Nuclear Engineering and Design 240 (2010) 2176–2184 Contents lists available at ScienceDirect Nuclear Engineering and Design journal homepage: www.elsevier.com/locate/nucengdes Simulation of containment jet flows including condensation Matthias Heitsch ∗ , Daniele Baraldi, Heinz Wilkening Institute for Energy, Joint Research Centre, 1755 ZG Petten, The Netherlands article info Article history: Received 20 January 2009 Received in revised form 30 June 2009 Accepted 3 July 2009 abstract The validation of a CFD code for light-water reactor containment applications requires among others the presence of steam in the different flow types like jets or buoyant plumes and leads to the need to simulate condensation phenomena. In this context the paper addresses the simulation of two “HYJET” experiments from the former Battelle Model Containment by the CFD code CFX. These experiments involve jet releases into the multi- compartment geometry of the test facility accompanied by condensation of steam at walls and in the bulk gas. In both experiments mixtures of helium and steam are injected. Helium is used to simulate hydrogen. One experiment represents a fast jet whereas in the second test a slow release of helium and steam is investigated. CFX was earlier extended by bulk and wall condensation models and is able to model all relevant phenomena observed during the experiments. The paper focuses on the simulation of the two experiments employing an identical model set-up. This provides together with other valida- tion exercises the information on how well a wider range of flowing conditions in a full containment simulation can be covered with a single set of models (e.g. turbulence and condensation model). Some aspects related to numerical and modelling uncertainties of CFD calculations are included in the paper by investigating different turbulence models together with the modelling errors of the differencing schemes applied. © 2010 Published by Elsevier B.V. 1. Introduction In the course of a severe accident the combined release of hydro- gen and steam into the containment of a nuclear power plant is likely. Whereas close to the release location a critical or sub-critical jet develops, in the far-field often nearly stagnant flow conditions can be found. The combination of these flow conditions is crucial for a realistic simulation of an accident scenario. The steam fraction of a multi-component jet is affected by the lower temperature of the surrounding atmosphere and the cold solid structures in the con- tainment. This leads to condensation in the bulk flow and on walls. These aspects were investigated in a series of experiments at the Battelle Model Containment (B-MC) called HYJET (Kanzleiter et al., 1996). Two selected tests of HYJET are simulated in this work. Ear- lier, a pure helium-in-air experiment from this family of tests was already simulated (Heitsch, 2000; Wilkening et al., 2008). One of the current experiments represents a slow vertical helium–steam jet with a nominal release speed of 4.9 m/s. The second test addresses a much higher jet speed of approximately 730 m/s. As a result of the simulation of both experiments more confidence in the selec- tion of the right turbulence model for containment mixing analyses ∗ Corresponding author. Tel.: +31 224 56 5162; fax: +31 224 56 5621. E-mail address: heitsma@gmx.de (M. Heitsch). (Heitsch et al., 2007) and additional information about the applied condensation model are expected. 1.1. Test facility The Battelle Model Containment was a complex large-scale test facility (dismantled in 1999) which was used in many test series investigating containment specific phenomena. It is built from concrete and consists of many interconnected compartments. The walls have steel liners. The total inner volume sums up to about 600 m 3 . The total height of the facility is about 9 m. The height of the individual compartments inside the B-MC is in the order of 2 m. Some of the compartments are curved others are cylindrical. A vertical cut through the facility is depicted in Fig. 1. This fig- ure shows several main compartments together with some flow connections which are relevant for the experiments under consid- eration. Another impression of the spatial shape and the curvature of many compartments can be seen in the computer model of the facility in Fig. 5. It should be mentioned that earlier experiments at the B-MC which included hydrogen deflagrations and tests with over- pressure a small leakage rate through the walls exists. But this leakage is not relevant for the experiments investigated here because these do not create high pressures and have a comparably short duration. 0029-5493/$ – see front matter © 2010 Published by Elsevier B.V. doi:10.1016/j.nucengdes.2009.11.024