Top Flooding Experiments and Modeling Estelle BrunetThibault EDF

Top Flooding Experiments and Modeling Estelle Brunet-Thibault (EDF), Serge Marguet (EDF) 1 october 25 -27, 2005 11 th International QUENCH Workshop

1 Top Flooding Phenomena 2 october 25 -27, 2005 11 th International QUENCH Workshop

1. 1 Scope of the top flooding physics in PWR • Hot leg injection • Condensation in steam generator tubes • In BWR quenching ring in upper plenum internals 3 october 25 -27, 2005 11 th International QUENCH Workshop

1. 2 Flooding Patterns Two quench fronts and two flooding patterns Co-current flow • Co-current flow pattern similar to bottom flooding Counter-current flow • Liquid single phase • Annular counter-current flow : water film and dispersed steam The location of the drying point of the water film is needed to consider the sharp change in the heat transfer between the region of dry and wetted wall and as a consequence to get the steep temperature decrease during quenching. 4 october 25 -27, 2005 11 th International QUENCH Workshop

1. 2 Counter Current Flow Limitation The phenomena Vertical : Horizontal : 5 october 25 -27, 2005 11 th International QUENCH Workshop

2 Top Flooding Experiments 6 october 25 -27, 2005 11 th International QUENCH Workshop

2. 1 ECCHO B – PERICLES (1/3) Flooding experiments at CEA Grenoble (France – ended 1991) 2 tests facilities (Nuclear technology vol. 107) • ECCHO B · 37 rods · Non heated shroud • PERICLES · 127 rods · Heated shroud Geometry RSM 1. 1 • Triangular pitch 12. 23 mm • Hydraulic diameter : 8 mm 33% reduction compared to PWR geometry 7 october 25 -27, 2005 11 th International QUENCH Workshop

2. 1 ECCHO B – PERICLES (2/3) Flooding experiments at CEA Grenoble (France – ended 1991) Test characteristics • • • Pressure : 1, 2, 2. 3 and 4 bars Initial temperature : 300 and 600°C Injected flow rates : 3. 6, 5. 4, 8. 1 g. s-1. cm-2 Stainless steel cladding Length of heated rods : 3. 6 m Measured variables Test matrix • 49 bottom flooding tests • 4 top flooding tests • 16 combined injection flooding tests 8 october 25 -27, 2005 • • • Inner cladding temperature of heated rods Fluid temperature Pressure Water injected flow rates Electrical power 11 th International QUENCH Workshop

2. 1 ECCHO B – PERICLES (3/3) Flooding experiments at CEA Grenoble (France – ended 1991) Findings • A combined top/bottom injection does not significantly improve the cooling efficiency due to high vapor velocities. • Counter-current flow rapidly limited Observations • The tests are not sufficiently prototypics to have conclusion for power plant applications · Pressure too low · The geometry is not a characteristic PWR geometry 9 october 25 -27, 2005 11 th International QUENCH Workshop

2. 2 UPTF experiments (1/3) UPTF Flooding experiments (Germany – ended 1991) Description (Nuclear Engineering and Design 133) • Full-scale (1: 1) representation of : · Upper plenum including internals · Downcomer · Four connected loops • Exact representation of : · Core barrel including core by-pass · Upper end-box and upper part of fuel element (0, 8 m) • ECC injection into : · 4 cold legs · 4 hot legs · Downcomer at two regions 10 october 25 -27, 2005 11 th International QUENCH Workshop

2. 2 UPTF experiments (2/3) UPTF Flooding experiments (Germany – ended 1991) Description • Core simulated by means of controlled steam and water injection supplied from external sources • Reactor coolant pumps and steam generators replaced by simulators • Breaks of variables sizes can be simulated in the hot and in the cold leg respectively Test characteristics • Primary system pressure : 20 bars • Primary system temperature : 485 K • ECC injection · 50 to 600 kg. s-1 for each hot leg injection port · 50 to 1100 kg. s-1 for each cold leg injection port • Onset of flooding at 10. 5 bars 11 october 25 -27, 2005 11 th International QUENCH Workshop

