Towards CFD Modelling of the Lower Plenum Mixing Test of the OSU High Temperature Test Facility

Abstract

Prismatic High Temperature Gas-cooled Reactors (HTGRs) are a concept approaching deployment, often as Small Modular Reactors (SMRs). Deploying these reactors requires modeling and simulation tools that have been validated for these systems but most System Thermal Hydraulic (STH) modeling and simulation tools were originally developed and validated for Light Water Reactors (LWRs). To provide a set of validation and verification problems, an OECD/NEA benchmark is underway that is based on the High Temperature Test Facility (HTTF) at Oregon State University (OSU). This is a 1:4 scaled integral effects experiment on the General Atomics’ (GA) MHTGR design.

The OECD/NEA benchmark consists of three separate problems to be analyzed. One of these is the Lower Plenum (LP) mixing exercise. This problem can be tackled in two cases: a code-to-code comparison study with fixed boundary conditions and a code-to-experiment comparison study with best estimate boundary conditions. In the framework of the European collaborative project GEMINI4.0, NRG is currently participating in the code-to-code study where the boundary conditions try to mimic the full power conditions of the experiment. The commercial code STAR-CCM+ is employed for the calculations. Preliminary tests showed that the problem cannot be solved with a steady state RANS solver. This will be briefly demonstrated in the paper. After that, the focus of the current work is with Unsteady-RANS (URANS).

The benchmark requires participants who run transient simulations to show grid independence as well as temporal convergence. The present article aims to showcase results for the coarse mesh with various time step sizes to show this temporal convergence.