Direct Numerical Simulations: Rod bundle

Share the news!

SESAME Newsletter #3 (March 2018) 

Access calendar of events

ALL ARTICLES

Access Newsletter here >


Direct Numerical Simulations: Rod bundle

WP2 bundleSnapshotOne of the main aims of the SESAME project is to provide reference data for flow configurations which are as relevant as possible for the applications. In this frame, the use of advanced, specifically-developed Direct Numerical Simulation (DNS) tools allows for the resolution of turbulent flows in non-trivial geometries, providing valuable information on cases for which the availability of reliable reference data is still scarce.

One of the most relevant flow configurations for Liquid Metal fast reactors is the bundle of heated rods. Such a configuration is being addressed by various SESAME partners within Work Package 2 of SESAME. Among these, the University of Modena of Reggio Emilia (UNIMORE) is performing DNSs of fully-developed turbulent LBE flow around an infinite, triangular array of heated rods. The simulations are carried out by means of an original numerical technique, developed in the frame of the FP7 THINS project, which combines the advantages of Finite-Volume Cartesian solvers, with a tailor-made treatment of arbitrarily oriented boundaries. In this way, ducts of arbitrary cross-section can be represented, without loss of accuracy, allowing for the simulation of relatively complex flows.

From a geometrical point of view, the case under investigation aims at reproducing the fully-developed flow conditions taking place in the NACIE full-scale bundle facility at ENEA, whose data will be employed for a cross-comparison with the reference DNSs. The Reynolds numbers investigated range from 10000 to 20000, approximately one order of magnitude less than the typical values of real-world applications. Such a choice is dictated by two main reasons: current computational limitations for fully resolved DNS, and the intention to explore mixed-convection regimes, i.e. regimes for which the influence of buoyancy significantly affects the flow and thermal fields. For these regimes, literature data are particularly scarce. Nevertheless, the chosen parameter values still lead to the development of a fully turbulent flow.

The results of this activity are expected to provide a deeper understanding of the turbulent structures in a bare rod bundle; a database of various turbulent properties will also be employed by SESAME partners for the validation of practical, customary turbulence models and the development of more advanced turbulent heat flux models, capable of reproducing mean flow and turbulence statistics also for mixed convection regimes.