Science and Technology

CFD-Based Turbulence Force Evaluation for Grid-to-Rod Fretting Phenomena

Article Background: Roger Lu - #83 - L1:CASL.P5.02 - Determine extent structural analysis amplifies (or damps) differences in pressure forces between different CFD codes for analysis of GTRF phenomenon (CASL.012) based on “CASL report CFD Turbulence Force Calculations and Grid-to-Rod Fretting Simulation”

CFD-Based Turbulence Force Evaluation for Grid-to-Rod Fretting Phenomena

R. Y. Lu and Z. Karoutas         Westinghouse Electric Co., LLC
M. A. Christon, J. Bakosi and L. Pritchett-Sheats         Los Alamos National Laboratory

Fuel rod wear due to grid-to-rod fretting (GTRF) is currently one of the main causes of fuel leakage and is responsible for over 70% of the fuel leaks in pressurized water reactors (PWRs) in the U.S. The Consortium for Advanced Simulation of Light Water Reactors (CASL) has identified GTRF as one of the Challenge Problems for VERA – the Virtual Environment for Reactor Applications. The GTRF challenge problem is being used to drive the requirements, development, and application of VERA and its components for predictive simulation of light water reactors.

GTRF is a complicated phenomenon, which includes flow excitation force, non-linear mechanical vibration, tribology, changing irradiated material properties and fuel assembly geometry. In the fuel assembly, fuel rods are supported by spacer grids. As the coolant flows through the fuel assembly, fluid forces generated by the flow field induce fuel rod vibrations. The flow-induced vibration causes small relative motions between the grid supports and the fuel rod, leading to fretting wear. Predictive simulations of GTRF involve turbulence flow, structural dynamics, contacts and wear. Computing flow turbulence force, which is the excitation force for fuel rod vibration, is a challenging aspect of GTRF simulation.

The highly turbulent flow in the fuel rod assembly is computed using large-eddy simulation (LES).   Instantaneous fluid forces on the fuel rod from the LES are used as input forces for a structure code to predict the fuel rod vibration response. Hydra-TH [ref], developed at Los Alamos National Laboratory, and STAR-CCM+ [ref], are used to compute the turbulent forces. Both results are used as excitation input forces for VITRAN (VIbration TRansient Analysis – Nonlinear). The VITRAN simulations are used to produce a comparison of the rod vibration and fretting wear work-rate based on input from Hydra-TH and STAR-CCM+.

Figure 1 shows an example of the instantaneous turbulent flow field from CFD simulation. The transient forces are computed on the center rod in the CFD model. The fuel rod is divided into 12 one-inch segments downstream from the tip of the mixing vanes. Transient forces acting on the fuel rod surface are integrated at each rod segment at each time step in two lateral components, for this analysis the two lateral (X and Z) components. Standard deviations of the force time series from Hydra-TH and STAR-CCM+ calculations are shown in Figures 2 and 3 for comparison. In vibration terminology, the standard deviation of a fluctuating quantity in the time domain is the overall vibration amplitude, or RMS amplitude.

  1. Results from Hydra-TH and STAR-CCM+ are in reasonable agreement, and show that the mixing vanes are the main source of turbulence that generates the excitation forces on the fuel rod. The excitation forces decay along the span downstream of the spacer grid.
  2. The Hydra-TH results with 14 million cells are relatively close to the STAR-CCM+ results with 48 million cells.
  3. Results from both Hydra-TH and STAR-CCM+ show the force in the X-component is slightly lower than the Z-component

The Hydra-TH and STAR-CCM+ both use LES models and have produced RMS force distributions that are comparable despite differences in the solution algorithms and meshes.

Figure 1: Snapshots of the instantaneous turbulent flow field from CFD simulation

Figure 2: Turbulent force on distribution along the rod, X-Component 7MX and 14X are results from Hydra-TH and STARX from STAR-CCM+

Figure 3: Turbulent force on distribution along the rod, Z-Component 7MZ and 14Z are results from Hydra-TH and STARZ from STAR-CCM+

VITRAN [ref] is used in this feasibility study by applying flow turbulence forces obtained from the Hydra-TH and STAR-CCM+ results to a fuel rod model. The fuel rod model has six support grids (see Figure 4). For each span, the several lumped RMS force segments are applied. The lumped forces are modeled as a white noise random signal with a frequency range from 5 Hz to 100 Hz.

Figure 4: A full-length fuel rod model with six spacer grids

Figure 5 show the comparisons of the fuel rod acceleration amplitudes from the VITRAN simulations that use CFD results as inputs. Figure 6 shows the comparison of wear work-rate. The comparisons show that the results of Hydra-TH with 14 million cells and STAR-CCM are very close in vibration amplitudes and work-rate. The computed work-rate differences based on the Hydra-TH results computed with a 14 million mesh and the STAR-CCM+ data computed using a 48 million element mesh differ by less 2% .

The VITRAN results show that the CFD modeling methodologies used by Hydra-TH and STAR-CCM+, while different, produce comparable RMS forces. Both codes show very good agreement.

Figure 5: Fuel rod acceleration vibration amplitude comparisons

Figure 6: Mid-grid work-rate comparisons