Abstract :
Due to the fact that reflooding of a nuclear reactor core is a very special and complex process, in RELAP5/MOD3, a different post-CHF wall heat transfer package is used when the reflooding is activated than when it is not. During reflooding, the model implemented by the author is activated, however, before reflooding is initiated, a wall heat transfer model model which was also in old versions of RELAP5 is used. This model is physically wrong and this has been pointed out on a number of occasions. The aim of this short note is to discuss this model, propose a new approach for the post-CHF wall heat transfer and show that when the correct version of the model in the code is used, the predicted peak clad temperatures can increase by 100 K. This temperature difference can be of paramount importance for the predicted peak clad temperatures during hypothetical LB-LOCAs in a PWRs.
