations in azimuthal heat transfer while the disc grid has the most uniform heat transfer.Copyright