Stratification in SNR-300 outlet plenum

In the inner outlet plenum of the SNR-300 under steady state conditions a large toroidal vortex is expected. The main flow passes through the gap between dipplate and shield vessel to the outer annular space. Only 3% of the flow pass the 24 emergency cooling holes, situated in the shield vessel. The sodium leaves the reactor tank through the 3 symmetrically arranged outlet nozzles. For a scram flow rates and temperatures are decreased simultaneously, so it is expected, that stratification occurs in the inner outlet plenum. A measure of stratification effects is the Archimedes Number Ar, which is the relation of buoyancy forces (negative) to kinetic energy. (The Archimedes Number is nearly identical with the Richardson Number). For values Ar>1 stratification can occur. Under the assumption of stratification the code TIRE was developed, which is only applicable for the period of time after some 50 sec after scram. This code serves for long term calculations. As the equations are very simple, it is a very fast code which gives the possibility to calculate transients for some hours real time. This code mainly has to take into account the pressure difference between inner plenum and outlet annulus caused by geodatic pressure. That force is in equilibrium with the pressure drop over the gap and holes in the shield vessel. For more detailed calculations of flow pattern and temperature distribution the code MIX and INKO 2T are applied. MIX was developed and validated at ANL, INKO 2T is a development of INTERATOM. INKO 2T is under validation. Mock up experiments were carried out with water to simulate the transient behavior of the SNR-300 outlet plenum. Calculations obtained by INKO 2T for steady state and the transient are shown for the flow pattern. Results of measurements also prove that stratification begins after about 30 sec. Measurements and detailed calculations show that it is admissible to use the code TIRE for the long term calculations. Calculations for a scram of the SNR-300 give a very similar behavior. After 50 sec stratification has already occurred. The temperature distribution looks a little bit more complicated as in the case of water models. This is caused by a non-uniform temperature profile at core outlet, but the global temperature pattern in the inner outlet plenum is not very much influenced by this more complicated boundary condition. The results show that in the case of the SNR-300 and its scram concept it is admissible to simplify the equations of thermal hydraulics for long term calculations of scrams. So very specific codes can be developed, which allow a high speed computing of transients. With such a tool it is easy to carry out parametric studies of the transient behavior of the reactor, to give a better understanding of dynamical behavior