2. 2 UPTF experiments (3/3) UPTF Flooding experiments (Germany – ended 1991) Hot leg injection – Findings • ECC delivery to the core occurs completely without delay • Water breakthrough occurs in front of the injecting hot legs • Rate and area of water breakthrough increase with decreasing core simulator steam injection Observations • Heterogeneous distribution of steam and water = geometry dependant phenomenon • Only LOCA scenarios 12 october 25 -27, 2005 11 th International QUENCH Workshop

2. 3 PARAMETER (1/2) Flooding experiments at LUCH Institute (Russia) Sources: • Presentation on 8 th CEG-SAM meeting “Fuel assembly tests under severe accident conditions” LUCH Institute Description of the facility • 19 rods · · · Geometry VVER 18 heated rod, 1 central rod non heated Zr 1%Nb cladding UO 2 pellets Heated length : 1. 275 m • Tungsten heater elements • Hexahedron shroud • Zr. O 2 insulation 13 october 25 -27, 2005 11 th International QUENCH Workshop

2. 3 PARAMETER (2/2) Flooding experiments at LUCH Institute (Russia) Scenario • Main coolant piping break with simultaneous ECCS failure • Restoring one ECCS channel at the stage of severe accident at Tclad > 2250 K • The core water flooding from top and bottom with total flow rate of 200 kg. s-1 Advantages • Severe accident scenario · Cladding temperature · Water flow rate • Prototypics rods 14 october 25 -27, 2005 11 th International QUENCH Workshop

3 Top Flooding Modeling 15 october 25 -27, 2005 11 th International QUENCH Workshop

3. 1 Top flooding modeling in CATHARE (1/2) General description • The model assumes a wetted wall with · A descending liquid film upstream of the quench front · A steep wall temperature gradient in the quench front region · A hot dry wall downstream of the quench front Characteristics (1/2) • This model takes into account : · Nucleate boiling in the descending film upstream of the quench front · Critical heat flux at the quench front · Transition boiling immediately downstream of the quench front 16 october 25 -27, 2005 11 th International QUENCH Workshop

3. 1 Top flooding modeling in CATHARE (2/2) Characteristics (2/2) • This model takes into account also : · Heat transfer to droplets sputtered off the film in the quench front region · Dispersed flow film boiling and wall-to-vapour heat transfer further downstream of the quench front Determination of the quench front velocity • Application of a local CCFL criterion of the Wallis type at the upper quench front with the critical vapour velocity given by : L is correlated on the basis of PERICLES and REWETT II experiments 17 october 25 -27, 2005 11 th International QUENCH Workshop

3. 2 Top flooding modeling in ATHLET-CD These information are kindly transmitted by GRS (Christine Bals) ATHLET quench front model • Determination of the upper quench front velocity · Yamanouchi correlation (only valid for vertical geometries) · Calculation of the Leidenfrost temperature with Schröder-Richter approximation · Validation of this model on FLECHT, FEBA, LOFT, SCTF and CCTF experiments ATHLET drift flux model • Determination of the amount of liquid available for top quenching · Parameters of this model are influenced by UPTF data concerning CCFL 18 october 25 -27, 2005 11 th International QUENCH Workshop

4 Conclusion 19 october 25 -27, 2005 11 th International QUENCH Workshop

CONCLUSION The aim of our study • Elaboration of a severe accident top flooding model integrable in ASTEC and in MAAP 4 including: · A model to determine the amount of water available for core quenching · A model to calculate heat transfer between top down flow and upper internals · A model to calculate heat transfer in the core upstream the upper quench front • Validation of the top flooding model · PARAMETER top and bottom flooding experiment · PERICLES top and bottom flooding experiments 20 october 25 -27, 2005 11 th International QUENCH Workshop