Numerical Simulation of Coolant Mixing at the ROCOM Test Facility with CFX-5

The work was aimed at the experimental investigation and numerical simulation of coolant mixing in the downcomer and the lower plenum of pressurized water reactors (PWR). For the investigation of the relevant mixing phenomena, the Rossendorf test facility ROCOM has been designed. ROCOM is a 1:5 scaled Plexiglas model of a German PWR (Fig. 1) allowing conductivity measurements by wire mesh sensors and velocity measurements by LDA technique. Due to the fact, that the mixing is dominated by turbulent mechanisms, it is assumed that the concentration field of a tracer solution can model both boron concentration and temperature fields. The salt significantly changes the conductivity of the water what can be measured by conductance methods. In the facility, so called wire mesh sensors are applied. The measured conductivities were transferred to a mixing scalar Qx,y,z(t) representing the contribution of the coolant from the disturbed loop to the mixture at the given position x,y,z.

It is calculated from the local instantaneous conductivity sx,y,z(t) by relating it to the amplitude of the conductivity range in the inlet nozzle of the disturbed loop.

Recent experiments at ROCOM, together with data on mixing obtained at the Vattenfall test facility; the Russian VVER-1000 mock-up and measurements at the VVER-440 NPP in Paks (Hungary) are integrated into the research project FLOMIX-R within the 5th Framework Programme of EC. The objective of the project is to obtain complementary and confirmatory data on slug mixing using improved measurement techniques with enhanced resolution in space and time. The experimental data will be used to contribute to the validation of CFD codes for the analysis of turbulent mixing problems. A few benchmark problems based on selected experiments will be used justify the application of various turbulence and turbulent mixing models for various flow conditions, to suppress numerical diffusion and to decrease grid, time step and user effects in the CFD analyses.

The CFD calculations were carried out with the CFD-code CFX-5. The ERCOFTAC Best Practice Guidelines, which have been specified for nuclear reactor safety calculations within the ECORA project, have been used when making sensitivity tests for: computational mesh, numerical schemes, convergence criteria, time step, boundary positions, boundary conditions, internal geometry modelling and turbulence models.
Based on these tests the production mesh was created and the final CFD calculations were performed.

In the case of stationary mixing (4 loop operation), the maximum value of the averaged mixing scalar in the downcomer (Fig. 2) and at the core inlet was found in the sector below the inlet nozzle, where the tracer was injected. There is a good agreement between the measurement and the CFD calculation, esp. in the averaged global mixing scalar at the core inlet (Fig. 3). In the calculation the maximum mixing scalar gives the same value at the peak in the same region compared to the experiment (94%).

For turbulent flows the CFD-Code CFX-5 were validated and can be used in reactor safety analysis. Due to the good agreement between measured results and the corresponding CFD-calculations efficient modules for the coupling of thermal hydraulic computer codes with three-dimensional neutron-kinetic models using the results of this work can be developed. A better description of the mixing processes inside the RPV is the basis of a more realistic safety assessment.