A Numerical Sutdy of the Influence of the Void Drift Model on the Predictions of the ASSERT Subchannel Code

Abstract

One of the factors which is important in order to ensure the continued safe operation of nuclear reactors is the ability to accurately predict the Critical Heat Flux (CHF) throughout the rod bundles in the fuel channel. One method currently used by the Canadian nuclear industry to predict the CHF in the fuel bundles of CANDU reactors is to use the ASSERT subchannel code to predict the local thermal-hydraulic conditions prevailing at each axial location in each subchannel in conjunction with appropriate correlations or the CHF look-up table. The successful application of the above methods depends greatly on the ability of ASSERT to accurately predict the local flow conditions throughout the fuel channel. In this paper, full range qualitative verification tests, using the ASSERT subchannel code are presented which show the influence of the void drift model on the predictions of the local subchannel quality. For typical cases using a 7 rod subset of a full 37 element rod bundle taken from the ASSERT validation database, it will be shown that the void drift term can significantly influence the calculated distribution of the quality in the rod bundle. In order to isolate, as much as possible, the influence of the void drift term this first numerical study is carried out with the rod bundle oriented both vertically and horizontally. Subsequently, additional numerical experiments will be presented which show the influence that the void drift model has on the predicted CHF locations.